最优路径搜索压缩感知水声通信信道估计

An improved underwater acoustic communication channel estimation based on A^(*) orthogonal matching pursuit algorithm

将信号恢复中最优路径搜索的A^(*)正交匹配追踪(A^(*) Orthogonal Matching Pursuit,A^(*)OMP)伪贪婪算法引入到水声通信信道估计中,可以有效改善正交匹配追踪(OMP)算法容易陷入局部最优的问题,并提出了一种改进型的A^(*)OMP水声信道估计算法。改进了路径初始化方式,同时为了避免过多迭代引起的未知误差,将前后两次迭代残差之差作为停止准则。在正交频分复用(OFDM)通信体制下,对OMP、A^(*)OMP和本文改进方法的估计误差和误比特率进行了仿真对比,随着信噪比的增加,改进方法未出现误差平台,且受导频间隔影响较小。仿真结果表明相对于OMP算法和传统A^(*)OMP算法,在高信噪比下改进方法的估计误差分别降低约2和1个数量级,海试数据结果验证了改进方法的可行性,其误比特率分别平均降低42.0%和4.7%。

A^(*) Orthogonal Matching Pursuit(A^(*)OMP),a pseudo greedy algorithm that can improve performance of OMP,is applied to the underwater acoustic channel estimation.Furthermore,by optimizing the initialization procedure and applying different criterion of stopping iteration in order to avoid unknown errors,we proposed a modified A^(*)OMP channel estimation method.Simulation and sea trial based on an Orthogonal Frequency Division Multiplexing(OFDM)OFDM system are carried out to test the performance.And the influence of the pilot interval is not obvious on performance of modified A^(*)OMP.The numerical simulation shows that the error of proposed algorithm is lower by 2 and 1 orders of magnitude respectively compared to OMP and A^(*)OMP.The results show that the proposed method outperforms the OMP and traditional A^(*)OMP algorithm in the estimation accuracy and the communication output bit error rate,and reduces the bit error rate by about 42.0%and 4.7%respectively.