5th International ICST Conference on Communications and Networking in China

Research Article

Laplace prior based distributed compressive sensing

Download426 downloads
  • @INPROCEEDINGS{10.4108/chinacom.2010.82,
        author={Liang Tang and Zheng Zhou and Lei Shi and Haipeng Yao and Jing Zhang and Yabin Ye},
        title={Laplace prior based distributed compressive sensing},
        proceedings={5th International ICST Conference on Communications and Networking in China},
        publisher={IEEE},
        proceedings_a={CHINACOM},
        year={2011},
        month={1},
        keywords={Bayesian compressive sensing (BCS) Laplace prior statistically interrelationship distributed Bayesian compressive sensing},
        doi={10.4108/chinacom.2010.82}
    }
    
  • Liang Tang
    Zheng Zhou
    Lei Shi
    Haipeng Yao
    Jing Zhang
    Yabin Ye
    Year: 2011
    Laplace prior based distributed compressive sensing
    CHINACOM
    ICST
    DOI: 10.4108/chinacom.2010.82
Liang Tang1,*, Zheng Zhou1,*, Lei Shi1, Haipeng Yao1, Jing Zhang1, Yabin Ye2
  • 1: Key Lab of Universal Wireless Communications, MOE, Wireless Network Lab, Beijing University of Posts and Telecommunications, Beijing, China
  • 2: European Research Center, Huawei Technologies Duesseldorf GmbH
*Contact email: tangliangbupt@gmail.com, zzhou@bupt.edu.cn

Abstract

Bayesian compressive sensing (BCS) utilizes the prior distribution of signal coefficients to reconstruct the original signal. The widely used prior is Laplace and Gaussian distributed. In this paper, we use the scene of L sets of signal sparse coefficients which are statistically related and take advantage of Laplace prior and statistically interrelationship among signals to propose the Laplace prior based distributed Bayesian compressive sensing. We provide the experiment result to demonstrating that the proposed method is an effective reconstruction algorithm and has a good performance.