NOISE REDUCTION IN A REMOTE MEASUREMENT SYSTEM BY USING DSP SOFTWARE METHODS

摘要

In this paper, we propose the use of the digital signal processing method to reduce unwanted noise instead of our previously proposed method of using repeatedly measured response waveforms and software average methods. We pass the measured waveforms with their excessive noise through digital FIR filters to reduce the influence of noise at either the local or remote computers. Here, we use three types of the digital FIR filters: a FIR filter designed with a Hamming window, with a Kaiser window and an equiripple linear-phase FIR filter based on a Parks-McClellan algorithm. In addition, we compare their filtering performance that depends on the Minimum Mean Square Error (M.M.S.E.) with respect to the different filter order N. Our experiment results show that the higher filter order N is not positively in proportion to the better of M.M.S.E.. However, all of the three filters can provide a range of improvement of from about 30 percent to 50 percent in M.M.S.E.. Finally, we compare the computation cost. The experimental results show that the computation cost is in proportion to the filter order N, but the computation cost is different from the varied digital filters.