Research Of Filtering Algorithm Based On Lateral Inhibition Of Auditory System

In recent years, with the rapid development of computer science and Microelectronics, the technology of digital signal process has also been developed rapidly. Fi ltering algori thm is one of the most important issues of DSP technology. People have proposeda variety of filtering algorithm for different signal characteristics in many application environments. The phenomenon of lateral inhibi t ion is a prevalence phenomenon in biology, for example in visual, auditory, tactile and other sensory systems of organisms. In this paper, we propose a new filtering algorithm based on lateral inhibition of human auditory system, and studied the theory of lateral inhibition mechanism and its application.This paper starts with a variety of lateral inhibi t ion phenomenon in different sensory systems, and summarizes the current lateral inhibit ion neural network mathematical model and its classification, and briefly discusses the appl icat ion of lateral inhibi t ion in the image processing. And then it details the human audi tory system, including the outer ear, middle ear and inner ear, and establishes the mathematical model of lateral inhibitionfi1tering algorithm according to structure of human auditorysystem.The simulation of this filtering algorithm has been implemented by MATLAB. By processing standard s inusoidal signa 1 with noise, we analyze the performance of the algoriihm. From the results of these experiments, we conclude that this algorithm has a dynamic filtering characteristic. It can change its filtering character ist ics in real time according to the characterist ics of the input signal.At last, this filtering algori thm has been used to speech enhancement, andits performance in speech enhancement has been analyzed according to the contrast with the result of the speech enhancement method using spectral subtraction.