A multiplication-free filter design method based on high-gain filter coefficients

Abstract

The present invention proposes a design method of a multiplication-free filter based on high gain of filter coefficients. A linear-phase transposed FIR filter is used. First, a set of tap coefficients of the original filter is designed to obtain the set of tap coefficients of the original filter. Do high-gain multiplication and rounding for each tap coefficient of the filter to obtain each high-gain tap coefficient; perform factor decomposition on each high-gain tap coefficient of the filter to obtain each decomposition factor, and finally according to the best CSD coding method to the obtained All decomposition factors are designed without multiplication coding, and the structure of multiplication-free filter is obtained. Compared with the existing non-multiplying filter design scheme, the present invention can be adapted to the multiplication-free design of high-gain filter coefficients, and reduces the number of adders and shift operations and the coding word length of filter coefficients, and reduces the realization of the filter. hardware complexity.

Description

technical field [0001] The invention belongs to the technical field of on-board signal processing, and in particular relates to a multiplication-free filter design method based on high-gain filter coefficients. Background technique [0002] Conventional FIR filters require many multipliers when implemented in hardware, the number of which is related to the filter order, which in turn affects the digital filter performance. However, the hardware resources of the spaceborne receiver are very limited, so it is necessary to optimize the design of the filter structure to reduce the hardware resource requirements. For example, the lookup table method and distributed algorithm are used to improve the multiplication efficiency. In addition, people continue to study the implementation of non-multiplying filters, replacing multipliers that require a large number of operations with shift operations and adders, and reducing the number of adders and the word length of sampled data throug...

AI Technical Summary

Problems solved by technology

[0004] The filter coefficient gain of the spaceborne receiver is usually very large. When using a multiplier to implement the filter hardware, a large number of calculations are required, which takes up a lot of resources. However, the existing non-multiplication filter implementation scheme is limited by the filter coefficient gain, which is prone to large quantization error

Images