用于干扰对消的稀疏约束卡尔曼滤波算法

无人机等小型平台上干扰机的收发隔离问题是影响其收发同时工作的关键。对于动态稀疏系统来说,常规的卡尔曼滤波算法并未考虑干扰信号耦合路径的稀疏性,对路径衰减系数的辨识精度不够,导致隔离性能不佳。针对此问题,从KF算法的修正步出发,将其等效为一个凸问题,并在此基础上,增加对待估计参数的稀疏性约束,重新推导了算法的修正步,提出了一种稀疏约束的KF算法,充分利用了待辨识系统的先验信息,提高了估计参数的稀疏倾向性。理论分析和仿真结果表明,新算法能够有效适用于动态稀疏环境下的系统辨识,提高了KF算法对于动态稀疏系统的辨识精度,并且能够将干扰机的收发隔离度提高3～5dB,改善了干扰机的收发隔离性能。

Sparse Constraint Kalman Filter Algorithm for Interference Cancellation

The transceiver isolation of jammers on small platforms such as UAV is the key to affect the simultaneous operation of transceiver. As for the dynamic sparse systems, the conventional Kalman filter (KF) algorithm does not consider the sparsity of the coupling paths of interference signals, and the identification accuracy of paths attenuation coefficients are not enough, so that the isolation performance is poor. Aimed at the problems that starting from the correction step of KF algorithm and regarding this as equivalent to a convex problem, on the basis of this, the sparse constraints on estimated parameters are added, the correction steps of the algorithm are deduced again, and a KF algorithm with sparse constraint (SC) is proposed, by so doing, this makes full use of the prior information of the system to be identified and improves the sparse tendency of estimated parameters. The theoretical analysis and the simulation results show that the new algorithm can effectively apply to the system identification in the dynamic sparse environment, and improve the identification accuracy of KF algorithm for the dynamic sparse system. Moreover, the isolation degree of transceiver of jammer can be improved by 3~5 dB, improving the isolation performance of jammer.