Improved LZW algorithm of lossless data compression for WSN

Wireless Sensor Network (WSN) nodes are powered by limited battery and communication consumes most of the energy. Therefore, it's not appropriate to transmit data directly in the networks while compressing data before sending is an effective method to save energy for nodes. According to the need for fully recovering the compressed numerical data, based on the LZW (Lempel-Ziv-Welch) algorithm, this paper proposes an improved lossless data compression algorithm for WSN nodes. In this algorithm, the span of data to be compressed is reduced by calculating increment between two adjacent data of sample sequence. The increment is replaced by a single character, where 1 ∼ 25 is orderly replaced by A ∼ Y, −1 ∼ −25 by a ∼ y, 0 by Z, and the data out of the span is marked by character z, so as to save memory space and improve data duplication possibility. Taking into account the different real-time requirements on data compressing, two specific algorithms for real-time transmission and latency tolerant transmission are given. Comparing test between the improved LZW algorithm and the original is made, and the results of testing data compression show that the two improved LZW algorithms have remarkable superiority in reducing dictionary length and increasing the compression ratio.