LAUSR

In this paper, a hybrid data compression scheme based on predictive compressed sensing (CS) and light weight lossless compression is suggested for wireless sensor networks (WSNs). CS based techniques are well motivated in WSNs not only for sparse signals but also by the requirement of efficient in-network processing in terms of transmit power and communication bandwidth even with nonsparse signals. This algorithm exploits prediction-based approach in which the difference between the actual measurements and the predicted measurements of the dataset is encoded using CS technique with reasonable error. The CS encoded data is further compressed using Huffman encoding to improve the compression ratio without any loss in quality. We analyzed the performance, data rate saving and inaccuracy introduced by the hybrid compression algorithm. The post processing analysis shows high compression ratios, with acceptable mean squared error.