高级搜索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

一种无线传感器网络中目标跟踪的自适应节点调度算法

胡波 王祺尧 冯辉 罗灵兵

胡波, 王祺尧, 冯辉, 罗灵兵. 一种无线传感器网络中目标跟踪的自适应节点调度算法[J]. 电子与信息学报, 2018, 40(9): 2033-2041. doi: 10.11999/JEIT171154
引用本文: 胡波, 王祺尧, 冯辉, 罗灵兵. 一种无线传感器网络中目标跟踪的自适应节点调度算法[J]. 电子与信息学报, 2018, 40(9): 2033-2041. doi: 10.11999/JEIT171154
Bo HU, Qiyao WANG, Hui FENG, Lingbing LUO. Adaptive Sensor Scheduling Algorithm for Target Tracking in Wireless Sensor Networks[J]. Journal of Electronics & Information Technology, 2018, 40(9): 2033-2041. doi: 10.11999/JEIT171154
Citation: Bo HU, Qiyao WANG, Hui FENG, Lingbing LUO. Adaptive Sensor Scheduling Algorithm for Target Tracking in Wireless Sensor Networks[J]. Journal of Electronics & Information Technology, 2018, 40(9): 2033-2041. doi: 10.11999/JEIT171154

一种无线传感器网络中目标跟踪的自适应节点调度算法

doi: 10.11999/JEIT171154
基金项目: 国家自然科学基金(61501124),上海市公安局科学技术发展基金(2017012)
详细信息
    作者简介:

    胡波:男,1968 年生,教授,研究方向为数字信号处理、数字通信和系统设计

    王祺尧:男,1993 年生,硕士生,研究方向为传感器网络、强化学习、序贯决策等研究

    冯辉:男,1980 年生,副教授,研究方向为分布式信号处理理论与应用

    罗灵兵:男,1992 年生,硕士生,研究方向为图像处理

    通讯作者:

    冯辉  hfeng@fudan.edu.cn

  • 中图分类号: TP393; TP391

Adaptive Sensor Scheduling Algorithm for Target Tracking in Wireless Sensor Networks

Funds: The National Natural Science Foundation of China (61501124), The Public Security Bureau Science and Technology Development Foundation of Shanghai (2017012)
  • 摘要: 在无线传感器网络目标跟踪的过程中进行节点调度,可以综合考虑跟踪误差和能量消耗,延长传感器网络的使用寿命。为了综合考虑节点调度的短期和长远损失,该文将问题建模为部分可观测马尔科夫决策过程(POMDP)以得到更优的调度策略,并提出一种近似求解算法C-QMDP。该算法利用马尔科夫链蒙特卡洛方法(MCMC)推导连续状态空间的置信状态的转移,并计算瞬时代价。使用状态离散化方法,基于马尔科夫决策过程(MDP)值迭代求解未来代价的近似值。仿真结果表明,相比现有POMDP近似算法,该文算法既可以降低跟踪过程中的累积损失,又可以将大量运算进行离线计算,减小了在线决策时的计算量。
  • 图  1  贝叶斯网络

    图  2  目标真实轨迹与估计轨迹比较

    图  3  目标跟踪过程中的累积跟踪误差

    图  4  目标跟踪过程中的累积总代价

    图  5  不同时刻节点调度策略比较

    表  1  C-QMDP算法

      算法 1 C-QMDP算法
     输入:置信状态(包括粒子状态和粒子权重):
      ${{{b}}_k} = \left({{S}}_k^{(1)},\tilde w_k^{(1)},{{S}}_k^{(2)},\tilde w_k^{(2)},·\!·\!·,{{S}}_k^{(N)},\tilde w_k^{(N)}\right)$
     输出:最优动作A
     (1) function C-QMDP( ${{{b}}_k}$)
     (2) $V\!\, \raisebox{4pt}{\hat} \leftarrow {\rm{MDP\_discrete\_value\_iteration()}}$
     (3)  for all control actions A do
     (4) ${{{b}}_{k + 1}} \leftarrow {\rm{Particle\_filter(}}{{{b}}_k},{{A}}{\rm{)}}$
     (5) for i = 1 : N do
     (6) ${\tilde{ S}}_{k + 1}^{(i)} \leftarrow {{S}}_{k + 1}^{(i)}$
     (7) end for
     (8) $Q({{{b}}_k},{{A}}) = L({{{b}}_k},{{A}}) + \beta \sum\limits_{i = 1}^N {\tilde w_k^{(i)}V\!\, \raisebox{4pt}{\hat}\left({\tilde{ S}}_{k + 1}^{(i)}\right)} $
     (9) end for
     (10) return $\arg \mathop {\max }\limits_{{A}} Q({{{b}}_k},{{A}})$
     (11) end function
    下载: 导出CSV

    表  2  C-QMDP与其余算法性能比较

    算法 跟踪误差 传感器功耗 总代价
    全部开启 1.3647 70.0722 97.3662
    CPA 2.9132 24.0217 82.2857
    CO-Rollout 1.8371 20.3493 57.0913
    C-QMDP 1.5542 20.0906 51.1746
    下载: 导出CSV
  • ALCALA J M, URENA J U, HERNANDEZ A, et al. Sustainable homecare monitoring system by sensing electricity data[J]. IEEE Sensors Journal, 2017, 17(23): 7741–7749 doi: 10.1109/JSEN.2017.2713645
    MARTELLI T, BONGIOANNI C, COLONE F, et al. Security enhancement in small private airports through active and passive radar sensors[C]. 17th IEEE International Conference on Radar Symposium (IRS), Krakow, Poland, 2016: 1–5.
    SHI W Y and CHIAO J C. Neural network based real-time heart sound monitor using a wireless wearable wrist sensor[C]. IEEE Conference on Circuits and Systems Conference (DCAS), Arlington, USA, 2016: 1–4.
    ANGLEY D, SUVOROVA S, RISTIC B, et al. Sensor scheduling for target tracking in large multistatic sonobuoy fields[C]. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Arlington, USA, 2017: 3146–3150.
    SONG R, WEI Q, and XIAO W. ADP-based optimal sensor scheduling for target tracking in energy harvesting wireless sensor networks[J]. Neural Computing and Applications, 2016, 27(6): 1543–1551 doi: 10.1007/s00521-015-1954-4
    YANG X, ZHANG W A, CHEN M Z Q, et al. Hybrid sequential fusion estimation for asynchronous sensor network-based target tracking[J]. IEEE Transactions on Control Systems Technology, 2017, 25(2): 669–676 doi: 10.1109/TCST.2016.2558632
    唐显锭, 冯辉, 杨涛, 等. 无线传感器网络中用于目标跟踪的节点规划算法[J]. 太赫兹科学与电子信息学报, 2014, 12(3): 355–361 doi: 10.11805/TKYDA201403.0355

    TANG Xianding, FENG Hui, YANG Tao, et al. Sensor scheduling for target tracking in wireless sensor networks[J]. Journal of Terahertz Science and Electronic Information Technology, 2014, 12(3): 355–361 doi: 10.11805/TKYDA201403.0355
    ZHANG H, AYOUB R, and SUNDARAM S. Sensor selection for Kalman filtering of linear dynamical systems: Complexity, limitations and greedy algorithms[J]. Automatica, 2017, 78: 202–210 doi: 10.1016/j.automatica.2016.12.025
    冉晓旻, 方德亮. 基于势博弈的分布式目标跟踪传感器分配算法[J]. 电子与信息学报, 2017, 39(11): 2748–2754 doi: 10.11999/JEIT170229

    RAN Xiaomin and FANG Deliang. Distributed sensor allocation algorithm for target tracking based on potential game[J]. Journal of Electronics&Information Technology, 2017, 39(11): 2748–2754 doi: 10.11999/JEIT170229
    SINGH P, CHEN M, CARLONE L, et al. Supermodular mean squared error minimization for sensor scheduling in optimal Kalman filtering[C]. IEEE Conference on American Control Conference (ACC), Seattle, USA, 2017: 5787–5794.
    ASGHAR A B, JAWAID S T, and SMITH S L. A complete greedy algorithm for infinite-horizon sensor scheduling[J]. Automatica, 2017, 81: 335–341 doi: 10.1016/j.automatica.2017.04.018
    SPAAN M T J. Partially Observable Markov Decision Processes[M]. Berlin Heidelberg: Springer, 2012: 387–414.
    ZOIS D S, LEVORATO M, and MITRA U. Active classification for POMDPs: A Kalman-like state estimator[J]. IEEE Transactions on Signal Processing, 2014, 62(23): 6209–6224 doi: 10.1109/TSP.2014.2362098
    ZOIS D S and MITRA U. Active state tracking with sensing costs: Analysis of two-states and methods for n-states[J]. IEEE Transactions on Signal Processing, 2017, 65(11): 2828–2843 doi: 10.1109/TSP.2017.2664049
    SHANI G, PINEAU J, and KAPLOW R. A survey of point-based POMDP solvers[J]. Autonomous Agents and Multi-Agent Systems, 2013, 27(1): 1–51 doi: 10.1007/s10458-012-9200-2
    LITTMAN M L, CASSANDRA A R, and KAELBLING L P. Learning policies for partially observable environments: Scaling up[C]. Proceedings of the 12th International Conference on Machine Learning, Tahoe City, USA, 1995: 362–370.
    RUSSELL S. Artificial Intelligence: A Modern Approach. Making Complex Decisions (Ch-17)[M]. Englewood Cliffs: Prentice-Hall, 2004: 645–692.
    HE Y and CHONG K P. Sensor scheduling for target tracking in sensor networks[C]. 43rd IEEE Conference on Decision and Control(CDC), Nassau, Bahamas, 2004: 743–748.
    CHONG E K P, KREUCHER C M, and HERO III A O. POMDP Approximation Using Simulation and Heuristics[M]. Boston, MA: Springer, 2008: 95–119.
    LI Y, KRAKOW L W, CHONG E K P, et al. Dynamic sensor management for multisensor multitarget tracking[C]. IEEE 40th Annual Conference on Information Sciences and Systems, Princeton, USA, 2006: 1397–1402.
    LI Y, KRAKOW L W, CHONG E K P, et al. Approximate stochastic dynamic programming for sensor scheduling to track multiple targets[J]. Digital Signal Processing, 2009, 19(6): 978–989 doi: 10.1016/j.dsp.2007.05.004
    BAR-SHALOM Y, LI X R, and KIRUBARAJAN T. Estimation with Applications to Tracking and Navigation: Theory Algorithms and Software[M]. New York: John Wiley & Sons, 2004: 199–266.
    ALIPPI C and VANINI G. A RSSI-based and calibrated centralized localization technique for Wireless Sensor Networks[C]. Fourth Annual IEEE International Conference on Pervasive Computing and Communications Workshops, Pisa, Italy, 2006: 301–305.
    NIU R and VARSHNEY P K. Target location estimation in sensor networks with quantized data[J]. IEEE Transactions on Signal Processing, 2006, 54(12): 4519–4528 doi: 10.1109/TSP.2006.882082
    ARULAMPALAM M S, MASKELL S, GORDON N, et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J]. IEEE Transactions on Signal Processing, 2002, 50(2): 174–188 doi: 10.1109/78.978374
    RUSSELL S. Artificial Intelligence: A Modern Approach. Probabilistic Reasoning Over Time (Ch-15)[M]. Englewood Cliffs: Prentice-Hall, 2004: 566–609.
    HE Y and CHONG E K P. Sensor scheduling for target tracking: A Monte Carlo sampling approach[J]. Digital Signal Processing, 2006, 16(5): 533–545 doi: 10.1016/j.dsp.2005.02.005
  • 加载中
图(5) / 表(2)
计量
  • 文章访问数:  2670
  • HTML全文浏览量:  643
  • PDF下载量:  79
  • 被引次数: 0
出版历程
  • 收稿日期:  2017-12-06
  • 修回日期:  2018-05-04
  • 网络出版日期:  2018-07-12
  • 刊出日期:  2018-09-01

目录

    /

    返回文章
    返回