PROJECT TITLE :

Code Compression for Embedded Systems Using Separated Dictionaries - 2017

ABSTRACT:

Engineers should think about performance, power consumption, and value when planning embedded digital systems; furthermore, memory could be a key issue in such systems. Code compression is a technique employed in embedded systems to scale back the memory usage. BitMask-primarily based code compression may be a changed version of dictionary-primarily based code compression. The essential purpose of BitMask is to record mismatched values and their positions to compress a bigger number of directions; it can be used completely or incorporated with the reference directions to decode the codewords. In this paper, we have a tendency to applied a small separated dictionary, and variable mask numbers were used with the BitMask algorithm to cut back the codeword length of high-frequency directions. Likewise, a novel dictionary choice algorithm was proposed to extend the instruction match rates. The totally separated dictionary technique was used to improve the performance of the decompression engine while not affecting the compression ratio (CR) (the compressed code size divided by original code size). Based mostly on the experimental results, the proposed methodology will achieve a seven.five% improvement within the CR with nearly no hardware overhead.