Research & Design Of Digital Audio Signal Processing Technology On The Basis Of FPGA

With the constant development of the science and the new technology, acoustical signal has been processed digitally. Meawhile, the development of Very Large Scale Integrated circuits(VLSI) has been developed,it make it possible to integrate complex digital signal processing system on monolithic circuit,.Compared to the traditional analog signals processing system, digital acoustica signal processing not only can get rid of disadvantage of analog signals,such as temperature drift and so on,but also can improve the precision which is in the signal processing. Meanwhile,it can finish the task in a very short time.This subject mainly studies about the related knowledge of over-sampling inΣ-ΔDAC, and design a interpolated filter which is used in the audio over-samplingΣ-ΔDAC.The role fuction of it is complete the over-sampling in the processing of audio signal, and improve the Signal-to-Noise Rate(SNR) of outcoming signal and over-samplingrate.According to the elementary knowledge of multi-rate change, in order to finish the sampled data interpolation, we needs to rely on the digital filter to complete the function, therefore, it is an unusual important question that how to finich the process of the filter effectively. When they are realizing the interpolation filter, it generally uses the method of the graduation filter,which is reduce the request of the exponent nunber of the filter, simultaneously can reduce the expenses on the hardware.The main purpose of this topic is to adopt a new method to achieve 128 times of the over-sampling rate compared to the original one.In this paper,it is introduce the signal quantizing noise model concisely, simultaneously, elaborate the technology which about the noise resharping inΣ-Δmodulator, also contain the structure and the performance of it. It has researched the principle and the characteristic in the half belt filter and CIC filter, and it has given the corresponding equation, also given the research discussion to the band width in the CIC filter. In this design, the interpolation filter has been divided into three parts. The first level is half belt filter; it have promoted 2 times of sampling frequences, and the second level used 1/4 belt filter to achieve 8 times of sampling frequences. Because the first two levels had already offered the enough band width, the last level completed 16 times of sampling by given third-level of cascade integratorcomb filter(CIC filter). After the MATLAB simulation, it obtains the frequency respond of all levels of filter separately, and obtains the an overall interpolation filter's frequency respond in the fourth chapter. Through the simulation, the overall interpolation filter's stop-band weaken has achieved about -75dB, and the pass band ripple controls 0.05dB, it has conformed to the audio frequency parameter area. In the hardware simulation, It has used simulation software Quartus II which belongs to Altera Corporation to confirm the function feasibility, and it also has used the CSD coding technique to make the FIR filter finish the function by the shift register and the accumulator. we obtained the correlation data through the simulation, and the test data induct in MATLAB, the result is that this interpolation filter have achieved the in terpolation function, by the original not smooth sine wave signal, the output obtained the smooth sine wave, and each parameter conforms to the audio frequency standard, has the feasibility.