Research Of Gauss-Seidel Fast Affine Projection Algorithm

The key of adaptive filtering is adaptive algorithm. Acoustic echo canceling is a technique that can update the estimate of actual echo path by an adaptive processing and discern the echo path by adaptive filtering. Researching different hardware and different adaptive algorithm is a primal methodology to solve the problem of acoustic echo canceling. This paper analyzes the relationships between the convergence of the Least Mean Square algorithm and convergence factor and number of taps. And the paper proves these relationships by MATLAB emulation. Then, the paper researches Gauss-Seidel Fast Affine Projection Algorithm (GSFAP) and researches it's Field Programmable Gate Array (FPGA) based design. Finally, the emulation a core algorithm of GSFAP based on FPGA is accomplished. The results show that: (1) In order to improve the effect of adaptive filtering in acoustic echo canceling, the important path is to adopt new hardware and new algorithm. (2) GSFAP is an excellent adaptive because that it converses fast and has low computing complexity and admits the taps'number greater than 1000. (3) It can not only reduce the algorithm but also make the algorithm's FPGA design easy to implement that vector computing replace matrix computing and the simplification of the Gauss-Seidel iterate expression equation. (4) In the process of designing of GSFAP based on FPGA, the paper decomposes the system in modular. That saves the design time and makes design easy. Another way of saving design time is associate the VHDL and Library element in Quartusâ…¡. Fast Affine Projection Algorithm (FAP) is good used of acoustic echo cancellation. It is difficult to implement the GSFAP in FPGA directly. But the optimization way given by the paper can implement this design.