Development of wavelet based signal compression algorithm using adaptive coder in FPGA

Signal communication is the method of sending data from one place to another. Signal compression is the technique, used to reduce the amount of data that must be transmitted in the signal communication. Algorithms that produce better compression ratio and less loss of data in the reconstructed signal are needed for signal compression. This paper presents the wavelet based signal compression algorithm for ECG signal MIT-BIH arrhythmia database record 117, LELECCUM signal and TIMIT audio signal using adaptive coder. The architecture for mean removal, haar wavelet transform, adaptive encoder, inverse haar wavelet transform and mean insertion are present in this paper. The mean removal is used, since the mean removed signal requires less number of binary bits for their representation. The haar wavelet transform is used to decompose the signal. The proposed adaptive encoder is used to minimize the number of bits transmitted. Experimental results prove that the proposed coder outperforms than other coders such as novel algorithm, Djohn, EZW, SPIHT in terms of simplicity and coding efficiency. For the proposed coder, a compression ratio of 9.344:1 is obtained for MIT-BIH arrhythmia database record 117 with a percent mean square difference as 1.2735. A smooth signal like LELECCUM, a compression ratio of 14.267:1 is achieved with a percent mean square difference as 1.3673. A compression ratio of 2.173:1 is achieved for TIMIT audio signal with a percent mean square difference as 1.0269.