Research On Energy-saving Technology Based On Data Compression Algorithm For Wireless Sensor Networks

Wireless Sensor Networks (WSN) have very limited energy supply, that efficient utilizationof energy has been a core area of research in WSN. Because of the communication operationsconsume most of the energy in WSN, and the communication module’s power is far more thanthe microprocessor module’s power, hence it’s a feasible solution to reduce the energy by usingdata compression methods. The purpose of this article is to find low algorithmic complexity andefficient data compression algorithms, so as to save energy for WSN.The efficient data compression algorithms are proposed in this paper after studying kinds oftraditional data compression methods and considering characteristics of the data collected byWSN and nodes. The major work is as follows:(1) Lossless data compression for WSN based on modified Bit-level Run-length Coding isproposed: designing a pretreatment process at first, which includes process of datatransformation and rearrange of bits, the pretreatment does not shrink the size of data but tomake the data easier to compress by RLE. Then the modified Bit-level RLE is used to compressthe rearranged bit stream, and Huffman coding applied to record the run-length. After that, theprocess of decoding is introduced. The performance of the modified Bit-level RLE was analyzedand compared with other Huffman Coding, the experimental results show that this algorithm getsbetter energy saving and compression ratio besides low complexity.(2) Lossy data compression for WSN is proposed based on Area Threshold method: allWSN data collected according to time can be considered as points in two-dimensional coordinate,then, the problem of data compression is equal to vector data compression issue. A new methodbased on area computation is proposed which using area as the threshold. The error caused bydata compression can be limited in certain area threshold. Meanwhile, the process of decodingand the methods to judge error are introduced. The performance of the Area Threshold method isanalyzed, the results indicate that proposed method has low algorithmic complexity and better compression ratio, and with good effect of saving energy.