针对无线传感器网络( WSN)中数据运算量大及传输速率受限问题，提出了在分布式扩散LMS算法中先后采用符号函数和改进的符号函数量化策略，实现对误差和估计信号的量化，以减少多节点无线传感器网络的运算量和通信量。仿真结果表明，采用符号函数的量化策略时，误差相对较大；采用改进的符号函数量化策略，在设定适当的量化门限时，所提出的算法性能十分接近非量化的扩散LMS算法，且优于非协作算法。特别对于BPSK信号，相较于传统的扩散LMS算法，性能更为优良。%To deal with the problem of heavy computation and limited data rate in Wireless Sensor Network ( WSN) , a quantization scheme with the sign and the modified sign functions in the diffusion least-mean square ( LMS) algorithm is proposed in this paper. It can effectively reduce the computational complexity and communication burden when applied to quantify the error signal and the estimated signal in WSN. Simulation results show that the performance of the proposed scheme with appropriate quantization threshold is approaching to the non-quantitative diffusion LMS algorithm and is superior to the non-cooperative algo-rithm. Especially for the BPSK signal, the proposed “modified sign-data” scheme outperforms the tradi-tional diffusion LMS algorithm.
展开▼
