Fault tolerant and channel aware target localization in wireless sensor networks that use multi-bit quantization

Technological advancements have made the development of low-cost, low-power, and multifunctional sensors feasible. Shortcomings of individual sensors such as limited accuracy, range, computational capability, energy reserves, and communication bandwidth may be overcome by organizing sensors into networks. Locating the position of a target or other entity is a very important application of sensor networks for commercial, military, and homeland security applications. Network bandwidth limitations make transmissions of analog data impractical, making localization based on quantized data important. Sensor faults and communication channel imperfections degrade the localization performance of a wireless sensor network. Extending the current work for a fault tolerant and channel aware (FTCA) network using binary data we developed a model for FTCA localization with multi-bit data. Simulations show that localization algorithms that are FTCA using the maximum likelihood framework and multi-bit quantization provide better localization performance in terms of root mean square (RMS) estimation errors which are close to the theoretical bounds defined by the Cramer-Rao lower bound (CRLB) even with a relatively small number of sensors.