论文部分内容阅读
本文阐述了线性预测差分编码法(LPCM)应用于模拟地震信号数据压缩的理论和方法;通过蒙特卡洛法(Monte-Caromethod)在DJS-130机上对容量为256的样本作了实验,取得了较好的效果;经压缩后的信息可使每抽样编码字长降低2比特以上,使量化信噪比增益等于13db以上;对影响压缩效果的各种因素进行了分析,发现在给定条件下最佳预测阶数为6,且增大直流电平可以明显地改善压缩效果,LPCM对信号压缩时具有自适应性;提出了硬件实现的设想。
In this paper, the theory and method of linear predictive differential coding (LPCM) applied to analog seismic data compression are described. An experiment on a DJS-130 with a capacity of 256 is performed by Monte Carlo method Better results; the compressed information can make each sample code word length reduced by more than 2 bits, so that the quantization signal to noise ratio gain equal to 13db or more; the factors that affect the compression effect were analyzed and found that under given conditions The best prediction order is 6, and increasing the DC level can obviously improve the compression effect. LPCM has the adaptability to the signal compression; and the idea of hardware implementation is put forward.