An Efficient and Robust Lossless Compression Scheme for Wireless Sensor Networks

Abstract

In wireless sensor networks (WSNs), a large number of tiny, inexpensive and computable sensor nodes are usually deployed randomly to monitor one or more physical phenomena. The sensor nodes collect and process the sensed data and send the data to the sink wirelessly. However, WSNs have limitations such as tight energy budgets, limited radio bandwidth, limited memory, limited computational capability, limited packet size and high packet loss rates. These constrains are important issues when designing compression schemes for WSNs. Data compression is one important tool that can maximize data return over unreliable and low rate radio links. Thus, due to the unreliable nature of the radio links in WSNs that result in packet losses, it is therefore very critical to propose a data compression scheme that is very robust to packet losses. In this paper, we propose a block based approach which allows each block of source data to be encoded independently to ensure unique decodability at the sink, thus leading to an efficient and robust lossless compression scheme for WSNs. Simulation results using various real-world sensor datasets show that a maximum percentage energy saving of 29.29% was achieved by our proposed scheme. In addition, although the compression performance of our proposed scheme is comparable with those of LEC, it is however 200% as efficient as S-LZW.