数据压缩的方法很多,实际应用中多采用变换加编码的方法,在允许一定的误差的范围内可以获得比无损压缩高得多的压缩率,而且常常大大简化处理算法。采用一种分块的动态归一化将需要压缩的数据收缩到[-1,1]的区间内,再采用Llyod算法对归一化的数据进行非线性标量量化编码降低每个采样点的比特位宽。算法简单,易于硬件实现,解码时只需查找码书和动态恢复。在50%压缩比情况下EVM值在1%以内。并针对该算法进行了MATLAB仿真和硬件代码的编写。%There are many data compression methods, and transform coding is mostly adopted in practical applications ,and it could achieve much higher compression ratio than the lossless compression within a cer-tain allowed error range while the processing algorithms is greatly simplified. The compressed data is shrunk into the range of [-1,1]with a blocked dynamic normalization, then the normalized data is coded with non-linear scalar quantization by Llyod algorithm,thus to reduce the bit width of each sampling point. This al-gorithm is simple in hardware implementation,and only to look for codebook and conduct the dynamic re-covery is needed for decoding. EVM value would be less than 1% at 50% compression ratio. Aiming at this algorithm, MATLAB simulation and hardware coding are carried out.
展开▼
