The Bionic Research Of The Human Ear Frequency Characteristic Based On FPGA

Neuroscience is the science of exploring human brain, mainly about human neuralsystem structure and function. At present, it is known that human brain has been the mostcomplex and intelligent organ in the biological field. The intelligent informationprocessing of human brain has become the forefront of the information processingtechnology, and the intelligent bionic about human brain is now the hotspot of intelligentinformation processing field.As to the sound processing, human ear has the incomparable superiority to theexisting artificial sound processing system. By simulating human auditory frequencycharacteristics, a new method for sound recognition and processing is provided.Nowadays, most research based on human auditory frequency are still in the stage ofsoftware stimulation, while only a small number of research has achieved the humanauditory frequency selectivity by using analog filters. But there are so many shortcomingson the usage of analog circuits, for example, the circuit usually takes more space, has poorstability and is inconvenient for changing parameters.In this paper, by studying the physiological structure of the human auditory whosefrequency characteristics is stimulated by a set of Gammatone filters, the technology ofauditory digital filter banks is adopted. After that, a method of implementing Gammatonefilter based on FPGA is advanced on the basis of software simulation. The design of wholesystem is accomplished in VHDL language on FPGA module and the system is testedonline by SignalTapII. Compared to the design method of analog filters, this digital designmethod has a variety of advantages of strong anti-jamming compatibility and shortdebugging period.