Adaptive Filtering Algorithm And Its Applications In Echo Cancellation

Along with rapid development of communication techniques~ digital signal processing and VLSI techniques, The subscriber demand higher and higher speech quality in communication. So Echo Canceller (EC) techniques has became a hot issue of competition in the famous communication company all over the world. This paper is just the result of cooperation with Huawei corporation to inspire the development of our country. Echo problem exists in many environments such as two-wire flill-duplex data transmissions PSTN system Audio teleconferencing system~晻, the solution is different in some way. This paper is focused on the EC in PSTN and is organized as follows: The background knowledge of EC such as how does the echo come into being? How can EC cancel the echo and main component of EC is introduced in the first chapter. In the second chapter, we mainly discuss many kinds of adaptive filter algorithms. Especially focused on algorithms which can made a tradeoff between performance and compute complexity. For real-time required PSTN system, the lower computational complexity can significantly decrease the cost of the Program Controlled Switch Device. Then we introduce the main algorithm of our EC, including Double Talk detection~. Nonlinear Processing~. Comfort Noise and realization with TMS320C6201 in the third chapter. In the last chapter, we give out the experiment results of our EC in the real communication environments, mainly the steady residual echo and rate of convergence. The results show that many specifications of our EC has reached or exceeded the ITU-T G.168 standard.