IIR Digital Filter Research And Design Based On FPGA

With the high speed development of digital signal processing (DSP) technology, the traditional analog signal processing technology has been gradually replaced by the sophisticated digital signal processing technology. Directly lead to the rapid development of DSP technology is the digital computer technology breakthrough and the rise of the integrated circuit manufacturing technology. Due to the saltant progress of digital computer specific digital hardware, the computer now became powerful, compact, inexpensive and fast. Many of the previously only analog system to achieve signal processing tasks are now transferred to cheaper, more reliable digital systems. In modern electronic systems, such as image, video, audio and communication system is needed to have the flexibility and real time in signal processing, the traditional DSP clearly can not meet these needs. But with the development of programmable logic devices, FPGA in flexibility and real-time signal processing advantage emerged, therefore digital signal processor based on the FPGA is widely used.This paper mainly studies the IIR digital filter design and implementation based on FPGA. The article first introduces the research background and significance, and the research status of digital filter design at home and abroad. Then based on the IIR digital filter design theory, separately from the conversion from analog filters to digital filters, IIR digital filter structure and the digital filter finite word length effect make a elaboration and discussion, compare and analysis the filter structure and the characteristics of their respective properties, and finally determine the structures used in this article.Secondly, considering the finite word length effects in the digital filter design process, to ensure that the filter stability, this paper presents a new stability constraint condition based on Argument Principle (AP), and put forward a weighted least square method to design IIR digital filter, and four comparative examples prove the effectiveness of stability constraints based on AP. Next, in the Matlab environment, combined with the filter selection and the discussion of the filter stability constraints, we designed with two order loop filter based on elliptic parallel structure, and through the Simulink dynamic simulation environment, the designed finite word length of IIR digital filter is dynamic simulated. And with the same design method, we design different center frequencies of IIR digital filter used in digital audio frequency for frequency separation and extraction. Simulation of the entire process is stable and smooth, and the selected frequency characteristic is good. Finally, in the Simulink environment, we could derive the generated Verilog code after the dynamic simulation of the finite word length IIR digital filter, and downloaded the code to FPGA development board for hardware test. Through continuous programming to control the development board ADC and DAC module, and through the use of an external signal source, spectrum analyzer, we could analyze and compare the effect before and after filtering.The test results of filter system showed that, the designed IIR digital filter in this paper basically to meet the design targets, and the selected frequency characteristic is a good. In practical application, we could modify and control filter coefficients flexiblely according to different needs, so as to change its filtering performance.