Distributed parameter estimation in unreliable sensor networks via broadcast gossip algorithms

In this paper, we present an asynchronous algorithm to estimate the unknown parameter under an unreliable network which allows new sensors to join and old sensors to leave, and can tolerate link failures. Each sensor has access to partially informative measurements when it is awakened. In addition, the proposed algorithm can avoid the interference among messages and effectively reduce the accumulated measurement and quantization errors. Based on the theory of stochastic approximation, we prove that our proposed algorithm almost surely converges to the unknown parameter. Finally, we present a numerical example to assess the performance and the communication cost of the algorithm.

[1]  Fabio Fagnani,et al.  Broadcast Gossip Averaging: Interference and Unbiasedness in Large Abelian Cayley Networks , 2011, IEEE Journal of Selected Topics in Signal Processing.

[2]  Ian F. Akyildiz,et al.  Wireless sensor networks: a survey , 2002, Comput. Networks.

[3]  Roberto López-Valcarce,et al.  A Diffusion-Based EM Algorithm for Distributed Estimation in Unreliable Sensor Networks , 2013, IEEE Signal Processing Letters.

[4]  Shaochuan Wu,et al.  Convergence of Gossip Algorithms for Consensus in Wireless Sensor Networks with Intermittent Links and Mobile Nodes , 2014 .

[5]  Soummya Kar,et al.  Gossip Algorithms for Distributed Signal Processing , 2010, Proceedings of the IEEE.

[6]  Tingwen Huang,et al.  Distributed parameter estimation in unreliable WSNs: Quantized communication and asynchronous intermittent observation , 2015, Inf. Sci..

[7]  Tingwen Huang,et al.  Average consensus in sensor networks via broadcast multi-gossip algorithms , 2013, Neurocomputing.

[8]  O. Nelles,et al.  An Introduction to Optimization , 1996, IEEE Antennas and Propagation Magazine.

[9]  Stephen P. Boyd,et al.  Randomized gossip algorithms , 2006, IEEE Transactions on Information Theory.

[10]  Anand D. Sarwate,et al.  Broadcast Gossip Algorithms for Consensus , 2009, IEEE Transactions on Signal Processing.

[11]  Bo Liu,et al.  A new approach to the stability analysis of continuous-time distributed consensus algorithms , 2013, Neural Networks.

[12]  Huijun Gao,et al.  Finite-Horizon $H_{\infty} $ Filtering With Missing Measurements and Quantization Effects , 2013, IEEE Transactions on Automatic Control.

[13]  T. C. Aysal,et al.  Distributed Average Consensus With Dithered Quantization , 2008, IEEE Transactions on Signal Processing.

[14]  Kai Cai,et al.  Quantized Consensus and Averaging on Gossip Digraphs , 2011, IEEE Transactions on Automatic Control.

[15]  Na Wang,et al.  Broadcast gossip algorithm with quantization , 2012, 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery.

[16]  Zhi-Quan Luo,et al.  Decentralized estimation in an inhomogeneous sensing environment , 2005, IEEE Transactions on Information Theory.

[17]  Qiang Zhang,et al.  Distributed Parameter Estimation Over Unreliable Networks With Markovian Switching Topologies , 2012, IEEE Transactions on Automatic Control.

[18]  Sonia Martínez,et al.  Distributed Coverage Games for Energy-Aware Mobile Sensor Networks , 2013, SIAM J. Control. Optim..

[19]  Santiago Zazo,et al.  Distributed static linear Gaussian models using consensus , 2012, Neural Networks.

[20]  Naomi Ehrich Leonard,et al.  Cooperative learning in multi-agent systems from intermittent measurements , 2012, 52nd IEEE Conference on Decision and Control.

[21]  Javad Lavaei,et al.  Quantized Consensus by Means of Gossip Algorithm , 2012, IEEE Transactions on Automatic Control.