An Efficient Code Compression Technique for Embedded Systems

Abstract

Memory plays a crucial role in designing embedded systems. Embedded systems are constrained by theavailable memory. A larger memory can accommodate more and large applications but increases cost, area, as well as energy requirements. Code-compression techniques address this issue by reducing the code size of application programs. It is a major challenge to develop an efficient code-compression technique that can generate substantial reduction in code size without affecting the overall system performance. We present an efficient code-compression technique, which significantly improves the compression ratio. Two previously proposed algorithm are evaluated. The first algorithm is dictionary-based method, provides a small separated dictionary is proposed to restrict the codeword length of high-frequency instructions, and a novel dictionary selection algorithm is proposed to achieve more satisfactory instruction selection, which in turn may reduce the average CR. The second algorithm is mixed-bit saving dictionary selection (MBSDS) the fully separated dictionary architecture is proposed to improve the performance of the dictionary-based decompression engine. This architecture has a better chance to parallel decompress instructions than existing single dictionary decoders Additionally, this paper offers a third algorithm namely, to combine the two previously proposed schemes along with run length encoding to compress the code.