Toward energy efficient multistream collaborative compression in wireless sensor networks

Wireless sensor networks possess significant limitations in storage, bandwidth, and power. This has led to the development of several compression algorithms designed for sensor networks. Many of these methods exploit the correlation often present between the data on different sensor nodes in the network; however, correlation can also exist between different sensing modules on the same sensor node. Exploiting this correlation can improve compression ratios and reduce energy consumption without the cost of increased traffic in the network. We investigate and analyze approaches for compression utilizing collaboration between separate sensing modules on the same sensor node. The compression can be lossless or lossy with a parameter for maximum tolerable error. Performance evaluations over real world sensor data show increased energy efficiency and bandwidth utilization with a decrease in latency compared to some recent approaches for both lossless and loss tolerant compression.