一种新的多普勒天气雷达回波数据无损压缩  

Novel lossless compression of Doppler weather radar data

作　　者：黄云仙[1] 马烁[1] 艾未华[1] 

机构地区：[1]解放军理工大学气象学院,江苏南京211101

出　　处：《解放军理工大学学报（自然科学版）》2012年第2期232-236,共5页Journal of PLA University of Science and Technology(Natural Science Edition)

摘　　要：为了解决多普勒天气雷达回波数据传输的实时性以及数据存储的有效性问题,提出了一种基于阶梯量化的多普勒天气雷达回波数据无损压缩新算法。针对天气雷达回波数据的特点,首先对天气雷达回波数据进行预处理,即将雷达回波数据进行格式转换;其次提出了一种阶梯量化方法,并采用MED预测进一步去除数据中的相关性;最后对量化和预测后的回波数据采用LZW编码以实现无损压缩。实验结果表明,新方法较其他方法大幅提高了回波数据的无损压缩倍数,可实现对多普勒天气雷达回波数据的有效压缩。Doppler weather radars produce large amounts of data,which this has important implication for the archivment and analysis of data.To solve the real-time Doppler weather radar data transmission and storage efficiency,a novel lossless compression algorithm based on step-quantization was presented.Considering the characteristics of the weather radar data,a pretreatment scheme that is the format transformation of the weather radar data was given firstly.Secondly,the method of step-quantization was proposed to quantize the radar data and the correlation of the data is further wiped off through MED prediction.Finally,the data after step-quantization and MED prediction was losslessly compressed using the LZW code.Experimental results show that this new method,compared with otheralgorithms,greatly improves the lossless compression ratio and is very effective in the lossless compression of Doppler weather radar data.

关 键 词：多普勒天气雷达 阶梯量化 MED预测 无损压缩 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]