一种利用信号周期性减少信息损失的数据压缩方法  

Data Compression Method Using Signal Periodicity to Reduce Information Loss

作　　者：姚登辉 孙正波 张晓勇 YAO Denghui;SUN Zhengbo;ZHANG Xiaoyong(Information Engineering University, Zhengzhou 450001, China;National Key Lab of Science and Technology on Blind Signal Processing, Chengdu 610041, China)

机构地区：[1]信息工程大学,河南郑州450001 [2]盲信号重点处理实验室,四川成都610041

出　　处：《信息工程大学学报》2020年第5期552-558,共7页Journal of Information Engineering University

摘　　要：对于某些资源受限的固定通信链路,采用满足一定要求的有损数据压缩方法是解决通信带宽问题的可行途径。在这种情况下,压缩算法应当充分利用信号本身的特点提高压缩性能。提出一种利用DCT压缩算法进行FSK信号有效传输的新方法,能够针对不同的输入参数结合信号周期性自适应调节压缩算法参数以达到最佳压缩性能。仿真试验表明,在信号参数不断变化且算法压缩比同为20的情况下,新方法与原JPEG标准下的DCT压缩方法相比,数据重构后的信号均方根误差百分比平均降低了33%,相关系数平均提高了26%,数据损失更低且稳定性更强。For some fixed communication links with limited resources,a lossy data compression method that meets certain requirements is a feasible way to solve the problem of communication bandwidth.In this case,the compression algorithm should sufficiently utilize the characteristics of the signal itself to improve the compression performance.A new method is proposed to utilize DCT compression algorithm for effective FSK signals transmission.Its innovation is employing a single period of local data to represent the whole data during compression,while the whole data is obtained by periodic extension of local data during decompression,so that the new algorithm can adaptively adjust the compression parameters according to different input combined with periodicity of signals to achieve the best compression performance.The simulation results show that when the signal parameters are constantly changing and the compression ratio of the algorithm is 20,compared with the original method,the root mean square error percentage of the signal using new method after data reconstruction is reduced by 33%on average,and the correlation coefficient is increased by 26%averagely,and the data loss is lower and the stability is stronger after compression.

关 键 词：数据压缩 DCT算法 2FSK 参数调节 周期性 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]