逆向捕获时间差的Voronoi声源定位机制

doi:10.3969/j.issn.1000-436x.2013.11.016

摘要/Abstract

摘要：

提出一种逆向捕获时间差的声源定位协议，可以使传感器网络在大部分时间里处于射频休眠状态，因此具有显著的节能效果；在利用时间差数据求解声源位置时，引入Voronoi图理论对搜索空间进行裁剪，以提高算法搜索求解的效率和成功率。理论分析和实验结果表明该声源定位机制具有明显的能量有效性、定位解算的精确性、快速收敛性和顽健性，适用于能量受限无线传感器网络中动态声源的实时精确定位。

关键词: 无线传感器网络, 声源定位, TDOA, 逆向捕获, Voronoi图

Abstract:

An acoustic source localization protocol based on counter captured time difference was presented. It can drive the sensors in network to “sleep”in most of the time, so as to achieve energy saving obviously. In the stage of computing the acoustic source position with the time difference data, Voronoi diagram was introduced to reduce the searching space to improve the success percentage and convergence speed of the algorithm. Both the theoretical analysis and experiment results demonstrate that this acoustic localization mechanism is energy efficient, in real-time and robust in localization computing. So it is suitable for dynamic acoustic source accurate localization in WSNs.

Key words: wireless sensor networks, acoustic source localization, TDOA, counter captured, Voronoi diagram

夏娜,倪成春,徐朝农,丁胜,郑榕. 逆向捕获时间差的Voronoi声源定位机制[J]. 通信学报, 2013, 34(11): 140-152.

Na XIA,Cheng-chun NI,Chao-nong XU,Sheng DING,Rong ZHENG. Voronoi acoustic source localization mechanism based on counter captured time difference[J]. Journal on Communications, 2013, 34(11): 140-152.

图/表 19

图1

图2

图3

表1

图4

图5

图6

图7

图8

图9

图10

表2

图11

表3

图12

图13

图14

表4

表5

参考文献 29

[1]	SAYEESH K V , KUMAR K V , PHANI G L . Fault repair algori hm using localization and controlled mobility in WSN[A]. of 2010 IEEE Global Telecommunication Conference (GLOBECOM 2010)[C]. Miami, Florida, USA, 2010.1663-1667.
[2]	LIU Y , HU Y H , PAN Q . Decentralized robust acoustic source locali-zation with wireless sensor networks for heavy-tail distributed obser-vations[A]. Proc of 2010 IEEE Global Telecommunication Conference (GLOBECOM 2010)[C]. Miami, Florida, USA, 2010.1-5.
[3]	MENG W , XIAO W D , XIE L H . Diffusion based projection thod for distributed source localization in wireless sensor networks[A]. Proc of 2011 IEEE International Conference on Computer Communications (INFOCOM 2011)[C]. Shanghai, China, 2011.537-542.
[4]	WANG Q X , ZHENG R , TIRUMALA A , et al. Lightning: a hard real-time, fast, and lightweight low-end wireless sensor election pro-tocol for acoustic event localization[J]. IEEE Transactions on Mobile Computing, 2008,7 (5): 570-584.
[5]	夏娜，唐树青，赵娟 . WSNs中节点协同声源定位协议研究[J]. 合肥工业大学学报 (自然科学版), 2011,34 (10): 1492-1495,1600. XIA N , TANG S Q , ZHAO J . Research on acoustic source localization protocol based on sensors collaboration in WSNs[J]. Journal of Hefei University of Technology (Natural Science), 2011,34 (10): 1492-1495,1600.
[6]	夏娜，郑语晨，杜华争等. 刚性驱动水下传感器节点自组织布置[J]. 计算机学报, 2013,36 (3): 494-505. XIA N , ZHENG Y C , DU H Z , et al. Rigidity driven underwater sen-sor self-organized deployment[J]. Chinese Journal of Computers, 2013,36 (3): 494-505.
[7]	DUHZ , XIAN , JIANGJG , et al. AmontecarloenhancedPSOalgo-rithm for optimal QoM in multi-channel wireless networks[J]. Journal of Computer Science and Technology, 2013,28 (3): 553-563.
[8]	XIA N , XU L N , NI C C . Optimal QoM in multichannel wireless networks based on MQICA[J]. International Journal of D stributed Sensor Networks, 2013,2013: 1-14.
[9]	KIM W . Support vector learning approaches for object localization in acoustic wireless sensor networks[A]. Proc of 2010 IEEE International Conference on Intelligent Systems[C]. London, UK, 2010.485-489.
[10]	KIM W . Event-driven Gaussian process for object localization in wireless sensor networks[A]. Proc of 2011 IEEE International Confe-rence on Intelligent Robots and Systems[C]. San Francisco, California,USA, 2011.2790-2795.
[11]	VAGHEFI R M , GHOLAMI M R . RSS-based sensor localization with unknown transmit power[A]. Proc of 2011 International ference on Acoustics, Speech and Signal Processing[C]. Prague, Czech Re-public, 2011.2480-2483.
[12]	YAN Y S , WANG H Y , SHEN X H . Efficient convex optimization method for underwater passive source localization based on RSS with WSN[A]. Proc of 2012 IEEE International Conference on ignal Processing, Communication and Computing(ICSPCC)[C]. Hong Kong, China, 2012.171-174.
[13]	ZENG W J , LI X L . High-resolution multiple wideband and nonsta-tionary source localization with unknown number of sources[J]. IEEE Transactions on Signal Processing, 2010,58 (6): 3125-3136.
[14]	LU L , WU H C . New direction-of-arrival-based source localization algorithm for wideband signals[A]. Proc of 2011 IEEE G l Tele-communications Conference(GLOBECOM 2011)[C]. Houston, exas, USA, 2011.1-5.
[15]	YOON J Y , KIM J W , LEE W Y . A TDOA-based localization using precise time-synchronization[A]. Proc of 2012 International Confe-rence on Advanced Communication Technology[C]. Chengdu, China, 2012.1266-1271.
[16]	QU X M , XIE L H . Source localization by TDOA with random sensor positions errors -part I: static sensors[A]. Proc of 2012 International Conference on Information Fusion[C]. Singapore, 2012.48-53.
[17]	MERHI Z , ELGAMEL M , BAYOUMI M . A lightweight collaborative fault tolerant target localization system for wireless sensor networks[J]. IEEE Transactions on Mobile Computing, 2009,8 (12): 1690-1704.
[18]	LIU Y , HU Y H . Robust maximum likelihood acoustic source localiza-tion in wireless sensor networks[A]. Proc of 2009 IEEE Global Tele-communications Conference (GLOBECOM 2009)[C]. Hawaii, 2009.1-6.
[19]	LU L , WU H C . Robust expectation-maximization algorithm for mul-tiple wideband acoustic source localization in the presence of nonuni-form noise variances[J]. IEEE Sensors Journal, 2011,11 (3): 536-544.
[20]	NAKAMURA K , NAKADAI K , ASANO F , et al. Intelligent sound source localization and its application to multimodal uman track-ing[A]. Proc of 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)[C]. San Francisco, California, USA, 2011.143-148.
[21]	ZHANG C , HUANG Y . Research on acoustic source localization using time difference of arrival measurements[A]. Proc of 2012 International Conference on Measurement, Information and Control[C]. Harbin, China, 2012.220-224.
[22]	KNAPP C H , CARTER G C . The generalized correlation method for estimation of time delay[J]. IEEE Transactions on Acoustic, Speech and Signal Processing, 1976,24 (4): 320-327.
[23]	JIA J J , LIU M J , LI X F . Acoustic localization algorithm using wire-less sensor networks[A]. Proc of 2009 International Conference on In-telligent Computation Technology and Automation[C]. Changsha, China, 2009.434-437.
[24]	VAKULYA G , SIMON G . Fast adaptive acoustic localizatio for sensor networks[J]. IEEE Transactions on Instrumentation and ment, 2011,60 (5): 1820-1829.
[25]	ALBANESE R , RUBINACCI G . Numerical procedures for the lu-tion of nonlinear electromagnetic problems[J]. IEEE Transactions on Magnetics, 1992,28 (2): 1228-1231.
[26]	VORONOI G M . Nouvelles applications des parametres continus a theorie des formes quadratiques[J]. Deuxieme Memoire Recherches Sur Les Parallelloedres Primitives, Journal Für Die Re ne Und Ange-wandte Mathematik, 1908,34: 98-287.
[27]	FU N , IMAI H . Voronoi diagrams on periodic graphs[A]. Proc of 2010 International Symposium on Voronoi Diagram in Science and Engi-neering[C]. Quebec, Canada, 2010.189-198.
[28]	XIA N , DU H Z , LI S J , et al. VSPSA for acoustic source localization in wireless sensor networks[J]. Ad Hoc ＆ Sensor Wireless Networks, 2013,19 (3-4): 277-304.
[29]	MIRZARGAR M , ENTEZARI A . Voronoi splines[J]. IEEE Transac-tions on Signal Processing, 2010,58 (9): 4572-4582.

参数符号	含义
S_i	第i个节点（i = 1, 2, 3，下同）
d_i	声源S到节点i的距离
v	声音在空气中的传播速度
D_reg	节点识别声音信号所需要的时间
T_{rf 1}	节点发送射频信号的启动时间
T_{rf 2}	节点接收并识别射频信号的时间
T_burst	节点发送射频信号持续的时间, T_burst≥T_{rf 2}
T_sat	广播“Satisfied”信号持续的时间，T_sat＞T_burst
R_tr	数据发送速率，单位bit/s
L_c	节点的通信帧的长度(bit)（包含同步字节Sy (bit)，帧头He (bit)，数据域Da (bit)，校验比特Ch (bit)

参数符号	参数值
v	340 m/s
D_reg	150 μs
T_{rf 1}	200 μs
T_{rf 2}	300 μs
T_burst	500 μs
T_sat	1000 μs
R_tr	34.7 kbit/s
L_c	150 bit
P_tx	120 mW
P_rx	68 mW
T	3 600 s

组别	节点数	是否采用Voronoi增强	平均定位误差/m	求解成功率^{（“求解成功”定义为当算法定位结果到声源的距离≤监测区域边长的1% 时，此次定位是成功的。）}
1	9	否	6.824	53%
1	9	是	2.281	72%
2	16	否	5.436	62%
2	16	是	2.019	81%
3	25	否	4.784	68%
3	25	是	1.533	88%
4	36	否	3.408	76%
4	36	是	0.907	92%
注：平均定位误差为200次实验的平均误差值。

实验分组	最小定位误差/m	最大定位误差/m	平均定位误差/m
1	0.172	3.751	2.245
2	0.137	4.613	2.097
3	0.122	4.085	1.916

节点数	发声间隔? /st	网络生存时间/h
节点数	发声间隔? /st	本文声源定位机制	传统TDOA方法
	1	94.3	39.5
8	5	122.5	44.6
	25	187.0	51.4
	1	89.2	17.9
16	5	119.8	21.1
	25	179.4	35.2
注：网络生存时间定义为从网络初始化到第一个节点能量耗尽的时间。