| With the rapid development of Internet of Things(Io T)and cloud computing technologies,the amount of data generated by Io T devices has exploded.On the one hand,it increases the data transmission and reception power consumption of edge Io T devices;On the other hand,it brings challenges to the storage space and data security of cloud servers.The research on Lempel-Ziv-Welch(LZW)data compression algorithm and Joint Compression and Encryption(JACE)algorithm can effectively solve the above two problems,but there are still some problems to be studied.One is that the existing two commonly used dictionary management schemes of the LZW data compression algorithm still have the problem of low compression ratio or slow compression speed;The other is that the existing JACE based on the dictionary compression algorithm still has the defects of large storage occupation,heavy computational burden,and low seed security.Therefore,the LZWm＿n algorithm and the SLZWm＿n algorithm have been proposed,based on the existing dictionary management schemes of the LZW data compression algorithm and JACE based on the dictionary compression algorithm.The dictionary management scheme of the LZWm＿n algorithm is based on dictionary partition and influence factor,which introduce spatial locality and a new dictionary management scheme to have a high compression ratio while maintaining a reasonable compression speed and decompression speed.SLZWm＿n algorithm is based on the key-stream generated by Pseudo Random Number Generator(PRNG)with Physical Unclonable Functions(PUF)as the seed to scramble the dictionary entries,xor the output encoding,and randomly insert new entries so that it has certain security while maintaining a high compression ratio.LZWm＿n algorithm and SLZWm＿n algorithm have been implemented on the Raspberry Pi3B+.The performance of the compression algorithm,the randomness of the key-stream,and the security of the encryption algorithm are tested based on Canterbury Corpus,NIST test project,and USC-SIPI image database respectively.The experimental results show that the average compression ratio of the implemented LZWm＿n algorithms with 8 dictionary sizes is greater than 2.40,the average compression time is less than 0.85 seconds,and the average decompression time is less than 0.40 seconds,compared with the existing improved LZW algorithm,it has a higher compression ratio while maintaining a reasonable compression speed and decompression speed;The implemented SLZWm＿n algorithms of 8 dictionary sizes maintain the same compression ratio as LZWm＿n algorithms;The key-stream used by SLZWm＿n algorithm passes the NIST randomness test;The correlation coefficient of the images encrypted by the SLZWm＿n algorithm drop to about 0.02;The Pixels Change Rate(NPCR)of the plaintext sensitivity test reaches more than 99.5%,and the Unified Average Change Intensity(UACI)is about 33%,which has certain security. |
