Research On Signal Recovery Algorithm For Wireless Sensor Networks Based On Compressed Sensing

With the gradual expansion of the scale of wireless sensor networks(WSNs), energy issue has become one of the key problems need to be solved urgently. As a new sampling theory, compressed sensing(CS) can take advantage of a lower sampling rate for accurate recovery of a signal to achieve energy saving effect. Standard CS theory consists of three parts, namely, the sparse representation of a signal, the design of a measurement matrix and the recovery of original signals. This paper mainly deals with the improvement for the recovery and the measurement part.Commonly used random measurement matrix doesn’t depend on the signal itself, it requires a large storage space. Furthermore, to generate a random matrix has high requirements for hardware which will bring excessive hardware costs. Through the in-depth study of LEACH algorithm, this paper develops CS_LEACH scheme based on LEACH algorithm combined with CS theory. According to this scheme, a measurement matrix based on nodes clustering is constructed. Experimental results show that CS_LEACH scheme effectively extends the lifetime of WSNs and solves the problem that random matrix needs storage in advance.In order to better study the recovery algorithms, this paper constructs the data acquisition framework of compressive sampling and online recovery framework(DAF_CSOR) based on CS_LEACH scheme and conducts several experiments to analyze the advantages and disadvantages of different recovery algorithms respectively based on 1? minimization, greedy algorithm and Bayesian recovery algorithm. Experimental results show that the joint action of the newly constructed measurement matrix and DAF_CSOR framework is superior to the conventional method in the terms of recovery speed and accuracy. SAMP algorithm performs best in recovery accuracy while it has a poor recovery speed. This paper proposes a new regularized sparse adaptive matching pursuit(N_RSAMP) algorithm which improves the original atomic selection method and utilizes regularization method to accelerate the algorithm convergence speed of SAMP algorithm. Experimental results show that the N_RSAMP algorithm performs much better than the other algorithms under our framework.