基于卡尔曼滤波的时变水声信道估计

为了能同时利用时变水声信道的簇状稀疏特性和时间相关性，构造了一种时变水声信道模型，并基于该模型对传统的卡尔曼滤波压缩感知算法进行改进。该方法主要利用前一时刻估计的信道状态响应来确定当前时刻信道的候选支撑集，并以此构造时变水声信道的状态转移方程。通过卡尔曼滤波迭代的方法计算候选支撑集上的系数，最后通过阈值法滤除误差原子。仿真结果表明：该方法能有效地利用水声信道间的时间相关性来提高信道估计的性能，同时由于水声信道存在簇状稀疏特性，因此经该方法也具有一定的鲁棒性。

Time varying underwater acoustic channel estimation based on Kalman filter

By using the cluster-sparse characteristics and temporal correlation of underwater acoustic channel, a time-varying underwater acoustic channel model is constructed and the traditional Kalman filter compressive sensing algorithm is improved under this model. In the proposed method, the channel state response at the previous moment is used to determine the candidate support set at the current moment, and then the state transfer equation for the time-varying underwater acoustic channel is established. The coefficients in the candidate support set are calculated by Kalman filtering, and the error atoms are filtered out by threshold method. The simulation results show that the method can effectively utilize the temporal correlation between underwater acoustic channels to improve the performance of channel estimation, and the cluster characteristics of underwater acoustic channels makes the method robust.