Home

HPA-RPD-008: personal dosimetry of RF radiation

Contents

1. 100 1 100 I eU Volunteer 5 E Volunteer 6 EE yd Tv4as W D eS Tv4as 80 4 o A A T A All Locations 2 m 2 70 m 3 3 E 60 E 0 o g 3 50 4 3 su BSS IS ee gt g 40 8 4 a a E 30 4 A AS eS ee L L 20 4 pgs ctc a O A eee E 10 A A eS E om M 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 I 100 I Volunteer 7 Volunteer 8 EM 3 a TV485 EM o o Q0 P 9 3 TV485 O ooh s a t mla e te da e te le co lt eae E er ae a oS All Locations 7 E TE E All Locations Doc SSS SSS Ss a SS eee E Living Room ee Kitch 2 Kitchen E men E 0 Bedroom E 0 Dining Room E office Bedroom zoaco RS E eg ee el AA EP EE EE a 9e oye eyed ym os Office car Bes ZZ l e i Outdoor LO LL LLL SPSS OA E Train Outdoor 0 7 42 A B awL 2 d L o ey c EE aS SS AS A a A 10 i 10 i N 0 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T aos Eee oe NES 3 70 All Locations 5 Living Room E o MAS ESTA ITA S AS SU Kitchen g Bedroom E A RR
2. Let us know if you would like a copy of the data from your use of the PEM or to be talked through it We can also send you a copy of the final report if you wish Thank you for helping us with this project Simon Mann NRPB 17 December 2004 96
3. Volunteer Consent Form MTHR PEM Evaluation Project Having read the Guidelines for Volunteers document would like to take part in the above project being carried out by the EMF Dosimetry Group at NRPB understand that my involvement in the project will require me to e Carry a personal exposure meter PEM with me for a period of one week sometimes mounted on my body while it logs my personal exposure to RF radiation e Identify up to five locations in my home and workplace where spend most of my time and arrange access so that spot measurements of RF signal strengths can be made on the days receive and return the PEM e Complete a location occupancy diary for the week showing when have been present at the spot measurement locations and where have been at other times e Complete a questionnaire on my perceptions of the PEM and how using wearing it affected me consent that the data arising from the above activities may be used by NRPB for the scientific purposes of the current study may receive a copy of the data and any publications arising from the project if wish understand that NRPB will not publish personal information such as my name and address have seen the spot measurement equipment and will make sure that the locations where it will be used in my home are tidy and safely accessible to NRPB staff understand that NRPB staff will take reasonable care of my home and possessions when the spot measur
4. Kitchen o o Bedroom S Jl HM 0 L L E 3 50 E 50 pira Car 2 40 Wo ee eL Z 2 Outdoor E 30 4 Pepe 22 22 22 2122 2 2 0 0 0 0 a L 20 4 jou ARAS AO L ee re 10 E EE ERR A l 0 4 o H SI 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 I 100 I Volunteer 7 Volunteer 8 EM w M3 3 E e p GSMx EM o o Qu a PETIT GSMx EE ode ee n 80 4 All Locations 2 PES 2 en Living Room 2 70 Kitchen ES O SSS m Bedroom 60 2 g E RA ESE a T E Office 3 50 Car 8 8 40 4 4 Outdoor 40 E Train E 0R 7 22 E 30 4 a 0 a ye cr 20 4 i 10 10 s E gal 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T x Volunteer 9 Sce GSMrx DES e Ze v All Locations 2 EM otra Living Room cy EENMEEHEENM D D A AP Kitchen Bedroom q IR Office E Music Room PA EEE RER EP A LEIE A A 2 Ka Outdoor T 0h 17 429 L ENSE s se s uei ee eem EREI vey ey E RN LN RRNA el et es Se 0 L Pii 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 87 PERSONAL DOSIMETRY OF RF RADIATION F6 PERSONAL EXPOSURE RECORDS DC
5. When the PEM is being carried by a moving person it is likely that the effect of the movement will be to average out the polar response over time so the lack of isotropicity would not be a problem The lack of an isotropic response would also not be a problem if the PEM were placed in a fixed position in a fading multipath environment because the same averaging of its response over time will take place Lack of isotropicity is a potential problem for the stationary PEM if there is a dominant field component in a particular direction In such situations it may over read or under read by the above amounts according to how it is aligned with the field 54 CONCLUSIONS AND RECOMMENDATIONS The polar response of the PEM should be measured in an appropriate test facility and the specification updated in the light of the results Any changes to the design that can improve the polar response should be enacted 7 1 6 Sensitivity The 50 mV m detection threshold of the PEM in each of its bands reflects the sensitivity that can be expected of an instrument of its size and it would probably be difficult to lower the threshold significantly without making the PEM larger Little practical advantage would be gained by improving the sensitivity by a small amount due to the range of exposures present in the population Much of the time the recorded data from the ten volunteers who each used the PEM for a week in this project were below detection
6. 45 7 44 5 5 4 34 2 32 8 4 6 27 5 31 9 4 8 19 1 20 8 mV m in the case of the kitchen measurements on deployment Summing these signals together gives an RMS total field strength of 112 mV m whereas the maximum individual signal strength is 45 7 mV m i e 7 8 dB lower In Section 3 5 2 the PEM has been noted to under read and in such situations give a reading more equivalent to the maximum individual signal strength in the band The PEM also seems to under read in the GSMrx and DCSrx bands and probably also due to multiple signals in the bands not being accumulated correctly to give the RMS total field strength The error seems to be less than with the TV signals probably because there were fewer signals of similar strength present 4 4 Summary of spot measurements The spot measurement data indicate that the PEM bands do not capture all of the signals that contribute significantly to exposure at the spot measurement locations Omission of the DECT TETRA E GSM and WLAN bands and of the TV band above 830 MHz would be a problem for use of the dosimeter in the UK Other signals frequently detected with the narrowband equipment were those used for wide area paging around 138 and 153 MHz and other ones of unknown origin at 469 8 and 961 0 MHz Signals were occasionally measured at other frequencies such as 168 185 206 and 453 MHz The maximum field strength of any of these paging and other signals was 12 4 mV m and con
7. others on a table of similar height to the bed and others on taller furniture such as a chest of drawers 48 VOLUNTEER EXPERIENCES AND PERCEPTIONS 6 4 Design of the PEM There was a clear consensus among the volunteers that the instrument was larger than ideal and that it should be made smaller if possible The exact text of the comments on the size of the PEM were as follows e Make much smaller e Far far too big to wear Very large and noticeable when worn on body e It s too big Quite big OK for a week but would need to be smaller if worn for a longer period It would be easier if smaller Too big for constant wearing Smaller would obviously be better It would be easier if it was smaller One volunteer observed that a good target size would be similar to a large mobile phone However two volunteers pointed out that the large size of the PEM meant that they were more likely to notice it and therefore less likely to forget to pick it up Views on the weight of the instrument were split roughly equally with only two volunteers strongly expressing that the PEM was too heavy when worn on the waist Carrying the PEM in some form of shoulder bag would probably lessen concerns about its weight There were no major problems with the colour and appearance of the PEM On the whole the volunteers were happy with its shape although two observed that it could be worn less obtrusively under clothes if it were th
8. 2 8 0 4 2 lt a 3 4 4 0 Poe ee A E er ee m Exerc OO a uL 0 L 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 8 EN a 00 340 1777107 UMTStx cs PET E falla fa eM l ee Tm ae 3 70 All Locations L Living Room io RR RR A O CREE Kitchen 2 Bedroom 3 cos AA A l Office y AAA A Ei id Outdoor 06 2224222222229 a EM er do ee AI erel e 10 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 90 APPENDIX F F9 PERSONAL EXPOSURE RECORDS UMTSRX 100 T 10096 I Volunteer 1 90 DE x ee os O A E O Volunteer 2 _ UMTSrx 90 UMTSrx OOOO SO Z E hoe eS ete Se ee oe pog a EN ee Een eras j ME All Locations m EEUU Living Room CoS Se er a Se E Study 3 p io Ex ce ccc Dining Room IO EE n Kitchen E Ex Bedroom 2 Bedroom 57 A A A Office HEEL LLL usus Office m An Lan o Car eea a O a a h a a E AAA AAA 2 Outdoor 3 s Outdoor IQ E TEE RS E T 5 30 i g 3 EE eek he Gee ee eee TES S Ee o B n eed ee ge ced RA 10
9. 2 Dining room 42 39 45 104 77 144 68 59 78 3 Bedroom 31 29 34 8 115 100 132 131 97 181 4 Office Work 6 5 5 7 7 5 4 5 Car 15 13 17 9 38 33 43 25 19 35 6 Outdoor 27 24 31 9 34 26 48 6 2 All 27 23 31 5 8 1 7 1 8 3 2 2 2 2 0 2 6 3 1 Living room 2 Dining room 3 Bedroom 4 Office Work 5 Car 14 12 16 25 23 27 2 8 1 7 5 8 8 5 6 Outdoor 17 13 24 3 15 13 17 4 38 34 44 2 3 All 15 11 21 2 9 2 2 4 0 6 15 11 21 9 0 8 3 9 7 4 1 Living room 2 Kitchen 2 7 2 5 29 2 3 9 2 9 5 4 2 3 Bedroom 4 Office Work 14 13 16 2 5 Car 53 39 73 39 29 53 46 34 64 25 22 29 6 Outdoor 67 58 77 50 32 84 18 13 25 8 12 10 13 7 4 All 9 0 6 7 12 12 9 2 17 9 8 9 1 11 4 1 3 5 4 7 5 1 Living room 16 14 17 2 2 Kitchen 3 Bedroom 4 Office Work 5 Car 26 23 30 29 21 40 21 18 24 12 9 3 17 5 6 Outdoor 9 1 8 0 11 19 18 21 16 15 17 5 4 4 8 6 3 4 7 Bathroom Home 51 47 55 5 All 52 48 56 11 7 9 15 18 14 25 8 1 7 1 9 3 1 Living room 55 51 59 2 Kitchen 53 49 57 13 10 19 14 11 20 8 9 0 7 8 10 6 3 Bedroom 68 63 74 4 Office Work 1 5 1 4 1 6 2 5 Car 26 24 31 49 36 68 40 30 56 27 23 31 6 Outdoor 29 27 32 26 24 28 54 40 75 33 25 46 3 7 Train 530 330 820 63 47 89 47 36 67 40 35 45 4 8 Study Home 98 85 110 6 All 4 7 4 4 5 1 4 5 5 4 8 6 3 2 4 2 3 9 4 5 8 1 Living Room 2 Kitchen 3 Bedroom 4 Office Work 90 79 100 2 15 14 16 4 5 Car 25
10. 90 TVs 8 4 d 3 70 All Locations 5 SSA SO SS AO O Study i PEAP A AA Kitchen 2 Bedroom 3 cov HN IS Office a a a E E a a IE m Outdoor yn Z a o ts EN us 8 ENS sh gh l 10 a _ _ _ _ __ _ _ _ 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 100 T Volunteer 4 e O Ed ES 80 2 ES Samples above threshold 2 a 8 S Sga S Z S Z Z 20 4 ELE SS SS E All Locations Living Room 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 100 T Volunteer 6 scs ERA A T ge E e A A z ka r l an a eee 3 DET LALALILm E Allocations 2 Living Room B ewL 2 2 Oo r sr l ar rar d c E rr bar ar arla Dan 2 8 0 4 2 lt a 3 4 0 4 0 E EME ERA MRE AA E AAA A AE BE SSS a e E O OO O OSO paa l 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 8 Ete cl El oo PET Me Eo A Ip fa eM A E N 3 70 All Locations L Living Room io O E ELE M E N le O Kitchen 2 Bedroom 3 cos AAA A Office y AAA A E ad Outdoor S 06 2224222222229 D EM lt l er AS AA erer adle 10 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field stre
11. Frequencies in bold are mid band frequencies for each of the nine PEM bands The GTEM cell system was used according to the test methods in Section 3 1 2 A field of 2 5 V m was established in the GTEM cell using the Holaday HI 6005 probe as a reference This was then removed and the instrument was exposed to the field for 15 PERSONAL DOSIMETRY OF RF RADIATION 1 minute in each orientation The instrument was then removed from the chamber whilst the next frequency in the series was set up On completion of the series the instrument was connected to a PC and the data downloaded using the supplied software For each frequency and orientation the mean value for each individual band as shown by the analysis software was recorded Values of 0 05 V m were taken as zero field A typical set of response data for one axis of one instrument is shown in graphical form as Figure 7 and an example data set can be found in Appendix C s n 001 X Axis CW response to 2 5 V m 200 180 160 BEM mTv3 OTV485 OGSM Tx MGSM Rx EIDCS Tx BIDCS Rx EIUMTS Tx MUMTS Rx 140 120 Response 100 Exposure Frequency MHz Figure 7 X axis response of one PEM with CW signals 3 3 1 In band response summary The results of pooling the data from the in band tests of all 8 instruments are shown in Table 4 The mean recorded field is shown with the standard error n 72 and is also expressed as the
12. supervising a small child because they wanted to be able to move more freely 6 6 Use of the PEM in future studies Questions were asked about how long the volunteers would be prepared to use the PEM in a future trial with and without keeping a written diary and the results are shown in Table 19 Generally a week was seen as the maximum time while keeping a written diary with seven of the volunteers indicating this period Without keeping a written diary some of the volunteers would be prepared to wear the PEM for longer so two to four weeks would seem practical One volunteer was prepared to use the PEM for three months with or without keeping a diary if they were not expected to wear it on their body or carry it every time they went out of the room for a few minutes The volunteers were then asked to consider how long they felt it would be reasonable to ask a member of the public to use the PEM in a future study and these results are also shown in Table 19 With a diary most of the volunteers felt a week would be acceptable 50 VOLUNTEER EXPERIENCES AND PERCEPTIONS but there were a number of reservations and one volunteer felt it would not be reasonable to expect a member of the public to keep a diary for any length of time One volunteer felt that much would depend on whether the PEM could be made lighter and smaller Another suggested that there should be a review at one week after which the trial could be terminated for an
13. tests but used fields of 2 5 0 25 0 125 0 10 0 08 0 06 and 0 05 V m There were difficulties in carrying out this measurement using fields close to the instrument s lower response limit Taking account of this no significant non linearity was found 3 4 Response to modulated signals An extensive series of tests was conducted to determine the response of the instrument to modulated signals In general these were conducted with one or two instruments at field strengths of 1 0 and 2 5 V m using both modulated and unmodulated signals for each of the X Y and Z orientations Confirmation of the consistency of field between modulated and unmodulated states was obtained by the use of a thermal power meter Rohde amp Schwarz model NRS 3 4 1 FM broadcast This was conducted at a frequency of 98 MHz using a frequency modulated signal with 150 kHz deviation at a rate of 400 Hz Responses were essentially identical with or without modulation at field strengths of 1 0 and 2 5 V m 3 4 2 Narrowband FM This is meant to be representative of communications systems such as PMR mobile radio services such as taxis found within the FM and TV3 response bands of the instrument A frequency of 198 MHz was used with a single tone modulating frequency of 400 Hz at 3 kHz deviation Again responses with or without modulation were essentially identical at field strengths of 1 0 and 2 5 V m 3 4 3 Digital audio broadcasting DAB Radio In the
14. 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T e A A TRE AA SN Volunteer 4 un k a 4 4 3 w 8 m 8 HA o MEL SS E S nn a 30 O o un LL l h l mu bl SSS Se SS se E 4 ll i A 2_ ___0GTEO gt OE Q_Q___D nn 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T ERE AA A EI Volunteer 6 wy a w 3 w 8 Ml 22 7 E 0 2 g S oo Besen lee l ele E 8 4 a e A ee E U 20 l U Ee N 4 U 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 8 Y 90 FM O O EE d 2 All Locations Oo Bless 22632222222 22242225 5 Living Room ro o MR Kitchen 2 Bedroom DES o cr Mna ed noces 2 ENS AO MS E 50 Office zoo E cec eec pis a Outdoor B awL 2 d 7 s Eccc ic 10 0 0 00 0 15 0 20 0 25 Electric field strength V m 83 PERSONAL DOSIMETRY OF RF RADIATION F2 Samples above threshold Samples above threshold Samples above threshold Samples above threshold Samples above threshold 100 T Volunteer 5 peer e wl a Ee ee eae ee ee ee le EE mo eomm meme o MER UR a L ME el ot coc ELE SS es E e a E 5 1 posed O Se SS SS Se eS 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Elec
15. 1 2 Test methods The majority of the tests were carried out in a large GTEM Cell EMCO model 5311 as shown in Figure 6 Specially shaped foam blocks of a low dielectric constant material Eccostock SH 2 Emerson amp Cuming were manufactured to hold the instrument so that when it was placed in the cell in either the X Y or Z orientations the origin of the x y and Z axes was always in the same position This position was at a height of 30 cm above the chamber floor and 70 cm below the septum The GTEM exposure system allowed known fields of up to about 5 V m to be established at the instrument test position at any frequency between 30 and 4200 MHz An Agilent E4483C Vector signal generator formed the primary signal source the output of which was amplified using a Mini Circuits ZHL42 W wide band amplifier Appropriate directional couplers and a power meter or RF millivoltmeter allowed monitoring of the drive level applied to the cell An RF change over relay operated via the GTEM door interlock system effectively allowed exposures to be started or stopped by closing or opening the chamber access door For the majority of tests the instrumentation was controlled by a general purpose PC running specially written control software to allow automatic input power level and hence exposure field control Where more than one signal was required Marconi 2031 and Hewlett Packard HP4831B signal sources were also used the outputs from the sources
16. 1 60 1 20 0 95 0 21 Home Living room Home Kitchen Home Bedroom Work Office 0 23 0 72 1 13 0 37 0 78 1 39 0 38 1 42 0 92 2 42 7 67 6 37 Home Living room Home Kitchen Home Bedroom Work Office 0 05 2 02 2 95 1 14 1 68 3 61 0 79 Home Living room Home Kitchen Home Bedroom Work Office Home Living room Home Kitchen Home Bedroom Work Office 0 63 6 57 0 32 2 85 0 73 7 24 3 37 OOOO OOo 0 Oo 4 Y Y 0 0 O 0 01 O1 01 Qn A E E EG 00 G GKM ION MIA Home Living room Home Kitchen Home Bedroom Work Office 1 70 4 74 9 36 1 96 2 02 Home Living room Home Kitchen Home Bedroom Work Office 2 12 0 73 1 17 2 67 0 63 4 81 2 68 0 57 4 03 Band averages 0 69 0 37 0 32 1 22 0 12 1 15 1 12 2 31 0 87 0 52 35 PERSONAL DOSIMETRY OF RF RADIATION 4 2 2 Results Table 12 indicates that the RMS field strengths summed over the PEM bands generally have a remarkable degree of consistency between the measurements made at the beginning and end of the week long trials Of the 51 points where comparison could be made 43 are within 3 dB and 20 are within 1 dB It is likely that the few extreme outliers in the distribution e g 9 36 6 57 4 74 dB arose due to changes in the RF spectrum over the week and so they may not have implications for the repeatability of the narrow
17. 5 4 5 4 5 2 5 5 Reclining 4 5 5 5 4 4 5 5 5 Scoring system 1 No problem 2 Minor problems 3 Somewhat problematic but still practical on balance 4 Fairly impractical 5 Totally impractical The table shows a wide variation in the reported practicality of wearing the instrument from volunteer to volunteer even within the same situation The volunteers generally found it practical to wear the PEM from the waist when standing or walking but not when sitting or reclining In these situations where the volunteer is not moving around it would be usual to take the PEM off and place it nearby 6 3 Placing the PEM near the body When the volunteers were not moving around they generally removed the PEM and placed it nearby When in their lounges volunteers were usually sat on a sofa or chair and placed the PEM on the arm of the chair or on a table next to the chair The volunteers indicated that it was easy to forget to pick up the PEM when they went to the kitchen or bathroom and that sometimes when they knew they would only be away for a few minutes they chose to leave the PEM behind When in the kitchen volunteers except for numbers 2 4 and 8 who wore the PEM tended to put the PEM on a work surface at a variable distance generally up to 1 2 m away from them When in the bedroom the volunteers generally placed the PEM near to the bed but at a variable height due to differing bedroom furniture Some put the PEM on the floor
18. T Volunteer 5 ee E ar gt a a ceca All Locations 8 7 4 4 B 0 2 o ca E s S 8 0 a a B e e a o 20 i co s s r za ese bese eee sds s 0 L 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 7 90 FM ooo EIE E CD TI A o I E qr 2 PS a G bow T HEN voy ee ee ae eee s 8 4 4 4 a HEUS BEA o A A tee 10 i H el 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T 3 Volunteer 9 90 FM 80 4 A 2 All Locations von Z r SS H G Se Se ee E ee 5 Living Room E a5 CA C I T RE Kitchen g Bedroom O Si Office j BE pese ee eee l Music Room ze Outdoor X L 4 4 d o ao a Z a ee eee eee eee eee ee ee N ee P e e SS 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 2 90 FM mm gt gt gt _ SSS 3 70 All Locations Tj Study zu RARA Sonn Sc one Kitchen 2 Bedroom DE Lube ntn EAS Er 50 Office o s Idi a Outdoor B 3wL 2 d o oy RAD i 1 BTE EEE II ESSEN I I EN R e 0 I nn
19. T E Office j os 2 40 Music Room E 30 o 20 10 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 85 PERSONAL DOSIMETRY OF RF RADIATION F4 PERSONAL EXPOSURE RECORDS GSMTX 100 I Volunteer 1 pos GSMtx Bere RU E a c ci 2 All Locations C EDD pee ee rl ee K Living Room E oo PET erp EE Dining Room o EMEN 1 12 2 0 2 2 Jone ele E 8 0 4 a E a o cay n al n emer L ee ee eg eer ere i 10 1 a AAA 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 I Volunteer 3 A EE E En 80 N j era ar rcr to aal E ee Samples above threshold Boo 8 8 8 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 j Volunteer 5 ce DSSS SS SSS Se SS co 80 4 T x All Locations 2 70 5 60 4 o 50 4 E B 405 a E 30 o 20 4 10 0 4 0 00 100 ji Volunteer 7 oo a ea aa eaeaaoaad ze ELE e n All Locations 2 B HEEL SS u Living Room 8 Kitchen EE ee E Bedroom H a A CHER CANTE hace AE Office 8 Car 8 oo eg ue Outdoor e Train too S y SS j EE onem i 10 l oy o 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 100 Electric field strength V m 90 x g O
20. Uncensor 4 program as available on the Internet from the Virginia Institute of Marine Science was used to implement Helsel s Robust Method This program fits a lognormal distribution to those data values that are above a detection threshold and then generates data values at random according to the fitted distribution to replace the censored values below the detection threshold Once all of the censored values have been replaced the arithmetic mean of all the data is evaluated Uncensor 4 can process a maximum of 1000 data values Consequently data sets from the PEM of greater than this size were sampled randomly to produce sets with exactly 1000 entries Also the floating point format of Uncensor 4 was fixed in the program and so the field values in V m were converted to mV m in order to improve the precision of the program output 2 4 Narrowband spot measurements The equipment used to make precision spot measurements comprised an ARCS Siebersdorf Miniature Biconical dipole antenna mounted on a wooden tripod and connected to an Agilent E4407 spectrum analyser The protocol under which this equipment was used is in Appendix D and the results are in Section 4 The antenna was inclined at an angle of 35 3 to the horizontal plane and its base plate allowed it to be rotated in steps of 120 Thus three orthogonal polarisations could be measured in turn by rotating the antenna These measurements were summed to obtain the resultant fiel
21. a small number of isolated events 3 8 4 Charging reliability Following some problems during the field trials all eight instruments were again set up in the laboratory for a run of 7 days and with a 120 s recording interval Each instrument was charged immediately before the run with the supplied charger until the charger showed that the fast charging was complete All were periodically inspected during the working day to confirm they were still operating as shown by the flash of the logging LED every 2 minutes Any instrument that stopped logging was immediately downloaded recharged and returned to the trial Six instruments ran satisfactorily throughout the trial One number 4 stopped after about 6 h and was recharged and restarted It then ran for 2 9 days before again stopping The second number 7 ran for 4 8 days before stopping In all cases logging was stopped because of low battery voltage In the light of the above there would therefore appear to be some question about reliability of the battery and charging arrangements It was noted that although the instruments use Nickel Metal Hydride NiMH batteries the supplied chargers are labelled as suitable only for Nickel Cadmium types 25 PERSONAL DOSIMETRY OF RF RADIATION 3 8 5 Temperature recording During these investigations it was noted that although the instruments were operated in a temperature controlled environment 21 1 C the maximum and minimum temperat
22. and with each of the 495 levels from 0 05 to 5 00 V m defined as a separate bin The data were then further processed to form curves similar to that in Figure 10 in which the x axis was a field strength threshold and the y axis was the percentage of the appropriately tagged values above that threshold 100 T T 90 80 All Locations Living Room 60 Dining Room 70 5096 40 4 3096 Samples above threshold 20 10 4 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Figure 10 Example of processed personal exposure record results for the exposure of one volunteer in the GSMrx band Figure 10 shows that the highest GSMrx exposures for Volunteer 1 occurred in the dining room where the field strength of 96 of the values i e 96 of the time was above 50 mV m Very little of the time was the field strength for any of the locations above 250 mV m A set of graphs similar to Figure 10 but covering all volunteer band tag combinations is shown in Appendix F 5 1 3 Sensitivity aspects The graphs in Appendix F show that the recorded field strength was below the 50 mV m PEM detection threshold for much of the time and this was analysed in more detail Among the 43 934 records taken across all the volunteers only 8 values were above the detection threshold in each of the FM TV3 and UMTStx bands Table 15 shows the number of values above
23. difference from the actual field value in dB The uncertainty in the table is the sum of both the uncertainty in the exposure field and the manufacturer s figure for isotropicity All of the responses except for that with the FM band are within the expected uncertainty These results are for continuous unmodulated signals and the responses with modulated signals are discussed in Section 3 4 16 LABORATORY TESTING Table 4 In band responses Band Mean recorded field V m dB rel 2 5 V m Uncertainty dB FM 1 94 0 05 2 20 1 50 TV3 2 43 0 08 0 24 2 00 TV485 2 56 0 09 0 19 2 60 GSMtx 2 90 0 12 1 29 3 10 GSMrx 3 33 0 07 2 49 3 10 DCStx 2 66 0 06 0 53 3 70 DCSrx 2 55 0 03 0 17 3 70 UMTStx 2 55 0 08 0 18 3 70 UMTSrx 2 10 0 04 1 51 3 70 3 3 2 Out of band responses A number of out of band responses were identified as a result of the CW testing An arbitrary response level of 10 or more of the exposing field in other words 0 25 V m or more was set as a significant out of band OOB response level There were several sets of conditions that generated OOB responses These were a FM band responses to frequencies below 49 75 MHz and above 125 145 MHz the FM band A TV4 amp 5 response to 430 MHz a frequency below the band edge TV48 amp 5 responses to GSMtx frequencies 880 895 5 and 912 MHz A DCStx response to a frequency in the DCSrx band 1780 MHz A DCSrx re
24. factor Spectrum graphs and pie charts can then be produced by the user for incorporation into reports Uncertainty Budget The spectrum analyser is calibrated through substitution for a power meter when connected to a source producing a particular modulated signal of constant power Uncertainties arising in measurements when using this calibration are associated with linearity and drift of the spectrum analyser the correction factor itself and the calibration of the power meter Since the antenna factor is derived including the cable its uncertainties include those in the cable 77 PERSONAL DOSIMETRY OF RE RADIATION loss A small amount of extra uncertainty is added to account for any coupling between the antenna and its surroundings Quantity Linearity Value dB Probability distribution Divisor Sensitivity coefficient Standard uncertainty dB Drift Correction factor Power meter Antenna factor 1 0 Normal Coupling 1 0 Combined standard uncertaint Expanded uncertainty k 2 Normal 78 APPENDIX E APPENDIX E Volunteer Location Diary This appendix contains a template diary that was used by the volunteers to indicate their whereabouts at a given time so the measurements with their PEM at that time could be interpreted The diary was supplied to the volunteers with their name and spot measurement locations already inserted lt was in the
25. field strength V m 100 T x Volunteer 9 ad DCSrx we NES 3 70 All Locations 5 Living Room E o ARPA NEU REEL T ERAS E S Kitchen g Bedroom 3 DLE i eee cele el a E e E A Ea Office S os EE n dL Music Room is Outdoor 0h 1 22 9 o 20 4 sete a m la m e l e tee see eet E CT ee ee S AE I o eee 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 89 PERSONAL DOSIMETRY OF RF RADIATION F8 Samples above threshold Samples above threshold Samples above threshold Samples above threshold PERSONAL EXPOSURE RECORDS UMTSTX 100 1 90 80 70 4 60 4 50 4 40 4 30 20 10 100 100 T Volunteer 1 2 UMTStx a A ee ap 3 70 All Locations 5 MEE AGA AAA E a Living Room roo Eee OR ES Dining Room 2 Bedroom 3 A O Office goo EE E EET A eens nm ioe Outdoor a 0 cary A l ceste Eee 10 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m ob eS ee eee eee x g ONES S sS N Z EN ccc ccc 2 __ _ __ ___ ____ ___ z_ _ 100 Volunteer 3 UMTStx 0 4 0 00 Volunteer 5 UMTStx sc a e Sl e e e A A O Volunteer 7 UMTStx AII Locations EEE RSS eR Li
26. how they might wear it on their person and where they might place it if they feel the need to remove it It will be ensured that that they have a suitable belt or other means of attachment The volunteers will be given the Consent Form and encouraged to ask any remaining questions Once they have signed the form they will be considered as included in the study Confirmation of arrangements A seven day interval during which the volunteer is able to use the PEM and during which their activities are expected to be fairly typical will be agreed The order in which the spot measurement locations are to be visited will be confirmed ideally starting with those in the workplace and then continuing with those in the home The precise dates times and meeting points for distribution of the PEM and the initial set of spot measurements will be double confirmed Written confirmation of the arrangements will be sent to the volunteer 63 PERSONAL DOSIMETRY OF RF RADIATION Distribution of PEMs The volunteer will be met at the agreed location time and the spot measurements will be made in the agreed sequence The location occupancy diary will be given to the volunteer and discussed so they are sure how when to complete it The feedback form will be given to the volunteer and discussed so they are clear what information is being sought The arrangements for collection of the PEM and repeat of the spot measurements a week later will be
27. in a village far from any masts The narrowband equipment was only able to measure signals in the second floor bedroom of this house which was above the stone walls and inside the tiled roof space 4 1 5 Field strengths outside the PEM bands Table 10 shows the strengths of TETRA WLAN DECT and other signals measured with the narrowband system These signals will contribute to exposure but fall outside the PEM s specified bands and are not accounted for TETRA signals make a measurable contribution in all but 10 of the 40 locations and they make a dominant contribution for the three locations in the house of Volunteer 1 DECT signals were measured in the houses of five of the volunteers and gave significant contributions to the total field Two of the volunteers 1 amp 6 had home computers with 802 11b wireless local area network capability in their houses The 2 4 2 5 GHz emitted signals were detected when the narrowband measurements were made in the same room as the computers but not when the measurements were made in other rooms Detailed examination of the signals grouped under the other category showed that truncation of several of the bands caused signals that should rightly be included in the PEM bands to be neglected TV signals were missed between 830 854 MHz and E GSM base station signals were missed in the range 925 935 MHz Paging signals in the 153 155 MHz range gave significant contributions with some of the voluntee
28. is designed to be carried by people sometimes mounted on their body and to record their exposures to radiofrequency signals over time The project comprises laboratory investigations and a volunteer trial to assess a the technical performance of the instrument and b the practical aspects of its use in studies PERSONAL DOSIMETRY OF RF RADIATION The project is supported by the UK s Mobile Telecommunications and Health Research Programme MTHR 2005 and its results should assist the appropriate use of the PEM by researchers and the future development of the PEM by its manufacturer 1 3 Report structure Section 2 of this report describes the study protocol the instrumentation and the analysis techniques that were used in this project Detailed results from the laboratory investigations of the PEM performance are described in Section 3 Section 4 describes spot measurements made with a precision measurement system at locations where the volunteers spent a significant proportion of their time with the PEM The logged data obtained from the volunteer trial are analysed in Section 5 and feedback from the volunteers on their experiences and perceptions of the PEM are summarised in Section 6 The overall results from the project are discussed and conclusions are drawn in Section 7 MATERIALS AND METHODS MATERIALS AND METHODS This section first describes the protocols used to gather and process data in this study The projec
29. made on deployment were compared with the PEM personal exposure readings when the diary showed the volunteer to be at the location in question This showed how well the spot measurements could be used to approximate personal exposures at the same location 2 1 2 5 Feedback questionnaire The volunteers completed a questionnaire reporting on their experiences and perceptions of the PEM at the end of their week long trial The questionnaire was divided into four sections the first of which considered how much of the time the PEM had been worn and where it had been placed when it was not worn The second section asked for views on the design aspects of the instrument and the practicality of wearing it in various situations The volunteers were asked to suggest any improvements for the design of the PEM and whether they felt using it had modified their behaviour in any way The third section asked about experiences such as whether the volunteers had felt self conscious with the instrument or whether it had attracted any attentions comment from others Finally the volunteers were asked how long they would be prepared to use the instrument if asked to do so again both with and without keeping a diary Similarly they were asked how long they felt it would be reasonable to ask a member of the public to use the PEM The questionnaire is in Appendix G and the information gathered using itis summarised in Section 5 MATERIALS AND METHODS 2 2 Persona
30. measured by the PEM and those in black are outside its specified frequency bands Spot measurement Electric field strength RMS mV m data source PEM frequency bands Other bands Total Volunteer Location FM TV3 TV485 GSMtx GSMrx DCStx DCSrx UMTStx UMTSrx TETRA WLAN DECT Other 1 Home Living room 86 7 39 113 36 170 1 Home Dining room 104 119 56 116 72 11 217 1 Home Bedroom 170 113 55 104 37 238 1 Work Office 9 18 21 2 Home Study room 4 5 6 2 Home Kitchen 4 8 3 9 2 Home Bedroom 11 11 16 6 23 2 Work Office 14 18 7 1 26 35 3 Home Living room 18 9 3 14 5 25 3 Home Kitchen 16 3 1 7 17 3 Home Bedroom 5 16 12 3 60 6 64 3 Work Office 4 7 1 8 4 Home Living room 8 3 23 1 11 28 4 Home Kitchen 18 2 11 6 22 4 Home Bedroom 7 9 11 33 2 14 39 4 Work Office 18 37 1 16 44 5 Home Living room 18 69 3 22 8 75 5 Home Kitchen 16 15 82 2 36 12 93 5 Home Bedroom 39 13 130 17 6 5 15 26 141 5 Work Office 7 15 17 6 Home Living room 11 3 7 13 6 Home Kitchen S 2 102 10 4 103 6 Home Bedroom 7 37 22 7 34 56 6 Work Office 20 42 1 43 63 7 Home Living room 3 3 7 Home Kitchen 3 3 7 Home Bedroom 5 6 4 11 14 7 Work Office gt 7 1 10 13 8 Home Living room 11 8 12 1 14 12 26 8 Home Kitchen 8 4 23 1 8 11 28 8 Home Bedroom 30 13 43 22 58 8 Work Office 23 6 14 27 9 Home Living room 11 28 16 13 11 38 9 Home Kitchen 4 2 45 30 12 17 66 9 Home Bedroom 6 13 25 28 24 9 48 9 Work Office 4 1 4 10 Home Livin
31. minimum displayed field is 0 05 rather than 0 00 V m and was in this test only recorded as such Finally a response test was made to confirm the operation of each measurement axis An 895 5 MHz the centre of the GSMtx band CW field of 2 5 0 2 V m was established in the cell at the reference position and the instrument placed on the foam block support for a 1 minute period in the X Y and Z orientations The instrument data was then downloaded and the mean recorded field for each band was noted With the exception of one unreadable set up file on one instrument s CD all instruments passed their acceptance tests However it should be noted that the battery chargers supplied have continental style mains connectors and an appropriate continental to UK 13A adapter will have to be provided by the user The certificates of calibration supplied with each instrument appear identical to each other with the exception of the serial number of the PEM In particular the performance data seem identical for each one being merely described as typical characteristics The equipment used for the calibration although mentioned is not given any calibration accreditation nor is there any mention of uncertainty 3 3 Response to unmodulated signals The instrument has nine response bands as detailed in Table 1 In addition to the in band performance it was necessary to evaluate the out of band performance Consequently a large number of test frequenc
32. the volunteers Much of the time the recorded data were below the 50 mV m detection threshold of the PEM and this lack of sensitivity seems likely to limit its ability to construct an exposure gradient within a study Nevertheless the PEM does seem to work well in discriminating the relatively high exposures of people who live near to mobile phone base station and television broadcast transmitters from those of people living elsewhere The data also generally showed higher exposures in all the bands when the volunteers were outdoors or travelling in cars and trains than when they were indoors The diaries showed when the volunteers were at particular locations and so the personal exposure records could by analysed to develop mean exposure estimates for locations even where much of the time the exposure level was below the detection threshold In a small number of cases where there were at least ten personal exposure values above the detection threshold and spot measurements were also available the two sets of data agreed well This suggests that spot measurements might be a reasonable surrogate for personal exposures occurring at indoor locations 45 VOLUNTEER EXPERIENCES AND PERCEPTIONS VOLUNTEER EXPERIENCES AND PERCEPTIONS This section summarises the results from the questionnaires completed by the volunteers following the their week long trials of the personal exposure meter PEM A template questionnaire as supplied to
33. 0 8 M4 E 04 B 0 o 3 50 4 oo Besen allen E E E g 40 4 EAM E II a a 5 30 S 30 l L o ips BS a ee eee eee eer en Se Pt A 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T 100 I NEUE oo DRXLESXEMNSEM LN YT Volunteer5 _ n 7 0x TNA Toms m m VE EE I Volunteer6 _ ind DCS n DCS o Bee ee ee ee eS 80 All Locations 2 S A mL 2 70 3 B 0 2 E 60 o g T 3 50 gt 4 4a g 4054 a a SA Id E 30 4 o L 20 20 10 10 0 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 Electric field strength V m 100 I 100 I Volunteer 7 Volunteer 8 pu DCSrx DCSrx ur ee pee co lt T Ee A E IEEE eie All Locations 7 E E ae All Locations cocu nn p El Living Room NINO SS I eg x 3 Living Room E Kitchen D x Kitchen EE O a p Bedroom DERE a a ia El o offi o Bedroom Zoaco RR A E ts el ee O ice MEE Office car as Ze EA OEO a e E Outdoor A E a a Outdoor E Train E T a R sme 7 L o E A A AAA 10 10 0 E A ee o l 0 4 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric
34. 0 1 5 11 5 45 18 74 15 13 115 11 6 61 2 8 61 191 21 38 0 302 7 16 4 UMTSrx All 2 1 19 14 46 4 25 29 83 223 1 0 0 0 0 0 0 0 0 0 2 0 1 0 0 1 0 0 0 0 3 0 0 0 0 1 0 0 0 0 4 0 0 0 0 0 0 0 0 0 B 1 0 12 5 36 1 9 29 0 6 1 0 7 9 4 3 13 0 83 7 4 2 Total All 5019 5040 5040 4797 5040 5022 5040 4524 4412 43934 number of 1 2070 1130 1143 698 634 737 943 793 775 records 2 458 808 462 165 421 424 796 321 107 3 1021 1736 1902 1810 1510 1891 1742 1718 1095 4 1025 874 910 651 720 946 744 754 515 5 194 343 236 269 473 216 263 734 47 6 158 149 384 1087 1075 808 353 152 1873 7 84 80 Considering the data in Table 15 and the graphs in Appendix F a range of exposures was evident in the TV485 bands across the volunteers and locations although 94 of values taken across all volunteers were below the detection threshold Of the values above the threshold 6996 were acquired from Volunteer 5 who as noted in analysing the spot measurement data see Section 4 1 4 lived in Southeast London and nearer to VHF UHF broadcast radio masts than the other volunteers 41 PERSONAL DOSIMETRY OF RF RADIATION Unlike the other bands the GSMtx DCStx and UMTStx bands do not contain signals that are continually present they contain the generally short duration transmissions from individual mobile phones to base stations while calls are taking place The phones may be used by the volunteers or by other nearby people Simple calculations suggest a mobile phone would
35. 24 27 10 16 12 22 7 21 19 22 6 Outdoor 19 16 22 7 6 5 6 10 10 7 6 14 7 All 4 3 3 8 5 0 6 3 4 7 8 8 2 6 2 4 2 8 5 15 13 17 8 1 Living Room 2 9 2 7 3 2 4 0 4 0 4 0 5 2 2 Kitchen 3 Bedroom 4 Office Work 5 Car 39 34 45 19 16 21 19 17 22 26 24 28 9 6 Outdoor 15 14 17 30 22 41 27 23 31 25 22 29 7 Train 60 45 85 3 14 13 15 4 70 65 76 2 8 All 7 1 5 2 9 8 8 8 7 7 10 8 1 6 0 11 13 12 14 3 1 Living Room 2 Kitchen 3 Bedroom 4 Office 5 0 4 4 5 8 4 5 Car 30 22 41 36 26 49 36 27 50 19 17 22 6 Outdoor 40 38 44 7 Dining Room 16 14 18 9 All 13 9 3 17 13 9 9 18 17 13 24 9 8 7 3 14 1 Living Room 15 14 16 2 Kitchen 31 23 44 3 3 Bedroom 60 56 65 3 4 Office 5 Music Room 100 95 110 56 51 60 6 Outdoor 24 18 34 36 23 60 46 29 77 Note estimates shown in red were derived on the basis of fewer than ten measured data points above the detection threshold The exact number of such data points is shown to the right of the confidence interval 43 PERSONAL DOSIMETRY OF RF RADIATION 5 2 2 Resulting mean field values Where small numbers of values above the detection threshold were used in the analysis the algorithm seems to sometimes become unreliable For example with Location 7 train for Volunteer 7 there were 2 values out of 80 above the detection threshold in the UMTSrx band and they were both at a level of 70 mV m In these circumstances it would not
36. 3 4 11 UMTS 3G handset The WCDMA 3GPP uplink modulation option of the Agilent 4438C signal source was used for this test with an RF frequency of 1950 MHz Exposing fields of 2 5 and 1 0 V m resulted in recorded mean fields of 2 51 and 0 95 V m respectively 3 4 12 UMTS 3G base station The WCDMA 3GPP downlink modulation option of the Agilent 4438C signal source was used for this test using a frequency of 2140 MHz Exposing fields of 2 5 and 1 0 V m resulted in recorded mean fields of 1 26 and 0 63 V m implying the recorded field was in error by about 50 or 6 dB 3 4 13 Radar Some Air Traffic Control ATC radar frequencies fall within the possible response range of the instrument The response to simulated radar signals at a frequency of 1300 MHz using several pulse widths and repetition rates 4 us and 660 Hz 34 us and 825 Hz 66 us and 430 Hz was investigated using a peak field strength of 2 5 V m No responses were seen 3 4 14 Summary of modulated signal responses The results for the signals in the bands to which the PEM is designed to respond are shown below in Table 5 expressed as a relative response the mean of the X Y and Z orientation responses at all field strengths tested The Analogue TV response was significantly higher than expected and both the DCSrx and UMTSrx responses were significantly lower Table 5 Summary of modulated signal performance Band Response dB SE
37. 33 7 100 7 3 0 1840 1875 T 4 0 4 7 8 3 6 9 2 6 93 9 2 0 93 2 19 2 1900 49 0 73 0 1925 2 6 3 3 9 6 155 7 1950 94 2 1975 2 1 80 1 2115 69 7 2140 71 0 2165 2450 66 3 71 APPENDIX D APPENDIX D Narrowband Spot Measurement Protocol This appendix contains the protocol that was used to carry out narrowband spot measurements with a small biconical dipole antenna connected to a spectrum analyser The procedure involves the use of carefully chosen settings for the spectrum analyser to reduce the effects of temporal fading and averaging of measurements over three heights in order to reduce the effects of spatial fading 73 APPENDIX D Spot Measurement Protocol for Environmental RF Exposures 80 MHz 2 5 GHz Equipment List 1 2 3 4 5 Spectrum Analyser Agilent model E4407B equipped with RMS averaging detector option Calibration by EMF laboratory is required to derive correction factors for specific modulated signal classes with respect to a power meter Receiving Antenna and Connecting Cable ARCS miniature biconical antenna calibrated with its cable to give antenna factors Supporting base plate allows rotation of the antenna through increments of 120 in order to cover three orthogonal polarisations Antenna Tripod To be made of wood plastic and with index marks to allow the antenna to be mounted quickly and r
38. 4 1 16 8 5 3 4 23 5 5 1 55 69 2 0 2 53 13 14 82 3 8 3 68 130 5 9 5 6 4 15 7 1 15 13 6 4 90 20 42 7 1 29 0 4 8 4 5 0 23 6 1 19 9 1 15 11 28 16 2 9 2 31 31 45 30 0 0 3 60 13 25 28 14 Note values shown in red were derived on the basis of fewer than ten measured data points in the personal exposure records above the PEM detection threshold The values in red were derived from small numbers of values above the detection threshold and as mentioned above were regarded as potentially unreliable 44 There were few situations where good data were available from both spot measurements and personal exposure means These were confined to the TV4 amp 5 band in the house of Volunteers 5 and 9 and the GSMrx and DCSrx bands in the house of Volunteer 1 In this limited set of cases the data are within 6 dB of each other suggesting that the spot measurements were representative of personal exposures at the locations 5 3 Summary of personal measurements On the whole the PEMs acquired data in accordance with their specifications although there were two functional problems encountered A software problem which has since been corrected by the manufacturer caused the corruption of all records where the instrument temperature was below 10 C Also one of the PEMs seems to not have been charged fully at the start of the run despite being charged in accordance with the instructions and this resulted in total loss of the data from one of
39. 40 Electric field strength V m 100 T Volunteer 9 aes DCStx cos S ee e ro ee ree ey ee 3 70 All Locations B CL IDO bei 5 Living Room zo oo A T Kitchen 2 Bedroom Ss gen eese eee Office G L P m Music Room Outdoor a NS E i ee een ee ee A E EERS f LSS un nn 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 2 gt DCStx 8 d d 3 70 All Locations 5 Boc az coca O Study ENDE cem oet Kitchen 9 Bedroom 8 cov NI Office sf RI Sa BE E a EE E Outdoor 3 2224222222229 o ENS sehe l 10 l 0 L riii 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 4 Eg oc E Des 80 2 Al 3 a Be a ma ma a En a Beer ers r l a h a r h r ape 0 Samples above threshold a SISSI RR S sS Z All Locations Living Room 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 100 T Volunteer 6 LA A SEM zo 80 70 60 4 50 4 40 4 30 Samples above threshold 20 4 10 0 4 0 00 All Locations Living Room 100 T Volunteer 8 cuu Eee l rr r b r rr l r e r h mes co A a I MEI 3 70 All Locations L Living Room io EEE RA A NEUERER CRISI Kitche
40. A c E x 9 tenen 3 Living Room Eg 0 4 E Bedroom E 00 Kitchen o offi o Bedroom Loo MERI UIS T AS E ua o es sy Be y El ofe car as ZZ ee l Outdoor LO PDD LLL LL SAS OH a ran Outdoor a 7 4 7 S s gt gt 5 B 3 gt 4 a o a cn o SS ee eS ee A AA N 10 10 i 0 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T Volunteer 9 oo UMTSrx oo a T E H T Hes ar 3 70 All Locations 5 Living Room E y ARPA ADLER SS Kitchen g Bedroom 3 A A I Office S BE oe A one eee eee Music Room ok Outdoor 0h 17 7429 o e s s s r ae ee ee la eee ee aes f 2o EMENESER Eta Mei cde LP m STE T o A 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 91 PERSONAL DOSIMETRY OF RE RADIATION APPENDIX G Volunteer Questionnaire This appendix contains a copy of the questionnaire that was e mailed to each volunteer after completing their week long trial of the PEM in order to obtain feedback 92 APPENDIX G Personal Dosimetry of RF Radiation Volunteer Feedback Form You have used the Personal Exposure Meter PEM for one week and we would now like you to answer some questions on your perceptions of the i
41. APPENDIX F F5 PERSONAL EXPOSURE RECORDS GSMRX 100 T 100 T Volunteer 1 T Volunteer 2 Box GSMrx etd GSMrx 80 4 A EE 2 70 3 70 oo B ew HM 2 2 amp 50 4 SO to Il t i Ei E Y 4054 4W4 a a E 30 4 o SE SE ee eee eret a o 20 4 Qj ee Oe a RES eae eee ee 10 10 0 4 un u __ _ _ 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T 100 T Volunteer 3 Volunteer 4 Nl rrr reb e nre reete ie GsMx Nox en GSMrx 80 i 80 a Be E 70 iB 70 E Gu E 60 4 g 50 50 E E g 40 8 40 a a 5 30 i 5 30 L 0 en c A o nen 20 4 10 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 i 100 I p Volunteer 5 E Volunteer 6 EN a a 1 O A GSM 1 A r GSMx 80 4 ELEC A A A A O E All Locations 9 20 al D 70 All Locations amp 1 E B ooo HAUS E EE eue escape get n Living Room E 60 E 0
42. Acceptance tests Response to unmodulated signals 3 3 1 In band response summary 3 3 2 Out of band responses 3 3 3 Linearity Response to modulated signals 3 4 1 FM broadcast 3 4 Narrowband FM 3 4 3 Digital audio broadcasting DAB Radio 3 4 4 Analogue TV broadcast 3 4 5 Digital TV broadcast 3 4 6 GSM 900 handset 3 4 7 GSM 900 base station 3 4 8 GSM 1800 handset 3 4 9 GSM 1800 base station 3 4 10 DECT 1800 3 4 11 UMTS 3G handset 3 4 12 UMTS 3G base station 3 4 13 Radar 3 4 14 Summary of modulated signal responses Response to multiple signals 3 5 1 Signals in different bands 3 5 2 Signals in the same band Electromagnetic immunity 3 6 1 Power frequency electric fields 3 6 2 Power frequency magnetic fields 3 6 3 TV receiver and PC monitor fields 3 6 4 HF fields Isotropy of response Capacity and integrity of data storage O qQ 1Oo 0101010000 DW Nanna PERSONAL DOSIMETRY OF RF RADIATION 3 9 3 10 3 8 1 Battery life 3 8 2 Storage integrity 3 8 3 Spurious recording events 3 8 4 Charging reliability 3 8 5 Temperature recording Low temperature tolerance Summary of laboratory measurements Spot Measurement Evaluations 4 1 4 2 4 3 4 4 Narrowband spectral measurements 4 1 1 Data acquisition 2 Post processing 3 Field strengths in the PEM bands 4 Location and source circumstances 5 Field strengths outside the PEM bands 4 1 6 Effect of varying the PEM detection threshold Repe
43. Consent Form This appendix contains a document that was drawn to the attention of all 270 staff at the Health Protection Agency s Centre for Radiation Chemical and Environmental Hazards located in Chilton Oxfordshire UK in order to inform them about the study and encourage them to take part in the volunteer trial The document explains the commitment that would be required and gives sufficient information for the volunteers to provide written informed consent to take part The consent form is also included in this appendix 65 APPENDIX B Personal Dosimetry of RF Radiation Guidelines for Volunteers Background Some people believe that mobile phone base stations or other radio transmitters near to them have affected their health due to the radio waves radiation that they emit However there are considerable challenges in assessing the exposure of people to radio waves during their everyday lives and little research has been carried out This project is being carried out by NRPB under the Mobile Telecommunications and Health Research MTHR Programme It aims to develop new ways to assess the exposure of people to radio waves in order to extend the possibilities for scientific research We will need the assistance of volunteers and we hope you will consider taking part The project will evaluate a newly developed personal exposure meter PEM that is able to log people s exposures over time and as they move around We will n
44. EN 00 4r TT pa E m l o 4 ANA 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0115 020 025 030 035 040 us 4 Electric field strength V m Electric field strength V m 100 T 100 T Volunteer 3 Volunteer 4 a ia nt pr an o em al UMTSx 7 A LS Le ea Xr UMTSx 80 80 S Bl oa S 70 8 mM E 04 B 0 o 3 50 4 ee een er d a SS s E g 40 8 4 a a E 30 E 30 L 0 20 4 S a ee ere en 10 o ee ee EE EE eee Br i 0 o MM 0 00 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 1 100 I er Volunteer 5 Volunteer 6 EE 9 3 1 r UMTS W D 17 UMTSx 7 Qoo A A A Se eee o O A A A 2 S eer 8 7 4 4 7 B 0 2 B 0 o g Boa o S 3 su BSS Tl s j 8 0 4 4 4 a a SA Id zoo o MM OEC a L e 5 A ee eee lee 0 T eee I m USES I assis o em 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 I 100 I Volunteer 7 Volunteer 8 pu UMTSrx UMTSrx O ore s a m t mla e te da pas te e le kann T ap IUE ET All Locations z E fee 2 All Locations EMEN ee ee on Living Room
45. ENDIX A Volunteer Trial Protocol APPENDIX B Guidelines for Volunteers and Consent Form APPENDIX C Laboratory Test Results APPENDIX D Narrowband Spot Measurement Protocol APPENDIX E Volunteer Location Diary APPENDIX F Processed Personal Exposure Records F1 Personal Exposure Records FM F2 Personal Exposure Records TV3 F3 Personal Exposure Records TV4 amp 5 F4 Personal Exposure Records GSMtx F5 Personal Exposure Records GSMrx F6 Personal Exposure Records DCStx F7 Personal Exposure Records DCSrx F8 Personal Exposure Records UMTStx F9 Personal Exposure Records UMTSrx APPENDIX G Volunteer Questionnaire INTRODUCTION INTRODUCTION 1 1 Background There are considerable challenges in assessing the exposure of individuals in the general population to radiofrequency signals These include the number and range of sources involved not least the personal use of mobile phones and the effect of the environment on ambient signal strengths as people move around Some people believe that mobile phone base stations or other radio transmitters near to them have affected their health due to the radio waves that are emitted It is important to respond to such concerns scientifically but little information is available on personal exposures to radio waves and how they might vary with factors such as distance of residence from a base station Exposure data for radio waves are generally reported in the form of spot measurements i
46. First the PEM could be used as a validation tool for exposure modelling techniques in which case it only has to measure 56 CONCLUSIONS AND RECOMMENDATIONS the field of individual signals from known transmitters Second the PEM could be used as a tool to measure the total exposure of subjects to all RF signals in which case it needs to sum the signals in the context of the chosen exposure metric With due heed given to the recommendations made above the PEM should be useful for both applications 57 REFERENCES REFERENCES Antennessa 2005 A 9 band isotropic selective personal dosimeter DSP090 URL accessed on 19 September 2005 http www antennessa com Bergqvist U Friedrich G Hamnerius Y Martens L Neubauer G Thuroczy G Vogel E Wiart J 2001 Mobile telecommunication base stations exposure to electromagnetic fields Report of a Short Term Mission within COST 244bis URL accessed on 19 September 2005 http www cost281 org activities Short term mission doc Cooper TG Mann SM Khalid M Blackwell RP 2004 Exposure of the general public to radio waves near microcell and picocell base stations for mobile telecommunications URL accessed on 19 September 2005 http www hpa org uk radiation publications w series reports 2004 nrpb w62 htm Cooper TG Allen SG Blackwell RP Litchfield I Mann SM Pope JM van Tongeren MJA 2004 Assessment of occupational exposure to radiofrequency fields and radiat
47. HPA RPD 008 Personal Dosimetry of RF Radiation Laboratory and Volunteer Trials of an RF Personal Exposure Meter S M Mann D S Addison R P Blackwell and M Khalid ABSTRACT This report describes laboratory testing and volunteer trials that were carried out to evaluate a personal exposure meter PEM that has been developed to measure exposure of the general public to radiofrequency RF radiation as from telecommunications base stations broadcast transmitters and from personal use of mobile phones The PEM is designed to measure the electric field strengths of radio signals in several different frequency bands where there are known to be transmitters that contribute significantly to public exposure The laboratory tests showed the PEM had performance broadly in line with that required for its intended purpose however there were several issues requiring further attention These include that the PEM does not sum together properly the fields of multiple signals in the same band and that there appears to be a battery charging reliability problem The PEM has a 50 mV m detection threshold in its bands and data from the volunteer trials suggest that this may limit the ability to construct an exposure gradient over the range of likely public exposures within a study Nevertheless the PEM does seem able to discriminate the relatively high exposures of people who live near to mobile phone base station and television broadcast transmitters fr
48. L DOSIMETRY OF RF RADIATION Table 7 PEM behaviour with multiple signals in the same band PEM band Individual signal E fields V m E field of signals together V m Signal 1 Signal 2 Signal 3 Recorded Predicted Error FM 1 66 1 98 1 77 2 65 34 FM 0 71 0 66 0 65 0 97 33 FM 0 71 0 86 0 91 1 03 1 44 31 FM 1 66 0 66 1 59 1 78 11 GSMrx 1 92 1 83 1 75 2 65 34 DCSrx 1 47 1 34 1 31 1 99 45 UMTSrx 1 63 1 65 1 73 2 32 34 The results are shown as mean field the mean of the recorded field in the X Y and Z orientations for each signal and for the recorded and predicted results It can be seen that the instrument does not respond correctly to multiple in band signals the response being significantly less than the RMS sum Thus the instrument will under read in situations where there is more than one simultaneous transmitter in a band This is a situation that will occur in practice particularly with broadcast FM TV4 amp 5 signals and with multi operator cellular sites 3 6 Electromagnetic immunity 3 6 1 Power frequency electric fields A parallel plate system with a large uniform field volume was used to establish a 50 Hz electric field strength of 5 kV m The instrument was placed in the field such that its x y and z axes were aligned in turn with the field for 1 minute each and then the logged data were downloaded No responses were seen 3 6 2 Power frequency magnetic fields A 0 6 m diamete
49. NE oS 2 E 2 g OMS sS S amp e Samples above threshold a 20 10 0X E 0 00 Volunteer 9 GSMtx All Locations Living Room ee eee ee ee ees Kitchen Bedroom e ill Office Music Room 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 2 GSMtx ee e 2 E a 5 fos bess t o a AE eccE 2 8 0 a r a A ee nes a oO ENS ee er 10 C E 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 4 is Ee a e E H a 80 2 g 70 4 i 60 4 S 50 4 2 8 40 a E 30 D 20 4 10 4 0 0 00 100 T Volunteer 6 sun EA ATA eee cea m EE c EET 3 0 All Locations 5 Living Room A NE T ER Kitchen io Bedroom 8 229 230 9 9 9 NU n CCS Office 9 40 se 2 Outdoor amp 3 4 4 0 Pee E E MAA E A d ere cs lt SSeS eS See OO EE uz l on LE MM 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 100 e E ee RSS A ese Volunteer 8 80 4 70 4 60 4 50 4 40 4 30 Samples above threshold 20 10 0 GSMtx 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 86
50. RF spectrum from 80 MHz to 2 5 GHz were made at the identified locations according to the protocol in Appendix D 4 1 1 Data acquisition The measurements were made with an ARCS miniature biconical antenna mounted on a wooden tripod and connected via a coaxial cable to an Agilent E4407B spectrum analyser The spectrum analyser was controlled from a laptop computer so that all of its settings were applied automatically and the only manual involvement during the measurement was to change the antenna height and polarisation when requested by the software The measurement was made in 13 sub bands each configured with appropriate bandwidths frequency resolutions and dwell times in order to measure the RMS voltage corresponding to any signals present Given the bandwidths used this required measurements at 14788 spot frequencies in order to avoid any gaps in the spectrum The measurements were made in three orthogonal polarisations and at heights of 1 1 1 5 and 1 7 m above floor level leading to nine sets of measured voltages stored in a table on the computer This data acquisition took around an hour to perform 29 PERSONAL DOSIMETRY OF RF RADIATION 4 1 2 Post processing Post processing of the data was carried out in order to derive the voltage corresponding to the resultant field strength at each of the three heights and then a spatially averaged value over the three heights was calculated on an RSS basis Finally a peak search alg
51. STX 100 T Volunteer 1 pots DCStx A RU I rcu c een m 3 70 All Locations 5 META IA NA AAA A Living Room Poo A OR T DERE Dining Room 2 Bedroom 8 Be le ee a a a d a a a a a a a lel Office goo A A E mo a Outdoor 3 4 o o n ee earner eee eee ee ee i 10 1 AAA AAA AAA AAA AAAAAAAAAAAAAA4AAA4A4 lt lt K lt AA 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 I Volunteer 3 o Ce ae pos 80 i 2 2 70 iz 60 o S 50 4 2 G g 405 3 E 30 i D O A ce E al a e ea Ka n n l an nn n n ba de DED LAUS ___ C O e 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 5 con ESS SSS aaa eee nes a O E 7 27222 E ewL 2 o ote oi E ERN ee TT a JE A gt o 20 10 0 4 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 ji Volunteer 7 NN 4 15 04 Verus seg DCStx ELE eSI All Locations PEE ub Renee Living Room 8 Kitchen a O SS Bedroom o IS us M DN ND DET CE dE Office 3 Car E SS ey a Outdoor Train o i A Ea E eee T 10 i O EA A 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0
52. Signal Frequency and Modulation FM 1 8 0 5 98 MHz WBFM z75kHz deviation TV3 0 4 0 5 220 MHz pseudo Digital Audio Broadcasting TV485 5 4 0 4 591 25 MHz Analogue PAL TV TV485 1 2 0 5 602 MHz pseudo Digital TV GSMtx 1 0 1 0 895 5 MHz GSM uplink 1 8 slots active GSMrx 0 7 0 4 947 5 MHz BCCH 950 MHz 8 8 slots active DCStx 0 2 0 6 1747 5 MHz GSM uplink 1 8 slots active DCSrx 3 0 0 2 1842 5 MHz BCCH 1840 MHz 8 8 slots active UMTStx 0 4 0 1 1950 MHz 3G RT TDMA uplink UMTSrx 5 7 0 9 2140 MHz 3G RT TDMA downlink 3 5 Response to multiple signals A number of tests were conducted to investigate the response of the instrument to multiple signals occurring both in the same or in different bands Outputs from the signal sources were combined using an appropriate power combiner and the resultant signal fed to the amplifier The fields of each individual component and of the combined signal 20 LABORATORY TESTING were checked with the HI 6005 probe In each case the RMS sum of the individual components corresponded closely with the field of the combined signal Finally a spectrum analyser coupled to the system was used to confirm that no unwanted spurious mixing products were present 3 5 1 Signals in different bands These are summarised in Table 6 All signals were modulated appropriately except where indicated GSMrx signals were unmodulated CW simulating the BCCH channel Table 6 PEM behaviour with two sign
53. UK this broadcast service occupies frequencies around 220 MHz which falls within the TV3 response band of the instrument As it was not possible to generate a true DAB signal a signal with a similar RF spectrum was generated by a combination of 100 amplitude modulation using random noise and simultaneous 3 0 MHz FM modulation using a 1 kHz triangle wave A 2 5 V m field gave a mean response of 2 31 V m 3 4 4 Analogue TV broadcast UK TV broadcasting is allocated 8 MHz spaced channels numbered 21 to 68 in the frequency range 470 to 854 MHz A single transmission was simulated by using the output from a commercial TV test signal source which provided both modulated video and sound carriers at frequencies of 591 25 and 597 25 MHz respectively CH36 Measurements using a field of about 0 38 V m resulted in mean recorded fields of 0 71 V m with little difference between results using modulated or unmodulated signals This error of about 45 4 dB is not inconsistent with the CW data since it was made with a single instrument at a single frequency where there is a pronounced peak in the instrument s response The CW data is a mean for measurements made with all 8 PEMS and for the three frequencies used for the CW tests 18 LABORATORY TESTING 3 4 5 Digital TV broadcast A similar approach to the DAB test was used as a DTV signal source was not available A 602 MHz carrier 10096 amplitude modulated with white noise and simultaneous
54. als in different bands Signals applied separately Signals applied together Change gt 10 Signal 1 Signal 2 Signal 1 Signal 2 Band E Vm Band E Vm Band E Vm Band E Vm FM 1 76 TV3 1 00 FM 1 70 TV3 0 97 No FM 1 09 GSMrx 1 65 FM 1 10 GSMrx 1 66 No FM 1 13 TV4 amp 5 1 04 FM 1 11 TV4 amp 5 1 02 No TV485 1 52 GSMtx 1 43 TV485 1 47 GSMtx 1 41 No TV485 0 97 GSMrx 1 56 TV485 0 64 GSMrx 1 55 Yes TV4 amp 5 34 TV4 amp b5T 1 43 GSMrx 1 59 TV4 amp 5 0 96 GSMrx 1 57 Yes TV4 amp 5 33 GSMrx 1 59 DCSrx 1 28 GSMrx 1 62 DCSrx 1 28 No modulated ATV T unmodulated ATV The majority of combinations of signals in different bands showed no significant change between single signal and simultaneous signal results However there was one important exception Recorded field strength of signals in the TV4 amp 5 band were significantly reduced by about one third in the presence of a GSM 900 Base station CW equivalent to the BCCH transmission signal 3 5 2 Signals in the same band These were carried out in several bands and are summarised in Table 7 Two or three signal sources were combined using an appropriate power combiner and the resultant signal was fed to the amplifier The fields of each individual component and the combined signal were checked with the HI 6005 probe In each case the RMS sum of the individual components corresponded closely with field resulting from the combined signal 21 PERSONA
55. as linked to an Excel spreadsheet showed frequent patterns of alternating extra records these having an invalid date such as 1 8 1900 no time and zero values for the battery voltage and temperature However it was possible to use the volunteer data by removing these null records The manufacturer has since provided a revised software release which corrects this problem 3 10 Summary of laboratory measurements In summary the laboratory investigations have shown that the PEM has e Broadly correct field strength recording for single signals e An incorrect response to multiple in band signals Errors in recording TV485 fields when a GSM 900 base station signal is present e An inaccurate calibration for UMTS base station signals Greater than specified departures from isotropy 26 LABORATORY TESTING In addition for use within the UK some further issues need to be addressed including the lack of provision for the UK TETRA band In any country the instrument s lack of response to DECT cordless phone signals is likely to be a problem The specification for the PEM was taken to be the Technical Data table on page 4 of the supplied user manual most of which is repeated in the Certificate of Calibration included with each of the instrument This is also the same as in the separate DSPO90 data sheet available from the manufacturer The isolation referred to in the calibration certificate is for modulated signals within t
56. atability of narrowband measurements 4 2 1 Data acquisition and processing 4 2 2 Results Comparison of narrowband and PEM spot measurements 4 3 1 Data acquisition and processing 4 3 2 Results Summary of spot measurements Personal Measurements with the PEM 5 1 5 2 5 3 Analysis of the logged data 5 1 1 Corrupt records 5 1 2 Analysis by location 5 1 3 Sensitivity aspects Personal exposure mean estimates 5 2 1 Derivation 5 2 2 Resulting mean field values 5 2 3 Correlation with spot measurements Summary of personal measurements Volunteer Experiences and Perceptions 6 1 6 2 6 3 6 4 6 5 6 6 6 7 6 8 Strategies for wearing the PEM on the body Practicality of wearing the PEM Placing the PEM near the body Design of the PEM Perceptions of the PEM 6 5 1 Effect on behaviour 6 5 2 Self consciousness 6 5 3 Questions from others 6 5 4 Safety aspects Use of the PEM in future studies Other comments Summary of volunteer feedback Conclusions and Recommendations 7 1 RF performance aspects of the PEM 7 1 1 Frequency bands Filter selectivity and signal discrimination Response to modulated signals Response to multiple signals Isotropicity NNNN nan 1 1 S 1 7 1 6 Sensitivity 7 2 Functional aspects 7 2 1 Software 7 2 2 Battery charging 7 2 3 Storage integrity 7 3 Ergonomic aspects 7 3 1 Wearing 7 3 2 Size 7 4 Placement when not on body 7 5 Use in future studies 8 References APP
57. band spot measurement procedure Nevertheless more than 63 e 33 of the points are within 1 7 dB and so this suggests the spot measurements in the PEM bands are repeatable to within 3 4 dB 95 confidence Similarly good repeatability is shown in the bands other than those logged by the PEM except for the one described as other This is to be expected because the signals in the parts of the spectrum described by this category may well be intermittent or of variable power e g paging and PMR signals 4 3 Comparison of narrowband and PEM spot measurements 4 3 1 Data acquisition and processing As described in Section 4 1 the narrowband spot measurement data were gathered at heights of 1 1 1 5 and 1 7 m and then processed to derive a spatially averaged resultant RMS electric field strength For the analysis in this section the data for 1 5 m height were used alone The peak search algorithm was used to extract individual signal frequencies and powers and then the signal field strengths were summed to derive the total RMS electric field strength in each of the PEM frequency bands Immediately after each of the above narrowband spot measurements was made both on deployment and collection of the PEM used by the volunteer the miniature biconical antenna was removed from the tripod and a second PEM was mounted on a special jig at the 1 5 m height position In this way spot measurement data were acquired with the second PEM for compar
58. be possible to fit a lognormal distribution through the two data points in a meaningful way and the algorithm gives a doubtful electric field strength of 70 65 76 mV m for the mean Where less than ten values have been used to derive the mean estimate the means are shown in red in Table 16 and the number of values used is shown to the right of the confidence interval The data generally show that mean estimates could be constructed much more frequently when volunteers were outdoors or travelling in cars and trains than when they were indoors This is because the fields were above the detection threshold for more of the time at these locations as can be seen in the figures in Appendix F 5 2 3 Correlation with spot measurements The personal exposure mean estimates in the above section were compared with narrowband spot measurements taken at the same locations and the results are shown in Table 17 where values for both sets of data were available Table 17 Difference between personal exposure mean estimates when the diaries showed the volunteers were at the spot measurement locations and narrowband spot measurements Circumstances Electric field strength V m Ratio dB Personal measurement Spot measurement Volunteer Location TV485 GSMrx DCSrx TV485 GSMrx DCSrx TV4 amp 5 GSMrx DCSrx 1 1 41 77 56 86 77 1 0 2 8 2 42 104 68 104 119 0 0 4 9 3 31 115 131 170 113 3 4 1 3 4 6 5 9 5 18 3 2 2d 3 9 16 2 9 15 4 14 3 6 6 9
59. being combined using appropriate hybrid combiners before being fed to the amplifier A spectrum analyser Agilent E4407B was used to confirm the spectral purity and composition of the signal being fed to the GTEM cell Where modulated signals were required the majority of these could be provided by the E4483C vector signal source with the exception of broadcast television both analogue and digital and digital audio broadcasting DAB signals Analogue TV signals at a low level were obtained from a commercial TV test source including the sound and colour sub carriers For Digital TV and Radio modulation an effective equivalent was obtained by using 10096 12 LABORATORY TESTING amplitude modulation with random noise and simultaneous frequency modulation using a triangle waveform Figure 6 EMCO Model 5311 GTEM test cell as used to generate known exposure fields for testing the PEM To establish the field within the GTEM Cell at the test position a Holaday HI 6005 3 axis electric field probe was used with a fibre optic output This had a calibration traceable to national standards with a typical uncertainty of 1dB In general each particular field condition for one orientation frequency and or field level lasted for 1 minute the opening and closing of the chamber door being used to control the period By adhering to a strict time format and using periods of no field before and after each exposure a series of tests c
60. bile phones and continuous signals as from GSM base stations in order to improve its band selectivity Table 2 Order and configuration of the measurements made by the PEM Order Frequency band Antenna Processing Number of Modulation number samples 1 FM 3 Min 20 2 TV3 3 Min 20 3 TV485 3 Min 20 4 FM 2 Min 20 5 TV3 2 Min 20 6 TV485 2 Min 20 7 FM 1 Min 20 8 TV3 1 Min 20 9 TV485 1 Min 20 10 GSMtx 3 Max 70 TDMA 11 GSMrx 3 Min 20 12 GSMtx 2 Max 70 TDMA 13 GSMrx 2 Min 20 14 GSMtx 1 Max 70 TDMA 15 GSMrx 1 Min 20 16 DCStx 3 Max 70 TDMA 17 DCStx 2 Max 70 TDMA 18 DCStx 1 Max 70 TDMA 19 DCSrx 3 Min 20 20 DCSrx 2 Min 20 21 DCSrx 1 Min 20 22 UMTStx 3 Average 20 23 UMTSrx 3 Min 20 24 UMTStx 2 Average 20 25 UMTSrx 2 Min 20 26 UMTStx 1 Average 20 27 UMTSrx 1 Min 20 MATERIALS AND METHODS 2 3 Imputation of missing data In many cases where the PEM was used to log electric field strength over a period of time the resulting data sets showed the field strength was below the detection threshold of 50 mV m for an appreciable proportion of the time In such cases it is possible to develop numerical approaches to impute the most likely value of the mean field strength and to develop an associated uncertainty estimate The approach applied in this project was the same as that of Cooper et al 2004 with occupationally acquired personal exposure records The
61. ch of the spot measurement locations will take around an hour to survey and the equipment used is quite bulky We would therefore ask that sufficient space is cleared for us to make the measurements and carry the equipment through your house in safety In particular we would ask that any breakables are cleared away Volunteer Location Diary The volunteers will be given a diary to complete in which they will identify the periods of time during which they have been present at each location Feedback on the Instrument On final collection of their PEM and on repeat of the spot measurements the volunteers will be asked to complete a questionnaire on their perceptions of the instrument such as how easy it was to wear and whether it modified their behaviour in any way How to wear the Instrument The PEM is supplied with a belt clip so it can be worn to the rear of the waist and so that the arms do not swing in front of its sensing antennas The photograph to the right illustrates the position although the research team will be happy to provide further guidance If you find the PEM inconvenient to wear on occasions please take it off and place it near to you There will be occasions when it would not be safe to wear the instrument for example when driving or in bed It is also not waterproof Please remember when you take the instrument off and where you place it so we can gain feedback on these aspects 68 APPENDIX B
62. confirmed The PEM will be programmed to record once every 2 minutes for 10080 minutes i e 7 days and its clock will be synchronised to that of the survey laptop computer The PEM will be started logging and given to the volunteer Collection of PEMs The volunteer will be met at the agreed location time and the spot measurements will be repeated in the agreed sequence The PEM will be collected and the data contained within it downloaded to a computer without delay The completed location occupancy diary will be collected from the volunteer The completed feedback form will be collected from the volunteer Note on Spot measurements Each spot measurement will first be made according to the Spot Measurement Protocol for Environmental RF Exposures 80 MHz 2 5 GHz and then with a specially configured PEM other than the one to be given to the volunteer This second PEM is mounted on the same tripod as used for the first measurement and at a single height of 1 5 m Its clock is to be synchronised to that of the survey laptop computer and it is to record once every 5 seconds for the full day assume 9 hours The start time of each spot measurement with this PEM is noted from the laptop computer and the PEM is manually rotated in azimuth through 45 every minute The stop time is them noted when a minute has been spent at each angular position over a full 360 64 APPENDIX B APPENDIX B Guidelines for Volunteers and
63. d out to examine the instrument performance in multi signal RF environments and to determine immunity from commonly encountered electromagnetic fields in other bands including 50 Hz electric and magnetic fields and domestic TV and PC monitor fields The details of the laboratory testing and its results are presented in Section 3 2 1 2 Volunteer trial Ten volunteers took part in the trial and each carried a PEM with them for a period of one week The first volunteer acted as a pilot for the procedures and received a PEM in December 2004 The main trial involving the remainder of the volunteers then took place in the latter half of January and early February 2005 The protocol used for the volunteer trial is given in Appendix A and the results are presented in sections 4 to 5 2 1 2 1 Personal measurements The personal measurements and results obtained from the volunteers using the PEMs are described in Section 5 A recording interval of 2 minutes was used as this could be sustained for a week on a single full charge of the PEM batteries The volunteers kept a diary while they used the PEMs and this indicated where they were at a given time The diary was used to aid interpretation of the logged electric field strengths PERSONAL DOSIMETRY OF RF RADIATION 2 1 2 2 Narrowband spot measurements In advance of receiving the PEM the volunteers were asked to list locations where they expected to spend most of their time over the trial week Ty
64. d not feel self conscious or ill at ease when using the PEM and so a question was asked about this Generally the volunteers felt comfortable with the instrument although it was regarded as better to wear the PEM under clothes or in a bag so it was not on view lt was noted that the PEM does not really resemble any other device in common use and so it is unlikely to be mistaken for anything else However one volunteer thought it might have been mistaken for some form of listening device Another noted that it gave them a lop sided appearance when walking with it under clothes and felt that this might have aroused suspicion 6 5 3 Questions from others The volunteers were asked whether they had been questioned about the PEM by people other than friends family and immediate work colleagues Few situations had arisen and any questions had been simply answered One person had remarked to a volunteer that the PEM looked like a baby monitor 6 5 4 Safety aspects The volunteers were asked to identify any situations where they had felt it unsafe to use the PEM and had chosen not to wear or carry it with then Driving and sleeping had been highlighted as examples of such situations in the guidelines for volunteers see Appendix B Other situations encountered by the volunteers were in an aerobics class and in a nightclub where it was felt the PEM might get damaged or stolen One volunteer chose to leave the PEM at home on a day when they spent some time
65. d strength The measurements were also made at three heights 1 1 1 5 and 1 7 m above floor level over which the field strength was averaged to reduce the effect of spatial fading The use of appropriate dwell times and an average detector mode on the spectrum analyser helped to reduce the effects of temporal aspects of fading LABORATORY TESTING LABORATORY TESTING This section describes the laboratory testing that was carried out to determine the electrical performance of the PEM as a stand alone instrument for the measurement of RMS electric field strength in the frequency ranges of its various bands see Table 1 3 1 Introduction 3 1 1 Co ordinate system First it is necessary to define a co ordinate system to describe the orientation of the instrument relative to the test field The x y and z axes of the instrument are shown in Figure 4 and also correspond to the orientations of the 3 antennas within the instrument Figure 4 Definition of the x y and z axes of the PEM The corresponding X Y and Z orientations of the PEM defined as where the corresponding axis of the instrument is vertical are shown in Figure 5 as seen through 11 PERSONAL DOSIMETRY OF RF RADIATION the open door of the GTEM test cell The origin of the co ordinate system was taken as the centre of the 3 axis sensor a InputEnd End LoadEnd gt End 3 Figure 5 Orientations of the PEM inside the GTEM cell 3
66. e measurements made effectively at a point in time and space when where a person may be present The measurements are generally of the electric field strengths and plane wave equivalent power densities Mann et al 2000 Cooper et al 2004 Sometimes the data are processed to develop exposure quotients in relation to guidelines such as those from the International Commission on Non lonizing Radiation Protection ICNIRP 1998 Many of the available spot measurement data have concentrated on signals from base stations but have not included contributions to exposure from sources operating in other parts of the spectrum e g Bergqvist et al 2001 Ofcom 2005 There are some published wide spectrum data and these suggest that exposure can be constructed by concentrating on parts of the spectrum where there are certain sources that are either powerful widespread or used near to the body Mann et al 2000 If a personal exposure meter PEM could be developed and used to reliably characterise RF exposures of people this might offer a way forward for scientifically robust research into the health of people living near to base stations Such an instrument could be used to assess directly the exposure of subjects in a study or as a tool for the validation of modelling approaches 1 2 Aims and objectives This project aims to evaluate a newly developed and commercially available personal exposure meter the Antennessa DSP090 Antennessa 2005 that
67. e Section 3 9 and such records were edited out The numbers of correct and corrupt records for each volunteer are shown in Table 14 Table 14 Total numbers of correct and corrupt records in the downloaded personal exposure data from the volunteers Volunteer number Number of correct records Number of corrupt records 1 5019 21 2 5040 0 3 5040 0 4 4797 243 5 5040 0 6 5022 18 7 5040 0 8 4524 21 9 4433 607 10 All data lost Only an empty file could be downloaded for Volunteer 10 despite the volunteer recalling that the indicator on the instrument had been flashing to denote normal operation over the week lt was suspected that the battery had failed with this instrument due to incomplete charging see Section 3 9 5 1 2 Analysis by location Each record in the personal exposure data was assigned a tag number based on where the diary showed the volunteer was present at that point in time The first four tag numbers were the spot measurement locations for each volunteer but a small number of other tag values were assigned as well e g travelling in car and time spent outdoors This allowed the recorded data to be partitioned and analysed separately for each location 39 PERSONAL DOSIMETRY OF RF RADIATION The PEM records the field strength in V m and to two decimal places A histogram was formed for each partitioned data set and also for the entire data set arising from each volunteer
68. each volunteer on returning the PEM is included in Appendix G 6 1 Strategies for wearing the PEM on the body The volunteers were not given prescriptive instructions about how and when they were to wear the PEM on their body but the belt clip and the method of mounting on the waist were drawn to their attention in the guidelines for volunteers document see Appendix B The volunteers were encouraged to experiment with wearing the PEM in different ways to find what suited them and their clothing style Volunteers 1 4 5 8 and 9 did not use any form of bag rucksack and reported varying degrees of success in wearing the PEM using its belt clip It was apparent that apart from Volunteers 1 and 5 these volunteers had struggled and had found it preferable to carry the PEM when walking and place it beside them at other times Volunteer 2 wore the PEM inside a small rucksack behind the shoulders This meant that the PEM did not interfere with movement when standing and walking but the rucksack had to be removed when sitting down There would be the possibility of the PEM rotating inside the rucksack to face the body thus shielding its sensors but this was avoided by hanging the PEM by its belt clip from a loop of tape inside the rucksack This volunteer also placed the PEM inside a large handbag on occasions Volunteers 3 and 7 used the PEM inside a bag strapped around their waist This had the comfort advantages of using a rucksack and also a
69. eed ten volunteers to use the PEMs to gather data and provide feedback on their experiences and perceptions The PEM can be worn on the body when walking or placed near to the body when stationary The project is not a study of people s health and the radio waves involved will only be those to which the volunteers would normally be exposed Commitment from Volunteers Wearing the Instrument Each volunteer will be given a PEM to carry with them for a period of one week e g Monday Monday while it logs their personal exposure from environmental radio transmitters and also from their personal use of mobile phones On the days the instrument is handed out and collected there will be a substantial time commitment from the volunteers during their normal working hours hence agreement to take part should be sought from line managers Spot Measurements The volunteers will each be asked to list up to five locations where they generally spend most of their time when not travelling Examples might be their office bedroom living room and kitchen On the days the PEM is given to and collected from the volunteers the research team will visit these locations and make spot measurements of the ambient RF field levels with the survey equipment shown overleaf T Only the time of mobile phone use and the exposure level are collected No information is collected on the call content or destination 67 PERSONAL DOSIMETRY OF RF RADIATION Ea
70. ehaviour and if so in what way Situation Comment 94 APPENDIX G C Your experiences with the PEM 5 Were there any situations where you felt self conscious when wearing the instrument Situation Comment 6 Were there any situations where persons other than friends and family and immediate work colleagues asked you to explain what the instrument was Situation Comment 7 Were there any situations where you felt it was unsafe to use the instrument and chose not to either wear or carry it with you Situation Comment 95 PERSONAL DOSIMETRY OF RF RADIATION D Future use of the PEM 7 What would be the longest period that you would be prepared to use PEM in any future study delete all but your response from the list While keeping a diary Without keeping a diary Not at all one day three days one Not at all one day three days one week two weeks one month three week two weeks one month three months months 8 What would be the longest period that you would feel it reasonable to ask a member of the public to use the PEM if they were involved in a future study delete all but your response While keeping a diary Without keeping a diary Not at all one day three days one Not at all one day three days one week two weeks one month three week two weeks one month three months months 9 Do you have any final comments
71. eive such signals Conversely only bands except for those for GSM mobile phone signals should be able to measure continuous signals as for broadcast radio and television With GSM base station signals broadcast carriers BCCH are emitted continuously at the maximum power level whereas other secondary carriers TCH are emitted intermittently with variable slot occupancy and power levels Consequently the signal discrimination function will prevent the secondary carriers from being measured in the appropriate band unless they have full slot occupancy Also the signal discrimination function will not prevent such carriers from being measured in the mobile phone transmit bands when they have partial slot occupancy 53 PERSONAL DOSIMETRY OF RF RADIATION The performance of the PEM band filters should be improved if possible and appropriate measurement of the secondary carriers from GSM base stations should be addressed 7 1 3 Response to modulated signals Across all eight PEMs the responses to appropriately modulated signals were broadly correct within 3 dB for individually applied signals However when considered individually one of the PEMs was found to under respond to UMTS base station signals by about 6 dB The calibration certificates supplied with the PEMs appeared to show typical responses for the PEMs rather than individual calibrations traceable to standards The PEMs should be individually calibrated so that the
72. ements are made will move any possessions that could easily be damaged when making the spot measurements and carrying the equipment through my home will inform the study team if there are any special hazards in my home that they should be aware of Examples of such hazards would be low ceilings dogs locked in particular rooms or missing floorboards understand that can revoke this consent at any time during my involvement in the project Signature Date Print Name 69 APPENDIX C APPENDIX C Laboratory Test Results This appendix contains a typical dataset showing the response of a PEM to CW continuous and unmodulated signals of various frequencies Spurious out of band responses are shown in red Response in GTEM cell to CW field strength of 2 5 V m X orientation PEM s n 001 Frequency PEM frequency band MHz 49 FM 29 9 TV3 TV4 amp 5 GSMTx GSMRx DCSTx DCSRx UMTSTx UMTSRx 75 80 6 90 83 4 98 107 3 106 85 4 125 51 9 145 9 0 6 3 176 198 68 5 86 5 222 78 4 380 5 0 12 7 430 57 2 475 71 3 2 3 590 152 0 650 123 5 820 880 160 4 166 8 87 7 895 5 182 8 189 7 912 150 4 131 1 940 947 5 16 2 127 2 164 3 955 122 9 1300 1715 124 0 1747 5 112 2 T9 1780 129 7 44 5 1810 4 4
73. en completed This is likely to result in the loss of some data in a practical study and it should be avoidable through design The data storage hardware should be reconsidered to make lost data recoverable after recharging of the batteries If redesign is not possible critical care will have to be taken in studies to fully charge PEMs immediately before use and download the data immediately on completion of logging 55 PERSONAL DOSIMETRY OF RF RADIATION 7 3 Ergonomic aspects 7 3 1 Wearing Volunteers generally found the PEM difficult to wear on their waist and not supplying an alternative method for them to carry the PEM caused them difficulties in this trial Several of the volunteers devised other ways of carrying the PEM on their person such as in a small rucksack in a bag worn on the waist or in a shoulder bag On balance the shoulder bag seemed the most practical since it could be rotated about the body when the volunteers sat down or were driving etc The PEM should be supplied with a shoulder bag so it does not have to be worn on the waist The bag should allow for the PEM to be attached inside so it cannot rotate to face the body 7 3 2 Size There was also a strong consensus among the volunteers that the PEM needed to be smaller to be used as a body worn instrument Weight was felt to be less of an issue than size It is likely that concerns over the size of the PEM could be ameliorated by supplying it in some form o
74. epeatably at heights of 1 1 1 5 and 1 7 m Control Computer and GPIB Interface Equipped with software to completely control the spectrum analyser settings and indicate when the operator is to change the antenna height or polarisation Notebook To write down accurate descriptions of each measurement position including its juxtaposition with respect to the base station of interest 75 PERSONAL DOSIMETRY OF RF RADIATION Location Characteristics The antenna shall be mounted on the tripod and such that it is at least 1 m from any conducting or dielectric objects other than the ceiling which might disturb its calibration The cable to the antenna shall be routed such that it is kept as far away from the antenna cones as possible If necessary it is to be realigned after each rotation Band Settings The spectrum analyser shall sweep according to the sequence of band settings in the table below which have been chosen on the following basis 1 The resolution bandwidths are as wide as possible while still being narrow enough to reliably resolve individual signals 2 The number of points is chosen to give a spectral resolution significantly finer than the resolution bandwidth 3 The sweep time is chosen to give a dwell time for each measurement point that is long enough to give a stable reading with the relevant signal modulations Frequency MHz Number of Resolution Sweep time Frequency Notes Start Sto
75. ession E colection 1 deployment Difference 20 x a Data were only processed if both the opening and closing field strengths were above 10 mV m in order to reduce the effect of noise on the results due to signal strengths being near the narrowband measurement detection threshold The results of this analysis are shown in Table 12 Values differing by more than 3 dB are shown in red 33 SPOT MEASUREMENT EVALUATIONS Table 12 Differences between narrowband spot measurements of electric field strength on deployment and collection of the PEM Individual signal strengths have been summed over the PEM and other frequency bands for situations where both data sets were above 10 mV m and expressed in dB Spot measurement data source Ratio of closing to opening RMS electric field strength dB PEM frequency bands Other bands Volunteer Location FM TV3 TV4 amp 5 GSMtx GSMrx DCStx DCSrx UMTStx UMTSrx TETRA WLAN DECT Other 1 Home Living room Home Dining room Home Bedroom Work Office 1 02 0 96 0 20 1 08 0 76 0 44 1 10 0 71 0 70 0 16 0 89 2 03 3 12 2 71 2 31 3 01 Home Study room Home Kitchen Home Bedroom Work Office 0 36 1 82 1 46 Home Living room Home Kitchen Home Bedroom Work Office 0 56 3 20 2 13 4 49 0 87 Home Living room Home Kitchen Home Bedroom Work Office 3 01 0 18 0 94 0 01
76. f shoulder bag as recommended above 7 4 Placement when not on body An important technical question when the PEM is taken off the body and placed nearby is where is the most appropriate position to put it in order for it to give a realistic estimate of personal exposures It is not easy to answer this question definitively in the absence of detailed experimental testing in houses where the exposure levels are consistently above the PEM detection threshold Poor placement of the PEM will give systematically higher or lower exposures and it may be possible to develop guidance in order to avoid poor placement Positioning the PEM on window sills will lead to over estimates of exposure and positioning it at floor level will lead to under estimates Since the laboratory testing has shown the PEM is not perfectly isotropic it is also interesting to ask whether the PEM should face towards or away from a volunteer when placed nearby One volunteer remarked that they always placed the PEM on its back so it was less likely to be knocked over Using this orientation would avoid the need to decide whether the PEM should face towards or away from the person being monitored Guidance should be developed for users of the PEM in order to avoid inappropriate placement when it is not carried on the body 7 5 Use in future studies It is possible to envisage two broad applications for the PEM and the technical requirements are different for each application
77. form of printed A4 sheets attached to a clipboard 79 APPENDIX E Personal Dosimetry of RF Radiation Volunteer Location Diary Location Number Please specify if elsewhere Time interval Start 81 PERSONAL DOSIMETRY OF RF RADIATION APPENDIX F Processed Personal Exposure Records This appendix contains graphs summarising the personal exposure data acquired for each volunteer over their week long trial of the PEM The data are presented with nine graphs covering a single frequency band on each page so they can be visually compared 82 APPENDI XF F1 PERSONAL EXPOSURE RECORDS FM 10096 i Volunteer 1 EM ee ee A eee ee el Gs z o TRN a 3 70 All Locations 5 Living Room E IQ ee E E Aa a E Dining Room 2 Bedroom PEERS 7 O A dee Office P MEM A Cu Outdoor E 30 i o LE ee re T T TA fp ee SS h n a a ba a J i 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 I EXE mum om AENA A E mms mmm we Volunteer 3 Tux m c eee qp 2 a 9 E 6 oo Be See e Ye Sl s rer ee s r l cete cse 8 8 4 4 4 a 0R 1727224222229 o 20 ous Bee ss e a Seis es 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100
78. g room 4 15 9 37 20 46 10 Home Kitchen 13 ri 24 20 34 10 Home Bedroom 8 17 27 28 44 10 Work Office 5 6 7 11 15 31 PERSONAL DOSIMETRY OF RF RADIATION 4 1 4 Location and source circumstances It is not the purpose of this project to carry out an exhaustive investigation into the relationship between where a volunteer lives or the sources present in their home and their likely exposure level Nevertheless certain observations can be made which help to explain the spot measurement and personal exposure data see Section 5 2 All ten volunteers worked in the same building although in different offices some of which faced in different directions The office measurements all yielded field strengths below the detection threshold of the PEM in all of its bands The building was in a rural area and not close to any base stations Volunteer 1 lived significantly closer to a mast than any of the other volunteers at a distance of around 300 m The mast had antennas for several GSM and 3G operators as well as TETRA installed The signals in this volunteer s house were several times stronger than those in the houses of the other volunteers Volunteer 5 lived in Southeast London and nearer to VHF UHF broadcast radio masts than the other volunteers This is evident in that the field strengths in the FM and TV4 amp 5 bands were greater in the house of this volunteer than the houses of the other volunteers Volunteer 7 lived in a stone house
79. have to be within around 100 m of a PEM to produce a field above the PEM detection threshold It is notable that when Volunteer 5 travelled into London on a commuter train the effect of passengers using their mobile phones in the train was pronounced A range of exposures is evident among the volunteers in the GSM and DCS mobile phone transmit bands as Volunteer 2 recorded only 11 values above the detection threshold in these bands over the entire week whereas Volunteer 3 recorded 745 When Volunteer 3 was in the bedroom 651 of these values were recorded in the GSMtx band overnight The values were at a level too low to be associated with a phone in the same house and given that the house of Volunteer 3 was in a sparsely populated area this seemed a puzzling result Discussions with Volunteer 3 revealed that the PEM had been placed next to a DECT base station on a bedside table and so it seems likely that the signals from this have been registered in the GSMtx band As noted in Section 4 1 4 Volunteer 1 lived at a distance of around 300 m from a mast and significantly closer to a mast than any of the other volunteers This is evident in the personal measurements for the GSMrx DCSrx bands in that field strengths above the PEM detection threshold are recorded for an appreciable amount of the time up to 9696 in the case of this volunteers dining room The readings from the other volunteers exceed the detection threshold for much smaller amounts
80. he designated response bands it does not refer to out of band signals The PEM has responses outside the designated bands as described in Section 3 3 2 however these do not necessarily have serious consequences The manufacturer s axial isotropy specification would seem to refer to measurements made with rotation around only one axis Z which is vertical in a vertical or horizontal field The measurements in Section 3 7 were made using rotation in three different planes and about one axis to get a clearer picture of overall isotropy which is considerably poorer than the axial isotropy given in the specification One important shortcoming in the instrument s performance was the apparent inability to deal correctly with multiple in band signals as shown in 3 5 2 This is a serious shortcoming in view of the co siting of TV and FM broadcast transmitters in the UK where for example five TV signals of similar field strength would be found in most locations Co sited mobile phone base transmitters could also be problematical The use of a signal discrimination technique to assign signals in adjacent bands to the appropriate logging band for example TV4 amp 5 and GSMtx to overcome the shortcomings of the RF band filtering is an innovative technique It can however mean that the lack of a correct response to multiple in band signals also applies when signals are in adjacent bands since from the point of view of the PEM they may considered as occu
81. he total time to execute the above is 63 minutes Post processing The output from the procedure as saved in the specified file is a table in an Excel spreadsheet containing ten columns of data and 14788 rows below the headings The first column is frequency and the remaining columns are the measurements for each of the nine polarisation height combinations The post processing proceeds as follows and is achieved by a second computer programme 1 The three voltage measurements for each polarisation measured at a given height are reduced to total voltage values corresponding to the underlying total field This is done on a root sum squares basis i e Vio Viet ME V2 2 The total voltage values for the three heights are summed on a root sum squares basis and then divided by the square root of three This is because the intent is to spatially average the power density i e Vavg Vio Viorz Voi 3 3 A peak search algorithm is used to extract the voltages and frequencies of individual signals from the table of spatially averaged voltages 4 The signal voltages are converted to power densities taking account of the cable losses the antenna calibrations 5 Modulation specific correction factors are applied for each band to account for the restricted spectrum analyser bandwidths Within each UHF TV channel a test is carried out to identify whether a digital or analogue signal is present before selecting the appropriate correction
82. iagram showing the internal functional arrangement and this is shown as Figure 3 This shows that the PEM contains filters switches and amplifiers which separate the signals from its sensing antennas into the different bands and then pass them to detectors A2 F4 co PEA BD c3 SIS bone pa A3 F8 ANTENNAS gt RF BOARD Figure 3 Arrangement of the RF functions within the PEM The PEM has to measure from the three sensing antennas and in the nine bands so a total of 27 measurements must be performed during each recording interval The order in which these measurements are performed is shown in Table 2 In order to conserve the batteries each amplifier is only switched on for the part of the recording interval when it is needed to make a measurement PERSONAL DOSIMETRY OF RF RADIATION During the periods when the amplifiers are switched on samples are taken from the detectors every 330 us so individual GSM bursts of 577 us duration are not missed The maximum values from sequences of 70 samples are recorded for the GSMtx and DCStx bands and the minimum values from sequences of 20 samples are recorded for the remaining bands other than UMTStx For UMTStx the average value from a sequence of 20 samples is recorded as shown in Table 2 Hence in addition to band pass filtering the PEM uses special techniques to differentiate between TDMA pulsed signals as from GSM mo
83. ies were used In addition to three frequencies per response band lower mid and upper a number of out of band frequencies were also selected The test frequencies are all shown in Table 3 For operational reasons the set of frequencies were divided into Low Band below 1300 MHz and High Band 1300 MHz and above 14 LABORATORY TESTING Table 3 Test frequencies Frequency MHz PEM band UK application 49 Licence free devices 75 Runway ILS 90 FM FM Radio Broadcasting 98 FM FM Radio Broadcasting 106 FM FM Radio Broadcasting 125 Aircraft communication 145 Amateur radio Paging 176 TV3 PMR 198 TV3 PMR 222 TV3 DAB radio 380 TETRA 430 Amateur Radio Licence free devices 475 TV485 UHF TV Broadcasting 590 TV485 ATC radar 650 TV485 UHF TV Broadcasting 820 TV485 UHF TV Broadcasting 880 GSMtx GSM900 Handset 895 5 GSMtx GSM900 Handset 912 GSMtx GSM900 Handset 940 GSMrx GSM900 Base Station 947 5 GSMrx GSM900 Base Station 955 GSMrx GSM900 Base Station 1300 ATC Radar 1715 DCStx GSM1800 Handset 1747 5 DCStx GSM1800 Handset 1780 DCStx GSM1800 Handset 1810 DCSrx GSM1800 Base Station 1840 DCSrx GSM1800 Base Station 1875 DCSrx GSM1800 Base Station 1900 1925 UMTStx 3G Handset 1950 UMTStx 3G Handset 1975 UMTStx 3G Handset 2115 UMTSrx 3G Base Station 2140 UMTSrx 3G Base Station 2165 UMTSrx 3G Base Station 2450 Ovens WLAN Bluetooth
84. increments of 10 were used with a recording duration of 1 minute for each position Due to time constraints only a limited number of bands could be used the middle frequency for the FM GSMtx and UMTStx bands was chosen and the response examined for rotation in the X Y X Z and Y Z planes and about the X axis as shown in Figure 8 A summary of results is shown in Table 8 and a representative polar response is shown in Figure 9 Table 8 Isotropic and rotational tests Band Specification X Z X Y Y Z X axis FM 10 5 dB 0 2 13 8 dB 0 2 2 3 dB 1 4 9 9 dB 1 3 2 8 dB GSMtx 2 0 dB 5 3 3 6 dB 4 4 1 5 dB 4 4 3 8 dB 5 1 2 0 dB UMTSrx 12 5 dB 0 2 7 8 dB 3 4 0 2 dB 0 3 11 0 dB 2 9 8 7 dB 23 PERSONAL DOSIMETRY OF RF RADIATION Y Z plane polar response 2140 MHz radial scale dB relative to 2 5 V m 0 340 70 4 10 0 330 2 30 310 300 60 290 70 280 l 80 270 l 90 260 100 250 110 20 75 420 220 g o 210 450 200 195 uo 160 180 Figure 9 Representative polar response 3 8 Capacity and integrity of data storage Tests were undertaken to establish the practical battery life and data retention time using data acquisition parameters likely to be used in the volunteer study 3 8 1 Battery life For the first test seven instruments were used all with a recording interval of 120 s and data collection periods of between 5 and 7 days As the test was conducted in a
85. inner It was observed that the less conspicuous the instrument appears the better As such it will be important to avoid bright colours logos and labels in a study There were mixed feelings over the belt clip and it was noted that it was only of use to volunteers who wear trousers Some volunteers felt it was not strong enough and some experienced the PEM detaching unexpectedly particularly when they sat down One asked for a belt clip that could be used at 90 to the current one so the PEM could be clipped to braces 6 5 Perceptions of the PEM 6 5 1 Effect on behaviour Using a body worn instrument such as a personal exposure meter inevitably causes some changes in the behaviour of the wearer The important question is whether any of these changes in behaviour affect their exposure The volunteers were asked whether there were any situations in this trial where they felt using the PEM had appreciably modified their behaviour 49 PERSONAL DOSIMETRY OF RF RADIATION The main issues were associated with filling in the diary which as one volunteer observed rapidly became a habit and keeping the PEM near the volunteer One volunteer felt they had moved around less in their house due to the burden of knowing that they would have to complete a diary entry Another highlighted forgetting to pick up the instrument on leaving a location because they then had to retrace their movements 6 5 2 Self consciousness Volunteers shoul
86. ion Radiation Protection Dosimetry 111 2 191 203 ICNIRP 1998 Guidelines for limiting exposure to time varying electric magnetic and electromagnetic fields up to 300 GHz Health Physics 74 4 494 522 URL accessed on 19 September 2005 http www icnirp org pubEMF htm Mann SM etal 2000 Exposure to radio waves near mobile phone base stations NRPB R321 National Radiological Protection Board Chilton UK June 2000 URL accessed on 19 September 2005 http www hpa org uk radiation publications archive reports 2000 nrpb_r321 htm MTHR 2005 Mobile Telecommunications and Health Research Programme URL accessed on 19 September 2005 http www mthr org uk Ofcom 2005 Audit of mobile phone base stations National measurement programme of the UK URL accessed on 19 September 2005 http www ofcom org uk advice telecoms_ifc telephony_con_guides mob_phone_base_stat 59 APPENDIX A APPENDIX A Volunteer Trial Protocol This appendix contains the protocol that was developed for the volunteer trial and which the project workers followed The guidelines for volunteers and the consent form to which this appendix refer are included in Appendix B The narrowband spot measurement protocol referred to is in Appendix D 61 APPENDIX A Volunteer Protocol for MTHR PEM Evaluation Project This document specifies the procedures for the involvement of volunteers who will use the Personal Exposure Meter PEM i
87. ion broadcast and GSM900 mobile phones and base stations were of insufficient width to capture all such active sources in the UK The TV3 band supplied to account for VHF broadcast television transmitters in France is unnecessary for the UK where such sources are not present Important omissions from the present band coverage of the PEM include TETRA emergency services radio DECT cordless phones and WLAN wireless computer networking The GSM900 and UHF TV bands on the PEM should be widened to cover sources active in the UK Additional bands should be provided for TETRA base stations DECT and WLAN 7 1 2 Filter selectivity and signal discrimination The performance of the PEM band filters reflects the difficulties of designing such filters With continuous signals not having the TDMA characteristics of real telecommunications signals there is evidence of significant out of band responses including measurements of signals with frequencies in some of the PEM bands by other PEM bands i e they could be measured twice In particular there is negligible rejection of continuous signals in the GSM900 mobile phone band by the filters for the broadcast television band In practice the manufacturer has designed the PEM to discriminate between continuous and intermittent TDMA signals in order to effectively improve the band selectivity This means that only the bands designed to receive signals from GSM mobile phones should be able to rec
88. ir readings are traceable to appropriate standards with defined uncertainties 7 1 4 Response to multiple signals The PEMs do not respond correctly to multiple signals present in the same band and Seem to give a reading more equivalent to the field strength of the strongest signal than to the appropriate result which would be the RMS field strength of all the signals combined This is a particular problem where multiple signals of similar strength are present as from broadcast television stations in the UK that may produce up to five signals of similar strength The response of the PEM to multiple in band signals should be examined to see if it can be rectified If this cannot be done consideration should be given to the additional uncertainty introduced in any studies carried out 7 1 5 Isotropicity The PEM sensing element incorporates three orthogonal electric monopoles mounted on the faces of a cuboid Such a configuration is potentially able to give a fairly isotropic response however the PEM has circuit boards and batteries of appreciable size adjacent to its sensor and these can be expected to degrade its isotropicity Measurements confirm that the sensor is significantly less isotropic than the 0 5 to 2 5 dB according to band claimed in its specification Nulls of at least 13 8 dB depth and peaks of at least 5 3 dB were identified Given the design difficulties it is difficult to see how this situation could be greatly improved
89. ison with the narrowband measurement data The PEM was known not to be as isotropic in its response as the precision narrowband equipment and so it was rotated through 45 steps in azimuth at 1 minute intervals while recording every 5 seconds After the measurements the data acquired over the eight minutes required to give a full 360 rotation were averaged over the rotation Where some of the data were below the detection threshold of the PEM numerical methods were used to estimate the mean see Section 2 3 Hence some of the PEM spot measurement values are reported as below the 50 mV m detection threshold The results were processed to express the ratio of the PEM spot measurements to the narrowband spot measurements in dB where both were above 30 mV m and they are shown in Table 13 Given the availability of data on deployment and collection of the PEM at four locations for each of the ten volunteers this gave 80 sets of data for comparison 36 SPOT MEASUREMENT EVALUATIONS Table 13 Ratios of PEM spot measurement data at 1 5 m height to corresponding narrowband spot measurements of electric field strength at the same positions where both values were above 30 mV m Spot measurement Ratio dB data source PEM frequency bands Total Volunteer Day Location FM TV3 TV485 GSMtx GSMrx DCStx DCSrx UMTStx UMTSrx 1 Deployment Home Living room 1 32 3 98 3 62 1 Deployment Home Di
90. l exposure meter 2 2 1 Physical characteristics The external appearance of the PEM is as shown in Figure 1 It has the approximate dimensions of 19 5 x 9 5 x 7 5 cm and weighs 0 45 kg The belt clip can be seen on the rear and generally the instrument would be worn hanging from the belt Figure 1 Front and rear views of the Antennessa DSP090 personal exposure meter During the volunteer trials the instrument was worn and carried on the person in various ways or placed in their vicinity and this is discussed in Section 5 2 2 2 Frequency bands The PEM is designed to measure the electric field strength in nine different frequency bands as shown in Table 1 and it has three orthogonal sensors in order to provide an isotropic response PERSONAL DOSIMETRY OF RF RADIATION Table 1 Specified PEM measurement frequency bands Band name Active sources in the UK Range MHz FM VHF broadcast radio 88 108 TV3 Digital audio broadcasting 174 223 TV485 UHF broadcast television 470 830 GSMtx GSM mobile phones 900 MHz 890 915 GSMrx GSM base stations 900 MHz 935 960 DCStx GSM mobile phones 1800MHz 1710 1785 DCSrx GSM base stations 1800 MHz 1805 1880 UMTStx 3G mobile phones 1920 1980 UMTSrx 3G base stations 2110 2170 Note tx and rx are abbreviations for the transmitted and received radio signals from the point of view of a mobile phone The first three bands cover the frequencies used for broadcas
91. ller in order for it to be used as a body worn instrument Weight was felt to be less of a problem than size A better way to use the current version of the PEM seems to be inside some form of shoulder or waist mounted bag since this allows it to be moved around the body easily to suit posture and to make it more discreet People differ in their preferred ways of wearing the PEM and it would be useful to develop a carrying bag that can be worn in several different ways Completing the diary was regarded as a burden by the volunteers and a number of ways to reduce this burden were suggested The format of the diary could be improved since the clip pad with A4 sheets used in this trial was not ideal A smaller spiral bound pocket sized notebook would be better Alternatively the PEM could be developed to allow entry of the location identifier via a keypad thus avoiding the need for a diary 52 CONCLUSIONS AND RECOMMENDATIONS CONCLUSIONS AND RECOMMENDATIONS This section gives the conclusions from the laboratory testing and volunteer trials of the PEM Recommendations following from the conclusions are italicised 7 1 RF performance aspects of the PEM 7 1 1 Frequency bands The personal exposure meters PEMs supplied to this project had specified measurement bands appropriate for FM broadcast radio GSM1800 DCS mobile phones and base stations and UMTS mobile phones and base stations in the UK The bands specified for UHF televis
92. llowed the bag to be rotated around the waist to a convenient position when sitting standing driving etc Volunteer 6 used a soft binocular case worn across body to contain the PEM The loose shoulder strap allowed the PEM to be moved around the body to a convenient position even when driving Volunteer 10 used a similar arrangement with a small shoulder bag about the size of a 1 litre bottle Particular difficulties reported with wearing the instrument from the waist with its belt clip included a The clip pushed off the belt when the volunteer sat down b When hung from the waist of loose trousers with an elasticated waist the PEM developed a pendulum effect while walking C The clip did not fit with all types of clothes e g dresses without belts d It was not practical to wear the PEM during exercise e g running cycling dancing e The PEM kept hitting things inside the house as the volunteer turned 47 PERSONAL DOSIMETRY OF RF RADIATION 6 2 Practicality of wearing the PEM The responses to the question about practicality of wearing the PEM in various situations are summarised in Table 18 Volunteer 10 did not assign any scores Table 18 Practicality of wearing the PEM in various situations as reported by the volunteers Situation Volunteer number and gender 1 M 2 F 3 F 4 F 5 F 6 M 7 F 8 F 9 F 10 M Standing 2 4 1 5 1 1 1 3 1 Walking 2 4 2 5 1 1 1 3 2 Sat at table 3 5 3 5 4 4 1 4 4 Sat on sofa 4
93. ly frequency modulated by a triangle wave with a deviation of 6 5 MHz was used to give a spectral occupancy similar to that of a DTV broadcast multiplex Fields of 2 5 and 1 0 V m resulted in recorded mean fields of 2 79 and 1 14 V m respectively differences of 1 0 and 1 4 dB 3 4 6 GSM 900 handset This was simulated using the Agilent E4438C signal source set to GSM uplink mode with 1 8 active slots A modulated field of 2 49 V m during transmitted bursts resulted in a recorded mean field of 2 80 V m without modulation the recorded field was 3 28 V m GSMtx band with an additional response in the TV4 amp 5 band of 3 54 V m This additional response which was evident from the CW response data was not present when the modulated signal was used Similarly with a field of 1 01 V m the modulated signal gave a mean in band response of 1 20 V m and the unmodulated signal an GSMtx band response of 1 30 V m and an out of band TV4 amp 5 response of 1 47 V m 3 4 7 GSM 900 base station This was simulated using the Agilent E4438C signal source set to GSM uplink mode with between 1 and 8 active slots at a frequency of 950 MHz A second signal source generating a CW signal of 947 5 MHz was used to simulate the Broadcast Control Channel BCCH which is transmitted continuously If this BCCH surrogate was omitted transmissions with up to 7 active slots were not recorded The mean field strength recorded with the BCCH surrogate car
94. n 2 Bedroom 3 cos AAA Office y AAA C d Outdoor S 06 2224222222229 a EM lt l eer d ne e c er eleal ne 10 o LL i 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 88 APPENDIX F F7 PERSONAL EXPOSURE RECORDS DCSRX 100 T 100 T Volunteer 1 T Volunteer 2 BO DCSrx ar DCSrx O ner ee er rer er ern er A e ee a 3 70 All Locations 3 70 All Locations 5 BE AR Aa Living Room 5 DM Study E POE o5 AL te omes te ide do de ta did Dining Room E MEET EU m RETE Kitchen E Bedroom g Bedroom EZ IN ER ERA LIE MMC te rt A E e e e L E 50 Office 3 50 Office o X Ne ee re me EEE es a Outdoor E Outdoor E 30 4 A Nee eee ee o RR A e ho tiec ey a ee AS eR ee a o un N ale en re a een a l eines RES Estee m 10 10 0 Q6 m M 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T 100 T Volunteer 3 Volunteer 4 Mili o rie rie m voee ia nt pr an cer n d DCS e oo ae aye A ae xm pes 80 80 3 70 3 w S 7
95. ngth V m 84 APPENDIX F F3 PERSONAL EXPOSURE RECORDS TV48 5 100 T 100 T Volunteer 1 T Volunteer 2 90 1V485 20 TV485 A A Se Eee ae e A TE a 3 70 All Locations 3 70 All Locations 5 DE u ae Living Room 5 a A T A Study E wo er TERR ENTM EC Dining Room E ERST A A Kitchen E Bedroom E Bedroom ee a Me nat Peer A at era cl RA RR RS EA RA E ER 50 Office E 50 Office PESE E EA A A E me 2 es a Outdoor A Outdoor 5 06 2217227222222 9 B 0 2 29 L a EE SR erp sl is Qj el ee ee ead eee eee 10 10 O 2 ie a o O MEME A y 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m Electric field strength V m 100 T 100 T Volunteer 3 Volunteer 4 Mi ri rie m eres e e mr eg e pan er jn dl Tv4as EE a RAN PAR Tv4as 80 80 L M a 70 4 g 0 E 60 4 E ewL c 2 2 2 3 50 4 MEUM n 3 3 S S g 40 2m KT a a E 30 E 30 L D 20 4 Pe ee ee ee 10 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 0 00 0 05 0 10 0 15 0 20 Electric field strength V m 0 25 0 30 0 35 0 40 Electric field strength V m
96. ning room 1 24 0 76 4 30 1 Deployment Home Bedroom 2 89 4 66 0 42 3 59 1 Collection Home Living room 0 05 2 45 1 13 1 Collection Home Dining room 0 61 0 16 3 52 1 Collection Home Bedroom 3 14 0 13 3 52 4 Deployment Work Office 0 83 0 04 5 Deployment Home Kitchen 6 82 3 36 5 Deployment Home Bedroom 4 00 1 27 3 88 5 Collection Home Bedroom 4 49 4 91 6 Deployment Work Office 1 86 0 92 6 Collection Home Bedroom 0 64 0 04 8 Deployment Home Bedroom 1 43 4 38 8 Collection Home Bedroom 1 12 2 82 9 Deployment Home Kitchen 0 23 2 26 Band averages 5 11 1 27 0 45 0 50 0 42 2 69 denotes that numerical techniques were used to impute the PEM spot measurement value 37 PERSONAL DOSIMETRY OF RF RADIATION 4 3 2 Results Table 13 shows that there were only 15 of the 80 situations 7 of which were with the same volunteer where the electric field strength in the PEM bands was great enough to allow a comparison between spot measurements made with the PEM and the narrowband equipment The absence of measurable signals with the PEM in the FM TV3 and UMTStx bands also meant no comparisons could be made for these bands The data for Volunteer 5 suggested the PEM was under reading significantly in the TV band and so the original signal strength data were examined in the narrowband measurements These showed that the TV part of the spectrum consisted of many signals some of which had similar strengths e g
97. nstrument noting that it is intended to be worn by members of the public involved in scientific studies Please complete the form electronically and e mail it to Simon Mann Volunteer name Meter received Meter returned Spot measurement Locations Location 1 Location 2 Location 3 Location 4 Location 5 Please read all of the questions before deciding how to answer each of them A Your use of the PEM 1 Estimate the percentage of the time that you wore the instrument on your body when at each of the spot measurement locations and indicate where you placed it when not on your body Location time Position and distance from body when not worn 93 PERSONAL DOSIMETRY OF RF RADIATION B Your Perceptions of the PEM 2 How practical did you feel it was to wear the instrument in the following situations Situation Score Comments Standing Walking Sat at table Sat on sofa Reclining Scoring System 1 No problem 2 Minor problems 3 Somewhat problematic but still practical on balance 4 Fairly impractical 5 Totally impractical 3 How did you feel about the design aspects of the instrument and what would you recommend for its future development Aspect Comment Weight Appearance Belt clip Other 4 Were there any situations where you felt using the instrument appreciably modified your b
98. nto the study It specifies the procedures to be used by the study investigators in managing the volunteer inclusion informed consent and supporting the acquisition of data Initial approach to potential volunteers An initial informal approach to potential volunteers is to be accompanied by distribution of the Guidelines for Volunteers document to those who show an interest in the project Those who are still interested in taking part will then be given a few days to discuss taking part with their line management and others who may be affected such as their family after which they will be contacted again Inclusion of volunteers Those volunteers who are able to take part are to be individually talked through the Guidelines for Volunteers document by the investigator This is to ensure they fully understand the nature of their involvement and to answer their questions The spot measurement equipment will be shown to the volunteers and they will identify through discussion with the study investigator the five locations where they normally spend most of their time and where spot measurements will be made on distribution and collection of the PEM The arrangements for making the spot measurements safely particularly when in the houses of the volunteers will be discussed and the importance of removing any clutter and breakables from the locations and access ways will be stressed The volunteers will be shown a PEM and encouraged to consider
99. o keep a diary One volunteer felt the PEM should be supplied with a bag having versatile straps that can be used around the waist or over the shoulder It was suggested that it would be helpful to be able to confirm that the PEM was working at any point in time especially with long recording intervals where the light only illuminates infrequently Some way of linking the unit s internal clock with events was suggested as possibly useful either with the unit displaying its own time so that can be used with a written diary or with the provision of event buttons The possibility of entering location data on the PEM via a keypad instead of keeping a written diary was also raised 51 PERSONAL DOSIMETRY OF RF RADIATION One volunteer developed an interest in the detailed technical aspects of the PEM It was suggested that the PEM could use a memory card similar to that in digital cameras in order to facilitate easy transfer of data and storage in a non volatile format The volunteer also suggested that developing a charging cradle on which the PEM could be placed at night next to people s beds would allow trials of longer than a week and with shorter recording intervals 6 8 Summary of volunteer feedback The feedback from the questionnaires highlights the importance of ensuring that a comfortable way of wearing the PEM is established for study participants There was a strong consensus among the volunteers that the PEM needs to be sma
100. ocations in the FM band and at 2 locations in the TV3 band The field strengths in these bands never exceeded the 50 mV m PEM detection threshold Signals detectable with the narrowband system were found in the TV4 amp 5 band in all but three of the volunteers houses but the field strengths were only above the PEM detection threshold in the house of one volunteer Signals in the GSMtx and DCStx bands were rarely detected with the narrowband system and the field strength was always below the PEM detection threshold suggesting that the signals were from distant phones No signals were detected in the UMTStx band Signals were detected with the narrowband system at 34 and 24 out of 40 locations in the GSMrx and DCSrx bands respectively However the fields were only above the PEM detection threshold at three locations all in the house of Volunteer 1 Signals were detected with the narrowband system in the UMTSrx band at three locations all in the house of Volunteer 1 and the signal strengths at two of these locations were above the PEM detection threshold Root sum squares Each voltage was squared then summed together and the square root taken 30 SPOT MEASUREMENT EVALUATIONS Table 10 Spot measurements of signal electric field strengths made with the narrowband measurement system spatially averaged over heights of 1 1 1 5 and 1 7 m and summed across the PEM and other relevant frequency bands Signals in red are too weak to be
101. of time typically less than 1096 in these bands and the strongest signals seem to be recorded when the volunteers are outdoor or in their cars rather than in the home Few signals are recorded above the detection threshold in the UMTSrx band 5 2 Personal exposure mean estimates 5 2 1 Derivation Numerical methods as described in Section 2 3 were used to impute the mean electric field strength in the PEM frequency bands for the different volunteers and locations The approach was likely to be unreliable for the intermittent nature of the signals in the three mobile phone tx bands and so these bands were excluded Also the FM and TV3 bands could not be considered due to lack of data above the detection threshold The personal exposure mean estimates are shown in Table 16 with the 9596 confidence interval as reported from the numerical algorithm not a measurement uncertainty Assumptions 2 W power as for GSM900 peak unity gain for the antenna and an inverse square dependence of power density on distance This gives an electric field strength at 100 m 77 mV m 42 PERSONAL MEASUREMENTS WITH THE PEM Table 16 Personal exposure mean electric field strength estimates in mV m 95 CI for various volunteer band location combinations Volunteer Location number Frequency band number TV485 GSMrx DCSrx UMTSrx 1 All 33 30 35 73 64 84 53 47 61 1 Living room 41 38 45 77 67 88 56 49 64
102. om those of people living elsewhere Currently it cannot measure signals from TETRA base stations wireless computer networks WLANs and digital cordless phones DECT but these capabilities could be added Recommendations have been made that should improve the usefulness of the PEM for validating exposure modelling techniques and as a monitor to assess RF exposure This study was funded as an adjunct to The Mobile Telecommunications and Health Research Programme MTHR www mthr org uk O Health Protection Agency Approval September 2005 Centre for Radiation Chemical and Environmental Hazards Publication October 2005 Radiation Protection Division 28 00 Chilton Didcot Oxfordshire OX11 ORQ ISBN 0 85951 568 0 This report from HPA Radiation Protection Division reflects understanding and evaluation of the current scientific evidence as presented and referenced in this document CONTENTS 1 Introduction 1 1 1 2 1 3 2 1 2 2 2 3 2 4 3 1 3 2 3 3 3 4 3 5 3 6 3 7 3 8 Background Aims and objectives Report structure Materials and Methods Overall study protocol 2 1 1 Laboratory testing 2 1 2 Volunteer trial Personal exposure meter 2 2 1 Physical characteristics 2 2 2 Frequency bands 2 2 3 Format of logged data 2 2 4 Recorded field strengths Imputation of missing data Narrowband spot measurements Laboratory Testing Introduction 3 1 1 Co ordinate system 3 1 2 Test methods
103. orithm was used to form tables of signal frequencies and field strengths and apply correction factors to account for power loss due to the restricted spectrum analyser bandwidths The individual signal frequencies and RMS electric field strengths were summed to evaluate the total RMS electric field strength present in each of the PEM s specified frequency bands see Table 1 The RMS field strength was also evaluated in the TETRA base stations band 390 395 MHz the 2 4 2 5 GHz band used by wireless local area networks WLANs and the DECT cordless phones band 1880 1900 MHz Finally the RMS field strengths of signals that did not fall in any of these 12 bands were accumulated as other The resulting narrowband spot measurement data taken on the day the PEM was given to each volunteer are shown in Table 10 Field strengths shown in blue should be measurable by the PEM as they are in its specified bands and above the 50 mV m detection threshold Fields shown in red and black should not be measurable by the PEM the former because they of insufficient strength and the latter because they describe signals of frequencies outside the specified bands The total column includes the nine PEM bands and the four other bands in the table 4 1 3 Field strengths in the PEM bands Table 10 shows that signals were not always detected in the PEM bands even with the sensitive narrowband system Signals were detected at 12 out of the 40 spot measurement l
104. ould be conducted in one session After downloading the session results each individual exposure could then be clearly and unambiguously identified For the majority of tests the instrument was configured for a 5 s measurement cycle and a maximum data collection time of 120 minutes To gain some idea of variability between instruments the unmodulated field tests as described later were conducted on each of the 8 instruments supplied but the other tests were generally conducted on one or two instruments 13 PERSONAL DOSIMETRY OF RF RADIATION 3 2 Acceptance tests Before the main laboratory investigations started all instruments were subjected to an acceptance test to confirm the completeness and functioning of the supplied components This comprised an initial visual inspection electrical safety testing of the supplied battery charger checking and installation of the supplied software CD and initial battery charging before a functional test Each instrument was then configured for a 5 s measurement period with a data collection period of 100 minutes and the internal clock was set to the current time UTC An initial file download operation was necessary as the instruments contained data when they were supplied Each instrument was placed in the GTEM cell started and left for 5 minutes with no field applied to establish the no field response The acquired data were then downloaded from the instrument and the results recorded The
105. p points bandwidth seconds Step 80 154 95 1500 30 kHz 4 5 50 kHz Includes FM Radio 155 389 9 2350 100 kHz 100 kHz 394 998 10 kHz TETRA base stations 469 9 100 kHz 854 1MHz UHF Television 923 5 1MHz 924 100 kHz GSM base stations 961 5 3 MHz 1804 100 kHz GSM base stations 1880 25 100 kHz DECT 2108 5 2170 UMTS base stations 2500 Includes WLAN 1 2 3 4 5 6 7 8 The video bandwidth shall be set to automatic and the detector to RMS mode The amplitude scale shall be in logarithmic voltage with 10 dB per division and a reference level of 5 dBm 125 7 mV The input attenuator shall be set to 5 dB Survey protocol The measurement equipment shall be set up at the survey location with all cables connected The height of the antenna is to be 1 1 m and it is to be in x polarisation 76 APPENDIX D The spectrum analyser is to be switched on and the control software is to be set running on the computer The instructions given by the software including specifying a filename in which the data are to be saved are to be followed The software will execute the sequence of sweeps listed in the above table and then ask the operator to move the antenna to the next polarisation y then z and height 1 5 then 1 7 m When all nine sets of data have been acquired and saved the measurement is complete The computer and spectrum analyser are to be switched off and the equipment dismantled T
106. pically these locations included their office at work and their kitchen bedroom and living room at home Spot measurements were made at these locations on the day of deployment and on the day of collection of the PEM The measurements and results are described in Section 4 The spot measurements were of signal electric field strengths over the RF spectrum from 80 MHz to 2 5 GHz and they were processed to derive the total field in each of the frequency bands of the PEM see Table 1 The equipment used for the narrowband spot measurements is described in Section 2 4 The data on deployment and collection of the PEM were compared to determine the repeatability of the narrowband spot measurement procedure as in Appendix D and to identify any significant changes in the RF exposure environment The data on deployment were also analysed to determine whether the signals present in the PEM frequency bands were of sufficient strength to be logged and whether there were any signals present in bands not covered by the PEM 2 1 2 8 PEM spot measurements The narrowband spot measurements both on deployment and collection of each PEM were compared with the readings from a second PEM placed at the same position This was to determine the degree of correlation between the PEM readings and precision measurements under realistic exposure conditions rather than in a laboratory 2 1 2 4 Spot measurements and personal exposures The narrowband spot measurements
107. quiet RF environment the number of events logged was taken as a measure of the intrinsic noise level of the instruments The results are summarised in Table 9 24 LABORATORY TESTING Table 9 Battery duration and data retention Serial no 1 2 3 4 5 6 7 Data collection 5 5 6 6 7 7 T days Download after 5 7 6 7 7 8 10 day Final battery 3 82 3 88 3 87 3 83 3 81 3 84 voltage Data retained Yes Yes Yes Yes Yes Yes No No of events 1 7 None 200 6 None Event bands TV48 amp 5 TV485 TV48 amp 5 TV485 22 UMTSrx 3 8 2 Storage integrity It is clear that a 7 day data collection period is practical as long as the instrument is downloaded promptly The 10 day download failed as shown in the final column of Table 9 the software showing no data present Further investigation showed that if the battery becomes discharged the logged data become irrecoverable This is a serious shortcoming in a device designed for use away from the controlled environment of the laboratory It is also surprising since a flash memory device which is inherently non volatile is used for logged data storage The manufacturer has explained that some data needed to access the logged results are not stored in a non volatile form 3 8 3 Spurious recording events One unit number 4 recorded a large number of events defined in this context as isolated records above the 0 05 V m threshold Unit numbers 1 2 and 5 also recorded
108. r Helmholtz coil system was used to establish a 50 Hz field of 100 pT The instrument was placed in each of the X Y and Z orientations in turn in the field for 1 minute and then the data were downloaded No responses were seen 3 6 3 TV receiver and PC monitor fields In order to evaluate the possibility of interference from these sources instruments were placed close to a number of PC monitors and domestic TV receivers No responses were seen 3 6 4 HF fields The instrument response to frequencies of 1 8 and 27 12 MHz was examined by exposing the instrument in a TEM cell exposure system The method was similar to that employed for the CW tests in the GTEM cell system and a field of 5 0 V m was used Whilst there was no response to the 1 8 MHz field for the 27 12 MHz exposures the mean field was recorded as being 0 59 V m in the FM band 22 LABORATORY TESTING 3 7 Isotropy of response In order to investigate the response of the instrument in differing orientations a jig was prepared Figure 8 which allowed rotation in different planes This was used in the GTEM cell where the field was substantially in the z direction te X Z Polar E X Y Polar E Y Z Polar T E X Axis Input End Load End gt Figure 8 Rotational jig and rotations used to examine isotropy of the PEM when inside the GTEM cell The experimental method was similar to that for other tests using a nominal 2 5 V m field Rotational
109. rier alone was very slightly higher than that recorded when both signals were present irrespective of the number of active slots This implied that the instrument response for multiple signals may be in error and it was investigated subsequently see Section 3 5 2 3 4 8 GSM 1800 handset This was simulated using the Agilent E4438C signal source set to GSM uplink mode with 1 8 active slots The results were similar to those for the GSM 900 handset tests 3 4 9 GSM 1800 base station In a similar way to the GSM 900 base station signal test described above the GSM framed transmissions were generated on a frequency of 1840 MHz with the surrogate BCCH carrier being on 1842 5 MHz Identical behaviour was seen the field strength recorded when just the BCCH carrier was present being very slightly higher than that when both signals were present 3 4 10 DECT 1800 DECT Digital Enhanced Cordless Telecommunications devices in the UK are allocated 10 frequencies in the range 1880 1900 MHz Most modern cordless telephones use DECT transmissions Although not specifically mentioned in the instrument specifications DECT was included in the tests as use of such phones could contribute significantly to personal exposure The Agilent signal source was set to generate a DECT modulated signal at 1890 MHz with 1 of the 12 possible timeslots active No response to fields of either 2 5 or 1 0 V m was seen 19 PERSONAL DOSIMETRY OF RF RADIATION
110. rring in the same band 27 SPOT MEASUREMENT EVALUATIONS SPOT MEASUREMENT EVALUATIONS Each of the ten volunteers was asked to identify locations where they expected to spend an appreciable proportion of the time over the week in which they would use a personal exposure meter PEM Spot measurements were made at four of these locations on the days of deployment and collection of the PEM The aims of making the spot measurements were as follows e To identify the dominant signals contributing to exposure at each location and to identify whether the PEM could be expected to measure these signals based on its specified sensitivity and filter bandwidths e To compare spot measurements made at the beginning and end of the week in order to examine repeatability of the measurement procedure and to assess whether the RF spectrum had changed significantly over the week e To compare spot measurements made with a precision narrowband system with those made with a PEM at the same fixed position in order to examine their degree of correlation in a real exposure situation e To provide spot measurement data at the locations that could be compared with personal exposure data gathered when the volunteer diaries showed that they were using the PEM at the same location The results from analyses of the spot measurement data are presented in this section 4 1 Narrowband spectral measurements Spot measurements of signal strengths over the
111. rs and a difficult question will be whether this band should be included in the PEM 32 SPOT MEASUREMENT EVALUATIONS 4 1 6 Effect of varying the PEM detection threshold Some indication of the improvement in signal measurability that could be gained from lowering the PEM detection threshold is shown in Table 11 This shows that even if the detection threshold could be lowered to 10 mV m there would be many locations where the field strength in the PEM bands would still be below the detection threshold Table 11 Number of the 40 spot measurement locations where signals would be measurable for a given PEM detection threshold PEM frequency Electric field strength detection threshold in mV m Ga 50 40 30 20 10 FM 0 0 1 1 4 TV3 0 0 0 0 2 TV485 3 3 4 6 15 GSMtx 0 0 0 0 2 GSMrx 3 4 5 7 17 DCStx 0 0 0 0 0 DCSrx 4 6 9 15 22 UMTStx 0 0 0 0 0 UMTSrx 2 2 3 3 3 4 2 Repeatability of narrowband measurements 4 2 1 Data acquisition and processing The spot measurement data in Section 4 1 were taken on deployment of the PEM The measurements were also repeated a week later on collection of the PEM and the two sets of measurements were compared to examine the changes in band field strengths Since the interest was in relative rather than absolute differences in the field strength levels the differences between the field data for each PEM band were expressed in decibels dB according to the following expr
112. siderably below the 50 mV m detection threshold of the PEM This suggests that there would be little merit in adding further bands to the PEM beyond those listed in the previous paragraph Comparison of the spot measurements made on deployment with those made a week later on collection of the PEM showed good repeatability of the spot measurement procedure with the data overall showing repeatability to within 3 4 dB based on 95 confidence with a normal distribution Spot measurements made with the PEM at the same locations as the narrowband equipment supported the observations made during the laboratory testing that the PEM does not sum the strengths of multiple in band signals correctly 38 PERSONAL MEASUREMENTS WITH THE PEM PERSONAL MEASUREMENTS WITH THE PEM Ten volunteers each carried a personal exposure meter with them for a period of one week while it logged the electric field strength in its bands Table 1 once every two minutes The volunteers kept diaries over the week so that the logged readings could be correlated with their whereabouts and the results are reported in this section A template diary is shown in Appendix E 5 1 Analysis of the logged data The logged data were downloaded from the PEM and exported into an Excel spreadsheet so they appeared as in Figure 2 5 1 1 Corrupt records Initially the 5000 logged data points were inspected for corrupt records e g where the temperature fell below 10 se
113. sponse to a frequency in the DCStx band 1810 MHz DCSrx and UMTStx responses to an OOB frequency 1900 MHz 000 G These responses reflect the difficulty of making perfect band pass filters In some cases these OOB responses are of little consequence For example the FM band responses render the instrument sensitive to pager PMR and other signals which whilst outside the specified band nevertheless represent actual exposure of the subject The same cannot be said of the TV4 amp 5 response to GSM frequencies c or the DCS responses d and e since recording a single frequency exposure in both bands at once will increase the recorded exposure However the response for modulated signals addresses these problems as will be described later see Section 3 4 3 3 3 Linearity In band linearity tests were performed on two instruments For each of the nine response bands a field of 2 5 V m was established in the test position in the GTEM cell using the appropriate mid band test frequency The cell line voltage was then reduced by a factor of two to generate a field of 1 25 V m and this process repeated a further three times resulting in a final field of about 0 156 V m No significant non linearity was seen A more detailed investigation of the linearity at low fields in the FM and TV4 amp 5 bands was carried out using one instrument This used a similar method to the earlier linearity 17 PERSONAL DOSIMETRY OF RF RADIATION
114. t radio and television in France however the TV3 band is not used for television in the UK The TV3 band is used for a variety of other applications in the UK including digital audio broadcasting DAB in the range 217 230 MHz The FM and TV485 bands have the same applications in the UK as France although the UK TV band extends somewhat higher in frequency to 854 MHz The other bands seem broadly as expected although the GSM bands do not include the E GSM portions as used in the UK which extend down to 880 and 925 MHz for the GSMtx and GSMrx bands respectively Other frequencies not covered by the PEM include those used by cordless phones wireless networks for home work computers and Terrestrial Trunked Radio TETRA Early analogue cordless phones which are still widely used in the UK operate around 30 50 MHz and tend to transmit only when a call is made More recent digital cordless phones are based on DECT Digital Enhanced Cordless Telephony and the base stations transmit in the 1880 1900 MHz band even when calls are not being made Similarly the transmitters in wireless equipped computers tend to transmit even when no data are being transferred The frequencies used are in the 2400 2500 MHz band with most wireless computer communication systems presently in use TETRA is a cellular radio system used by the emergency services in the UK and its base stations use the 390 395 MHz band 2 2 3 Format of logged data The user can program
115. t used a newly developed RF personal exposure meter PEM whose operation is detailed The PEM records the electric field strength present in several different frequency bands at regular intervals so it can be downloaded at a later time Spot measurements of electric field strength were made with an antenna on a tripod connected to a spectrum analyser at positions where the PEM was used and this system is described Finally a questionnaire that was used to gather feedback from the volunteers who used the PEM in the trials is outlined 2 1 Overall study protocol The study comprised an initial series of laboratory investigations to characterise the functionality and electrical properties of the PEM followed by a volunteer trial to investigate the practical aspects of its use As the volunteer trial progressed further laboratory investigations were carried out to examine matters that arose 2 1 1 Laboratory testing Eight PEMs were supplied to the study and the initial laboratory testing of these instruments was carried out in the final quarter of 2004 Functionality tests examined the performance of the PEM software as well as the hardware and suggestions for improvement were made to the manufacturer leading to revised issues of the software during the course of this project Tests of electrical performance examined parameters such as filter selectivity response to modulated and unmodulated signals linearity and isotropy Tests were also carrie
116. the PEM to make measurements at set recording intervals and for a set total duration within the constraints of its battery life and available memory In addition to the field strengths in the nine bands each record contains a measurement of the battery voltage and the temperature so that the data appear as in Figure 2 when exported to Microsoft Excel MATERIALS AND METHODS Sample Date Time Battery mV Temperature C gi TV3 TV4 amp 5 GSMtx GSM DCStx DCSRx 1 10 12 2004 13 02 46 4017 23 9 0 05 0 05 0 05 0 05 0 19 0 05 0 07 0 05 0 05 2 10 12 2004 13 04 46 3989 23 7 0 05 0 05 0 05 0 05 0 21 0 05 0 10 0 05 0 05 3 10 12 2004 13 06 46 4030 23 5 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 4 10 12 2004 13 08 46 4062 23 3 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 5 10 12 2004 13 10 46 4010 23 2 0 05 0 05 0 05 0 05 0 08 0 05 0 09 0 05 0 05 6 10 12 2004 13 12 46 3986 22 9 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 7 10 12 2004 13 14 46 3999 23 0 0 05 0 05 0 05 0 19 0 06 0 05 0 05 0 05 0 05 8 10 12 2004 13 16 46 4008 23 0 0 05 0 05 0 05 0 05 0 06 0 05 0 05 0 05 0 05 9 10 12 2004 13 18 46 3992 22 9 0 05 0 05 0 05 0 05 0 07 0 05 0 05 0 05 0 05 10 10 12 2004 13 20 46 4014 22 9 0 05 0 05 0 05 0 05 0 08 0 05 0 05 0 05 0 05 Figure 2 Example of logged data from the PEM 2 2 4 Recorded field strengths The exact physical meaning of the field strengths recorded by the PEM depends on its detailed internal design The manufacturer supplied a d
117. the detection threshold for the remaining PEM frequency bands The sequence of location numbers in the table reflects the order in which the locations appear in the key to the relevant graph in Appendix F 40 PERSONAL MEASUREMENTS WITH THE PEM Table 15 Number of personal measurements above the PEM detection threshold for various volunteer band location combinations and the corresponding total number of records Frequency Location Volunteer number Total for all band number 4 2 3 4 5 6 7 8 9 volunteers TV485 All 66 19 225 84 1809 33 67 104 205 2612 1 T 1 0 2 252 1 12 1 10 2 16 0 2 1 183 1 0 0 4 3 27 1 4 3 1215 0 0 15 1 4 3 1 2 0 1 1 1 4 0 5 14 13 51 27 40 10 39 67 45 6 2 3 166 28 17 20 13 17 145 7 11 1 GSMtx All 96 7 709 97 161 33 79 36 112 1330 1 79 0 8 14 15 1 12 0 10 2 3 0 2 1 19 3 3 0 1 3 3 0 651 14 11 0 1 0 2 4 4 0 5 14 14 3 3 8 4 5 0 3 4 10 13 2 4 27 0 6 7 4 37 44 48 24 20 1 95 7 36 35 GSMrx All 2825 8 48 115 141 28 76 108 257 3606 1 1389 1 0 0 0 1 4 0 0 2 439 1 0 0 1 0 0 1 1 3 911 0 0 0 0 0 0 0 0 4 0 0 0 0 2 1 0 0 0 5 17 2 32 29 96 7 18 107 0 6 69 4 16 86 13 19 34 0 256 7 29 3 DCStx All 4 4 36 59 98 18 35 12 89 355 1 3 0 2 7 2 1 0 0 0 2 0 0 0 0 8 3 1 1 0 3 0 0 9 5 10 0 0 0 0 4 0 0 0 0 5 1 1 0 0 5 0 2 3 5 36 2 1 11 0 6 1 2 22 41 18 11 20 0 89 7 19 12 DCSrx All 1606 9 56 79 282 41 58 116 315 2562 1 599 1 0 0 0 1 2 0 1 2 395 1 2 0 1 0 0 1 0 3 539 0 0 0 0 0 0 0 0 4 1 0 1 0 0 4 0
118. threshold and this lack of sensitivity seems likely to limit its ability to construct an exposure gradient within a population study Nevertheless the PEM does seem able to discriminate the relatively high exposures of people who live near to mobile phone base station and television broadcast transmitters from those of people living elsewhere 7 2 Functional aspects 7 2 1 Software The PC software had a number of bugs when the PEMs were first supplied to the study and these were communicated to the manufacturer leading to revised versions of the software All major issues encountered now seem to have been resolved 7 2 2 Battery charging There seems to be a problem with the reliability of the battery charging arrangements for the PEM in that the chargers sometimes indicate charging is complete when it is not It was noted that although the PEMs use Nickel Metal Hydride batteries the supplied chargers are labelled as suitable only for Nickel Cadmium types The charging reliability problems should be resolved before the PEM is used in further studies If studies take place before they are resolved the battery voltage should be examined each time after charging to ensure the PEM really is fully charged 7 2 3 Storage integrity The storage of data in the PEM is effectively volatile in that the data are irretrievably lost if the battery becomes discharged for example due to the data not being downloaded promptly after a logging run has be
119. tric field strength V m 100 ji Volunteer 7 2c SPTE TUS ELE eSI AII Locations O EE Living Room Kitchen o Ml a a aa Bedroom B a A ben O DO IO E Office Car EE o il R Outdoor Train oo RES A 7 7 og SS a a a ce 10 i N 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 9 90 TV3 cos ES EE ec de 70 All Locations ME co uu c RUE CU AA AA Living Room o PER E NO E e EN Kitchen Bedroom cou ee E et Office Music Room un Buc cc SS SS SS El Outdoor DE a m ENDE Sele Sla epa E S SS Se SSS set i oer A AE m E 0 4 100 I Volunteer 1 90 TV3 ov HESPERIA iru c c l NES 70 All Locations MEA EIE dE Living Room o EE Dining Room Bedroom Ba a O Office Car a a es ee eee a Outdoor 3 7 o e ad lo es ce en eee 10 L nn a na 100 90 Volunteer 3 N Z EN ccc ccc 2 Am x g ONES S sS PERSONAL EXPOSURE RECORDS TV3 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m TV3 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 2
120. ures recorded varied quite widely outside these limits Since the maximum temperatures could reflect elevated temperatures following battery charging these were ignored however 2 out of 7 instruments that completed the 7 day duration test had both mid run record 2500 and end of run recorded temperatures that were up to 5 C lower than the laboratory temperature 3 9 Low temperature tolerance Following the volunteer trials it became apparent that there were intermittent problems with data loss or instrument malfunction Inspection of the records suggested that there might be a problem when the instrument temperature dropped below 10 C Two instruments were used to investigate this possibility They were set to a 30 s recording rate fully charged then left in the laboratory for 2 h to reach thermal equilibrium after charging They were then started and placed in a laboratory refrigerator at 4 C overnight for 19 h They were taken out and left for 2 h in the laboratory at 21 C before downloading was attempted The LED was still indicating correct operation when the instruments were removed from the refrigerator and before downloading Even so both had corrupted data it being clear from the battery temperature trace that once the instrument temperature approached 10 C malfunctioning occurred The recorded duration was incorrect and the battery voltage and temperature graphs showed that zero values were often logged Inspection of the data
121. ving Room Kitchen SS SS Bedroom B TT O H TTT E Office Car EE o il Outdoor Train en 7 7 7 Dog cate ERS a a um 10 i N 0 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 Electric field strength V m 100 T Volunteer 9 Ei UMTStx cos Er c eL c m 70 All Locations MI oo Ses eT Sie one por Fe N Living Room Se E E A REE E ae Em Kitchen Bedroom cou e T E D Office Music Room un Bu SS SS SS el Outdoor a i E Dl E S le oo Re E A A A A eee EE E Da pp a 0 00 0 05 0 10 0 15 0 20 0 25 0 30 Electric field strength V m 100 I Volunteer 2 Nl oie Wl UMTSk cw S eeu 3 All Locations a iia e Study 2 A A le Kitchen 2 Bedroom 3 s pq SS A AO AA Office Car Ge aH 2 Outdoor E 30 i E EN lt nel a d nlar l n dl dl n e BO O O O O a ha S i gt _ 100 T Volunteer 4 5 uox poa G UMTStx 80 2 Samples above threshold a CONES 5 oS S Z Z All Locations Living Room Al 3 a s l aa S a r na la a N l a ma a en l co Ma h on I 0 Am 0 00 0 05 0 10 0 15 0 20 0 25 0 30 0 35 0 40 100 T Volunteer 6 NN A a ee ANN UMTSk HL ce A ccc 2 3 T E E cw es ein no ag ee N Leane 2 Living Room B ewL 2 osw r sr l ar rar d c E E bar ar arla ar r l aar rl
122. y volunteer who wished to leave it If member of the public were not expected to keep a diary again it would seem possible to use the PEM for longer with 1 2 weeks seeming acceptable to the volunteers Table 19 Number of volunteers indicating particular periods of time that they regarded as the maximum for the PEM to be used in future trials Maximum period regarded as User of the PEM acceptable Volunteer in this trial Member of the public With diary Without diary With diary Without diary Not at all 1 3 days 1 j 1 week 7 2 8 4 2 weeks 1 2 1 month 4 2 3 months 1 1 This volunteer could not find a comfortable way to wear the PEM but indicated they would be prepared to use it for longer if a way could be found 6 7 Other comments At the end of the questionnaire the volunteers were asked if they had any final comments and a range of views were expressed Keeping a diary was regarded as tedious and particularly so because it was not easy to sit down to write when wearing the PEM The diary was given to the volunteers in the form of printed A4 sheets on a clipboard and it was suggested that it would have been more convenient as a smaller spiral bound book that could fit in a pocket One volunteer observed that having to complete the diary was a useful prompt to remember where they had put the PEM when they were not wearing it and that they might have tended to lose the PEM more if they had not had t

Download Pdf Manuals

Related Search

Related Contents

Kenmore Elite 6.1 cu. ft. Electric Range w/ Dual True Convection - Black Owner's Manual  Z 23150 - Leroy Merlin  fiche technique JURA IMPRESSA F8  Convert JPG to PDF online - convert-jpg-to-pdf.net  sensION+ PH3 HACH_LANGE 0710.indd  バルーフプロセッサ  Omega Vehicle Security HHM290 User's Manual  Instructions  FM/AM PLL Synthesized Clock Radio ICF-C273  2600SEG210 NORMA PARA AIRE RESPIRABLE 1.0 PROPÓSITO  

Copyright © All rights reserved. Failed to retrieve file