Several Bits Are Enough: Off-Grid Target Localization in WSNs Using Variational Bayesian EM Algorithm

Yan GUO  Peng QIAN  Ning LI  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E102-A   No.7   pp.926-929
Publication Date: 2019/07/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E102.A.926
Type of Manuscript: LETTER
Category: Digital Signal Processing
wireless sensor networks,  target localization,  quantized data,  compressive sensing,  variational Bayesian method,  

Full Text: PDF(314.3KB)>>
Buy this Article

The compressive sensing has been applied to develop an effective framework for simultaneously localizing multiple targets in wireless sensor networks. Nevertheless, existing methods implicitly use analog measurements, which have infinite bit precision. In this letter, we focus on off-grid target localization using quantized measurements with only several bits. To address this, we propose a novel localization framework for jointly estimating target locations and dealing with quantization errors, based on the novel application of the variational Bayesian Expectation-Maximization methodology. Simulation results highlight its superior performance.