%0 Journal Article
%T Robust Data Compression for Irregular Wireless Sensor Networks Using Logical Mapping
%A Thanh Dang
%A Nirupama Bulusu
%A Wu-chi Feng
%J ISRN Sensor Networks
%D 2013
%R 10.1155/2013/253257
%X We propose RIDA, a novel robust information-driven data compression architecture for distributed wireless sensor networks. The key idea is to determine the data correlation among a group of sensors based on the data values to significantly improve compression performance rather than relying solely on spatial data correlation. A logical mapping approach assigns virtual indices to nodes based on the data content, which enables simple implementation of data transformation on resource-constrained nodes without any other information. We evaluate RIDA with both discrete cosine transform (DCT) and discrete wavelet transform (DWT) on publicly available real-world data sets. Our experiments show that 30% of energy and 80¨C95% of bandwidth can be saved for typical multihop data networks. Moreover, the original data can be retrieved after decompression with a low error of about 3%. In particular, for one state-of-the-art distributed data compression algorithm for sensor networks, we show that the compression ratio is doubled by using logical mapping while maintaining comparable mean square error. Furthermore, we also propose a mechanism to detect and classify missing or faulty nodes, showing accuracy and recall of 95% when half of the nodes in the network are missing or faulty. 1. Introduction With the continued evolution of sensor networking hardware, the ability to deploy large numbers of sensors is becoming possible [1]. Typically, sensor networks are deployed to gather environmental information with the sensors cooperating to forward data to a data sink. One of the main challenges with such sensor networks is the need to minimize wireless transmissions to conserve energy at sensors [2]. There are several basic ways to minimize the network traffic: in-network storage, data aggregation, and data compression. With in-network storage, sensors store data locally and only transmit data in response to a query [3¨C6]. The network discards old data to store newer data. With data aggregation, different sensors aggregate their data (sum, average, etc.) and only transmit the result to the sink [7¨C11]. Some applications may need raw data. For example, scientists might want to collect raw temperature data in an area to model and forecast the weather, as well as for archival. For such applications, data compression is preferred [6, 12¨C16]. Compression can be applied to the data stream from a single sensor [16]. Good compression performance can be obtained this way as a sensor typically generates similar data over time. However, if the compressed data from a single sensor is
%U http://www.hindawi.com/journals/isrn.sensor.networks/2013/253257/