Energy-Efficient Platform Designed for SDMA Applications in Mobile
Contents
1. Modify SurNo The photo of DataTruck is depicted in Figure 7 and the comparison between DataTruck and other mobile nodes such as Robomote 7 and XYZ 8 is listed in Table 2 Il A SDMA Space Division Multiple Access SDMA is a channel access method based on creating parallel spatial pipes next to higher capacity pipes through spatial multiplexing and or diversity by which it is able to offer superior performance in radio multiple access communication systems DESIGN OF MULTI ANTENNAS ARRAY Table 2 Comparison between DataTruck and other mobile nodes ma ARM9 AVR Atmel ARM7 OKI S3C2440A 8535 ML67Q5002 Communicanon CC2431 UART CC2420 module Maximum speed cm s 130 180 15 20 Storage K a eo oa kbps External A D ae Maximum running time hr XYZ is a node move along a string so it cannot be compared in these aspects Fig 7 A node of DataTruck The kernel part of SDMA is smart antenna Smart antennas also known as adaptive array antennas multiple antennas and recently MIMO are antenna arrays with smart signal processing algorithms used to identify spatial signal signature such as the direction of arrival DOA of the signal and use it to calculate beamforming vectors to track and locate the antenna beam on the mobile target In SDMA system the beams are like multiple space division channels It provides a new domain named space domain expect other three domains time domain freque2. T and sensor 2 wants to send data x The received vector at the DataTruck can be written as 2092 y hix haza n 2 where 7 is channel noise We can see from Eq 2 that each data stream faces an extra source of interference from the other data stream An idea that can be used to remove this inter stream interference from an interested sensor is to project the received signal y onto the subspace orthogonal to the one spanned by the other channel vector That is we choose and es as the filter vectors for sensor 1 and sensor 2 respectively which satisfies ejh2 0 and e5h 0 Hence the received signal can be decoded as 3 te ejy ejhiz ejn LQ 5y ejhore e5n After processed this way the inter stream interference nulling can be achieved can be any vector that lies in V4 which is the space orthogonal to ha however to maximize the received signal strength e should lie in the same direction as the projection of h onto Vy e 2 should be similarly chosen and es can be unit vectors because increasing the length of them will not increase the SNR ej and gt can be expressed as follows e1 Jah hy h22 h T 4 eo hia hiy kial h From Eq 3 we can see that the signal part of x and ren are e h z and ed5he2z2 respectively Since le hy lt lhl and e hall lt lholl we can further see that the projection operation
3. Xing Tian Wang Weijia Jia and Minming Li Rendezvous design algorithms for wireless sensor networks with a mobile base station In MobiHoc 2008 7 K Dantu M Rahimi H Shah S Babel A Dhariwal and G S Sukhatme Robomote enabling mobility in sensor networks In IPSN 2005 8 D Lymberopoulos and A Savvides Xyz a motion enabled power aware sensor node platform for distributed sensor network applications In IPSN 2005 9 S R Gandham M Dawande R Prakash and S Venkatesan Energy efficient schemes for wireless sensor networks with multiple mobile base stations In Globecom 2003 10 Liu wei Li li Zhang jin Smart An tenna Design Ba sed on MUS IC and LMS Algorithms Electronic Sci amp Tech Jan 15 2009 11 Wu Renbiao Kang Xiao Zhong Lunlong Hu Tieqiao Design and Realization of Smart Antenna Based on DSP FPGA Journal of Civil Aviation University of China Vol 27 No 1 February 2009 12 Miao Zhao Ming Ma and Yuanyuan Yang Mobile Data Gathering with Space Division Multiple Access in Wireless Sensor Networks in 27rd Annual IEEE Conference on Computer Communications INFOCOM 2008 13 User s manual of S3C2440A TEXAS INSTRUMENTS Samsung Electronics http www samsung com Products Semiconductor 14 Intel mote iMote http www intel com research exploratory motes htm 2094
4. always reduces the length of h unless h is already orthogonal to the spatial signature of the other data stream This is the overhead for nulling out the interference Hence the effective channel for x would be in deep fading whenever the projection of h onto q is small A similar situation is also applicable to x Therefore for given transmission power of each sensor not any two sensors can successfully transmit data to DataTruck simultaneously To ensure the DataTruck can successfully decode the received signal the follow criteria should be satisfied Pri P e hy gt o SNR P le hi o gt 5 Pro P le3h2 gt dg SN Ro P leSh2 Jo gt dy where Pry Pro SNR and SN Ro are received power and SNF of the received data from the two sensors respectively F is denoted as the transmission power of each sensor and o is the receive sensitivity threshold while 61 is the SN F threshold for the DataTruck to correctly decode the received data Any two sensors that satisfy this criteria can successfully make concurrent data uploading to the DataTruck Such two sensors are said to be compatible 12 Therefore when we select the trajectory of DataTruck the position of compatible nodes should be took into consideration to find the shortest path of DataTruck and collect maximum size of data Sensor Node 1 i b e p XH x gt gt y H D gt Sens
5. in robotics and low power embedded systems mobile nodes 3 4 5 6 7 8 10 are becoming a viable choice for the sensing applications mentioned above These mobile nodes may be mobile data ferries or mobile data relays which responsible to collect or relay data from the sensor nodes to base station In this approach a small number of mobile devices referred to as data ferry roam about sensing fields and collect data from sensors As a result significant network energy saving can be achieved by reducing or completely avoiding costly multi hop wireless transmissions On the other hand if the mobile node has sense ability called mobile sensor it can cover more area than a stationary sensor over a period of time because it can move to anywhere to capture the event Mobile nodes can promote the network performance efficiently We know for a random deployment in static sensor networks the sensor density should increase as O log kloglogl to provide k coverage in a network with size of In 3 authors prove an all mobile sensor network can provide coverage over the field with a constant density of O k independent of network size l Mobile Wireless Sensor Network MWSN can be mainly divided into two categories One is all nodes in the network are mobile However these nodes only have limited mobility to maintain the steady of network 5 The other category is widely researched that part of nodes in the network are mobile and the o
6. set of peripherals multiple power options and a suitable memory configuration The CPU core of S3C2440A is a 16 32 bit ARM920T RISC processor which offers outstanding features The ARM920T implements MMU AMBA BUS and Harvard cache architecture with separate 16KB instruction and 16KB data caches each with an 8 word line length By providing a complete set of common system peripherals the S3C2440A minimizes overall system costs and eliminates the need to configure additional components Operating System and Communication Protocol Stack To make DataTruck interoperable with other devices we ported Chipcons IEEE 802 15 4 compliant medium access control protocol which we operate inside the ARM Linux operating system DataTruck Support API In order to support the special Communication Subsystem TAN y CC2431 ee as gt 802 15 4 Power Subsystem fee y Tracking Coret SV AEA A EA AEE Interface H t INTI ON OFF 3 3V 25V E E d Mobility Subsystem Voltage Samsung mad H bridge H Geared Motor Regulator S3C2440 eee eee ARM9 4 Light Sensor i onori 3 d Thermometer Sensor i K rC W Subsystem H Range Sensors i pane oy Supervisor ee aad Wake Up SRAM 256K X 16 GPIO amp Peripheral Connectors Fig 2 The hardware design graph of ma
7. this hybrid network static nodes send data to rendezvous points at first and mobile node only needs to visit the rendezvous points to save the data delay transmission for the moving trajectory of mobile node is shortened Although the above schemes can perform data gathering in MWSN well there still exist some disadvantages For example when the DataTruck in the intersection area of communication range of several nodes it should gather data one by one which increases the transmission delay In this paper we improve the performance of data gathering in WSNs by considering two critical factors mobility and space division multiple access SDMA technique To the best of our knowledge this is the first work that introduces SDMA technique to mobile sink node design and explores the utility of a joint design of mobility and SDMA technique in data gathering schemes The rest of this paper is organized as follows Section II introduces hardware design of DataTruck and the compare of other mobile nodes In section III we design a dual antennas system on DataTruck using smart antenna technology based on SDMA Section IV presents experiment and simulation results and section V concludes the paper II HARDWARE DESIGN A The core circuit board The architecture of DataTruck is depicted in Figure 2 The main processing unit is a S3C2440A ARM920T RISC microcontroller 13 We found this processor to be an appealing choice since it provides a rich
8. A A FPGA B DDC Hp AD6645 De Kk S g z 2 S 2S 5 g 2 X gn gt 2 188 X DPRAM DBE gt Ly E D DDC H AD6645 5 t A y y y Get weight of Get desired signal throug System synchronization Cholesky_LS respreading control Fig 8 The architecture of smart antennas system C Selection of compatible source nodes Due to the smart antenna cannot null the interference which in the same direction of desired signal not any couple of sensor nodes can send data to DataTruck In order to revert to the original data the source couple of nodes should satisfy certain conditions which we call this couple of nodes is compatible 12 In this section we use Linear Decorrelator Strategy LDS to analyze this problem To guarantee that the decorrelator operation is successful we need to limit the number of simultaneous data streams to no more than the number of receive antennas In other words since the DataTruck is equipped with two receiving antennas at most two sensors can send data simultaneously to the DataTruck Figure 9 shows the transceiver architecture of SDMA with the linear decorrelator For simplicity we will use h to denote h hj2 which represents the complex channel coefficient vector or called spatial signature between sensor i and the two receive antennas of the DataTruck h and h are the two columns of the channel coefficient matrix H Suppose sensor 1 wants to send data
9. IEEE WCNC 2011 Service and Application Energy Efficient Platform Designed for SDMA Applications in Mobile Wireless Sensor Networks Xiwei Zhang Guihai Chen Department of Computer Science Nanjing University Department of Computer and Information Hohai University Nanjing Jiangsu China zxw bhhu edu cn gchen pnju edu cn Abstract Although advance network planning and dense node deployment wireless sensor networks WSNs may achieve the required performance it still face the fundamental challenge of meeting stringent power and time requirements using nodes with limited sensing capacities To better cope with the power consumption problem mobile sensor nodes can be introduced to dynamically reconfigure the sensor network capacity in an on demand manner Through data gathering and relaying mobile nodes can reduce the amount of data transmitting between the static nodes then conserve the power of these nodes to prolong the lifetime of network In this paper we describe the DataTruck a new open source sensing platform specifically designed to support our experimental research in mobile sensor networks which is used to collect or relay data from static sensors The DataTruck node is designed around the S3C2440A ARM920T RISC microprocessor and the IEEE 802 15 4 compliant CC2431 radio from Chipcon Mobility is enabled with an additional accessory board that allows the node to drive its 4 linear motion actuators To red
10. ONCLUSIONS In this paper we introduced the design and implementation of DataTruck a mobile sink node for data gathering which has high performance We have shown the software and hardware design of DataTruck and for the purpose of gathering data efficiently we design a smart antenna system on DataTruck Through experiments we can see that the mobile sink can save the energy of the network and with the dual antennas it reduces the average data transmission delay apparently ACKNOWLEDGMENTS The work is partly supported by China NSF grants 60721002 60825205 61073152 the Fundamental Research Funds for the Central Universities and Hohai Science Fund grant 2009424211 REFERENCES 1 G Tolle J Polastre R Szewczyk D Culler et al Amacroscope in the redwoods In ACM SenSys pages51 63 2005 2 K Mayer K Ellis K Taylor Cattle health mon itoring using wireless sensor networks In IASTEDCCN 2004 3 W W V Srinivasan and K C Chua Trade offs between mobility and density for coverage in wireless sensor networks In MobiCom 2007 4 G Wang G Cao and T L Porta Movement assisted sensor deployment in 23rd Annual IEEE Conference on Computer Communications INFOCOM pp 2469 2479 2004 5 S Chellappan W Gu X Bai D Xuan B Ma and K Zhang Deploying wireless sensor networks under limited mobility constraints IEEE Transactions on Mobile Computing vol 6 no 10 2007 6 Guoliang
11. a Average data delay s w D O O N O Ob 0 10 20 30 40 50 60 70 80 Number of sensing in 5 minutes b sensing frequency VS average transmission delay Fig 10 The relationship between sensing frequency and lifetime and average transmission delay From Figure 10 a we can see while the frequency of data sensing increases the lifetime of network is reduced rapidly with no mobile sink because there are hot spot problem in this network the nodes near base station will die quickly for 2093 relaying all data transmitted from other nodes In Figure 10 b the average transmission delay is not changed when using DataTruck for data gathering this is because the DataTruck picks up data from all nodes when it moving along the trajectory regardless of the frequency of data sensing We use simulations to verify the performance of DataTruck with dual antennas Suppose a sensor network with 40 static nodes evenly distributed in 100x100 meter square field There are 30 data rendezvous points on the trajectory of DataTruck which means in these rendezvous points there is at least one pair of compatible nodes can send data to DataTruck We suppose the communication range of nodes is 30 meter and the data amount of each node send to DataTruck is 1M bytes We analyze the performance of DataTruck through adjusting the moving speed of it and the transmitting rate of static sen
12. e bone as at Joon oR Te schematic of wireless RF communication circuit used in our system is depicted in Figure 5 Aci PE h T ijji CH Fig 5 Schematic of CC2431 in DataTruck 2091 The means of the notations in the table are as follows the message package is start and end with it SurNo address of source node DesNo address of destination node Length package size datan nth sensing data want to transmit to next node CRC cyclic redundancy check code While DataTruck enters the sensing area it will send linking requests to static nodes and set CC2431 in receiving mode Address resolution will be done if DataTruck received sensing data correctly otherwise it send a retransmission signal When the DesNo in the received package is matched with the current DataTruck address the package will be handled in the local node Otherwise the DataTruck modifies the SurNo of the package as the current DataTruck address and relay it to the next node or base station The data transmission workflow is described in Fig 6 Send linking requests N ae Y Address resolution Y Set CC 2431 to receiveing mode Handling data Modify SurNo CRC check Set CC2431 to i sending mode y Retransmission Relaying data Fig 6 The data transmission workflow Y Data relaying Y i
13. gnal data is despreaded in DEMODULE module and then is transmitted to DSP to respread The data after respreading is used to be the desired signal for weight solving of Cholesky LS module Another goal of these data is send to ERRORCODE_ RATIO module for solving the error data rate which is the parameter of the state of system The sensing data from AD6645 were sent to DDC for wave tracking which means the signal frequency will be nulled from 40MHz after AD sampling The signals will be orthogonalized and sent to X DPRAM module and DBF module X DPRAM module translates these 4 line concurrent data to 32 bit serial data and sends these data to DSP for weight solving using EDMA mode through EMIFB interface The functions of DSP mainly include scheduling and computing For instance in DSP there are some modules which are used to control the synchronization state of system and regenerate the desired signal after respreading etc The system needs initialization and the Reset PGA function is called to set the parameters of FPGA After initialization DSP reads synchronization point and respreading data from FPGA A to determine whether the system is enter the synchronization state Then the DSP recalculate the weight and update the weight value in the FPGA B and compute the error code rate at the same time Y FPG
14. in processing unit features of DataTruck we implemented a specialized protocol that consists of the following modules Power Manager Module To support long term deep sleep we have implemented a supervisor circuit outside the processor operated by a real time clock RTC with two interrupts With this circuit the node has software control to transition into a deep sleep mode by disabling its main power supply regulator The RTC is directly powered by the batteries and not by the on board voltage regulator This allows the RTC to keep track of time with its own oscillator when the voltage regulator powering the rest of the board is disabled Device Drivers Module This component includes the low level interfaces to all the devices and the I O peripherals on the DataTruck including the radio the Real Time Clock the ADC the UART the timers and the DMA controller Using this module applications can easily use all the devices connected to the various I O interfaces of the node Radio Manager Module This module is responsible for configuring the communication subsystem Applications can call this component to change the transmission power level of the radio and or its security configuration Frequency Manager Module The frequency manager is responsible for changing the operating clock frequency of the node while preserving the correct timing of the operating system the Zigbee MAC layer and the application running on the node B The motor dr
15. ip of wireless communication circuit The CC2431 is a true System on Chip SoC solution specifically tailored for IEEE 802 15 4 and ZigBee applications It enables ZigBee nodes to be built with very low total bill of material costs The CC2431 combines the excellent performance of the leading CC2420 RF transceiver with an industry standard enhanced 8051 MCU 128 KB flash memory 8 KB RAM and many other powerful features The CC2431 is highly suited for systems where ultra low power consumption is required This is ensured by various operating modes Short transition times between operating modes further ensure low power consumption This is ensured by various operating modes Short transition times between operating modes further ensure low power consumption It need less than 0 6uA current consumption in standby mode where external interrupts can wake up the system Fig 4 Appearance of CC2431 The picture of CC2431 is shown in Figure 4 and the D Software design In order to identify each node in mobile sensor network which contains more than one mobile node and several static sensor nodes each node must have unique address and specific transmitting format The address of mobile node is composed of a CC2431 code which can be a fixed value and a node code We assume the node code is an increasing integer value like 0 1 2 3 and etc The transmitting format is listed in Table 1 Table 1 Transmitting format of nodes se
16. ive circuit DataTruck uses L298P as motor circuit to drive its 4 linear motion actuators The L298 is an integrated monolithic circuit in a 15 lead Multiwatt and PowerSO20 packages It is a high voltage high current dual full bridge driver designed to accept standard TTL logic levels and drive inductive loads such as relays solenoids DC and stepping motors Two enable inputs are provided to enable or disable the device independently of the input signals The emitters of the lower transistors of each bridge are connected together and the corresponding external terminal can be used for the connection of an external sensing resistor An additional supply input is provided so that the logic works at a lower voltage The motor drive circuit is depicted in Fig 3 The speed of motor is calculated using formula 1 which is adjusted by controlling the PWM duty cycle through programming Suppose V nar 18 Maximum rotation speed of r Fi motor duty cycle is D the average speed of motor is Va Va Vines x D 1 In this formula D t T is a duty cycle T is a value of cycle of a timer which is preset by the processor and is the time of high level in the cycle which is set in the program to satisfy t lt T 2090 paie IN4 148 IN4 148 AME gt D D GND D p Fig 3 The schematic of motor drive circuit C Wireless RF communication circuit DataTruck uses CC2431 CHIPCON CC2431 2007 as the control ch
17. ncy domain and code domain Although when the user is utilizing the same frequency and address code in the same time they can use the space division channels to promote the capacity of communication system B Design of dual antennas system In this paper we are mainly consider the case when the DataTruck is equipped with two antennas because it is not hard to mount two antennas on the DataTruck while it will likely become difficult and even infeasible to mount more antennas due to the constraint on the distances between antennas to ensure independent fading Now we will present the design of a smart antenna system with dual antennas Smart antennas have two main functions DOA estimation and Beamforming In this paper smart antenna achieves DOA and identifies the directions of the received signals around the array antenna using the MUSIC Multiple Signal Classification algorithm 10 We control the direction of the main beam by the LS DRCMA least squares de spread re spread constant module algorithm algorithm 11 and thus track the desired source signal and at the same time generate deep nulls in the direction of interfering signals Here we use dual FPGA and DSP chip to design the smart antenna system which is shown in Figure 8 In Figure 8 the CORRELATE module in FPGA B uses the signal which is transmitted from FPGA A after beamforming to synchronize all signal data The synchronization point is delivered to DSP and the si
18. or Node 2 gt amp m Xo DataTruck Fig 9 Linear Decorrelator Strategy LDS in SDMA IV EXPERIMENTS AND SIMULATIONS To compare the proposed solution we have made two kinds of experiments to test the performance of DataTruck In the real experiments we use DataTruck with single antenna to gather data from static sensor nodes There are 20 static nodes in the room and DataTruck knows the position of each node it visits all nodes along a fixed trajectory and each node is visited once in every round The DataTruck is equipped with an infrared ranging module which is used to avoid the barrier When DataTruck find the barrier it will turn left or right with an angle of 30 degree then it will go to the position along prior direction using the electronic compass module We compare the lifetime of network and the average data transmission delay between using DataTruck and multihop transmitting directly with different sensing frequency of static nodes The results are shown in Figure 10 5004 l x using DataTruck 450 no mobile sink 400 350 5 5 300 x O iN 250 f NG 5 ra gt 200 amp 150 100 50 W 0 10 20 30 40 50 60 70 80 Number of sensing in 5 minutes a sensing frequency VS lifetime of network using DataTruck no mobile sink _ N Q 2
19. sor nodes which is showed in Figure 11 The performance is the average of the results in 1000 simulations 2500 single antenna dual antennas 2000 1500 1000 1 het Average Total Time s ek w 0 20 40 60 80 100 120 140 150 Number of Sensor Nodes a moving speed is 0 5m s transmitting rate is 90Kbps 3000 single antenna dual antennas 2500 2000 E i T xr _ a 1500 at at or fod pe aK a D oO Ke ig aK D E 4 1000 x ae D a ES a 500 2 AT 0 0 20 40 60 80 100 120 140 150 Number of Sensor Nodes b moving speed is 1m s transmitting rate is SOKbps Fig 11 The relationship between number of nodes and the average total time of a data gathering From Figure 11 we can see that using dual antennas systems outperform non SDMA algorithm and the improvement turns to be more evident when the network becomes denser with more sensors This is reasonable because more sensors make data uploading time dominant and provide more opportunities to utilize SDMA for concurrent data uploading Thus DataTruck with dual antennas is suitable for data gathering when the density of sensor nodes is high The figure shows if there are 100 sensor nodes in the field the data delay is reduced 40 when the DataTruck using dual antennas than single antenna V C
20. ther nodes are static so the network 1s called hybrid network In this network static nodes are responsible for data sensing and the mobile nodes move to them to collect or relay the data to base station A classical application of a hybrid network is depicted in Figure 1 In this paper we design a mobile sink node named DataTruck Mobile sink usually has no sensing ability so the main function of DataTruck is collect or relay data from other static sensors DataTruck is a new sensor node platform designed to support mobility experiments in sensor networks Although our design is driven by the research requirements of our group extra effort was taken during the design phase to specify a feature set that is complimentary to existing 2089 platforms and can serve multiple aspects of research and education in sensor networks The DataTruck platform is built around a S3C2440A ARM920T RISC microprocessor from Samsung Semiconductor and a CC2431 radio with a 250kbps raw data rate from Chipcon The choice of the S3C2440A microcontroller provides a wealth of peripherals and flexible modes of operation The Chipcon radio and its use with an IEEE 802 15 4 compliant MAC protocol make our node interoperable with other sensor nodes available in the community such as Telos and Micaz O O ki E basestation rendezvous A point static node gt wireless link __ moving trajectory Fig 1 In
21. uce power consumption a long term sleep mode is supported through different power supplying methods for main board and clock Furthermore we integrated a smart antenna system to gather the data from multiple static nodes concurrently which transmitting data using the same frequency of channel The experiments show that DataTruck collects data efficiently to reduce the average data transmission delay by using SDMA technology Keywords wireless sensor networks mobile sink SDMA smart antenna I INTRODUCTION In recent year wireless sensor networks WSNs have been used to monitor physical or environmental conditions collect or transmit sensing data These networks can serve as an infrastructure for a number of applications including surveillance medical monitoring agricultural cultivation facility monitoring and entertainments 1 2 For a number of these applications sensor nodes could be deployed in a vast area or in harsh environments As a result a large network deployment may require excessive sensor nodes in order to achieve satisfactory sensing performance Moreover although dense node deployment may initially achieve the required performance it does not adapt to dynamic changes of network 978 1 61284 254 7 11 26 00 2011 IEEE conditions or physical environments For instance death of nodes due to battery depletion or physical attacks can easily cause coverage holes in a monitored battlefield With recent advances
Download Pdf Manuals
Related Search