基于FPGA的并行RICE解码技术研究与实现
发布时间:2018-05-05 21:23
本文选题:无损解压缩 + RICE算法 ; 参考:《计算机工程与科学》2017年06期
【摘要】:RICE算法在无损压缩系统有着广泛的应用。由于RICE算法采用了变长的自适应熵编码,因此在解码时需要对压缩流进行逐位判断和解析,这给高速解压缩的实现带来了困难。现有的RICE解码实现在解码速度和通用性上都不理想。针对RICE算法中自适应熵编码的特点,设计了一种基于有限状态机和查找表的并行RICE解码结构,可在FPGA上完成8比特宽度的并行解码,解码速度最高可达176 MB/s;同时,该解码结构适用于编码参数k变化的情况,具有很强的通用性。
[Abstract]:RICE algorithm is widely used in lossless compression system. Because the RICE algorithm adopts variable length adaptive entropy coding, it is necessary to judge and analyze the compressed stream bit by bit in decoding, which brings difficulties to the implementation of high-speed decompression. The existing RICE decoding implementation is not ideal in decoding speed and generality. According to the characteristics of adaptive entropy coding in RICE algorithm, a parallel RICE decoding structure based on finite state machine and lookup table is designed. The parallel decoding of 8 bits width can be accomplished on FPGA, and the decoding speed can reach 176MB / s. The decoding structure is suitable for the change of the encoding parameter k, and has strong generality.
【作者单位】: 中国科学院遥感与数字地球研究所;中国科学院大学;
【基金】:中国科学院遥感卫星数据处理与深加工系统专项改造项目(Y52203101A);中国科学院遥感与数字地球研究所青年基金(Y3SJ6100CX)
【分类号】:TN791
,
本文编号:1849348
本文链接:https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/1849348.html