一种基于最优观测矩阵的自适应贝叶斯压缩信道感知联合机制

A Joint Mechanism of Adaptive Bayesian Compressed Channel Sensing Based on Optimized Measurement Matrix

该文采用基于概率模型的贝叶斯压缩感知方法,从最大后验概率角度,给出了压缩信道感知的一般流程。在此基础上,利用自适应贝叶斯压缩感知将信号的重构和观测矩阵的设计结合,使这两个环节不再相互独立。同时,提出一种基于最优观测矩阵的自适应贝叶斯压缩感知联合机制,通过减少观测矩阵的相关度以及对观测矩阵的自适应设计,使得信道的重构效果更佳。另外可利用重构过程中得到的差错栏,对重构精确度进行衡量。仿真表明:在相同的实验条件下,该联合机制相比传统的重构算法,具有更好的抗噪声能力和重构精度。

In this paper, a common process of compressed channel estimation is given by the Bayesian Compressed Sensing （BCS） which is based on the probability principle of Maximum A Posteriori （MAP）. In the process, signal reconstruction and measurement matrix design as two separate steps can be combined together by Adaptive BCS （ABCS）. Meanwhile a joint mechanism of ABCS and optimized measurement is proposed by reducing the coherence and the adaptive design of measurement matrix to get a better reconstruction performance. Furthermore the error bars obtained in the process of reconstruction can be used to measure the accuracy of the reconstruction. Simulation results show that under the same conditions, the joint mechanism shows better anti-noise ability and recovery accuracy than those of the traditional reconstruction algorithm.