Sensor communications has been a bottleneck in automotive vehicles. It is extremely important for the automotive industry to improve vehicle safety, reliability, fuel economy, reduce emissions and cost. Powerful electronic control units and the associated distributed sensors and actuators have been introduced. Revolutionary changes from the conventional wired sensors to wireless sensors are required to improve vehicle performance, reduce vehicle weight and cost. This research project undertakes a multidisciplinary and cross-layer approach to develop UWB intra-vehicle sensor network theory and methods. Packet detection theory will be established considering channel propagation statistics and multiple access interference for an intra-vehicle sensor network consisting of 50 mission critical sensors and 50 non-mission critical sensors. Methods will be developed to achieve the packet loss rate of in the detection stage. Reliable MAC protocols will be developed for multi-channel direct-sequence spread spectrum ALOHA UWB sensor networks to guarantee that the maximum delay is ms and the probability of message failure is for message delivery through non-line of sight intra-vehicle channels. Network performance will be analyzed including message throughput, packet loss rate and BER in the presence of multiple access interference considering UWB channel propagation effects and transmission power uncertainty. A UWB intra-vehicle sensor network testbed will be designed and built using cross-layer approaches for real time control and computing. This project can help to reduce the parts cost for intra-vehicle sensor communications and improve fuel economy, reliability and safety. Students will be trained in communications, networks, real-time operating systems and programming in embedded systems.
