Underwater acoustic channel estimation algorithm based on recursive logarithmic minimum average P norm

Applying least squares regularization could enhance the convergence performance and robustness of the system in hydroacoustic channel estimation applications. Unfortunately, the channel prior information is not well utilized in the current LMP algorithm research, which reduces estimate performance. Thus, we present a recursive algorithm that builds upon the recursive logarithmic least mean P-norm algorithm (RlogLMP) and considers the impact of constraint terms on system characteristics. Building upon the RLS algorithm, we integrate the paradigm constraint term with the RlogLMP method to enhance the speed of convergence and stability of the system. Integrating a paradigm constraint into the RlogLMP algorithmic framework facilitates the utilization of its inherent properties to enhance the efficiency of solving the system. Theoretical study and simulation have shown that the recursive-paradigm RlogLMP algorithm improves both the steady state error and convergence speed of the system in comparison to the RlogLMP method and the LMP algorithm.