Shared Channel Ordered Transmissions for Energy-Efficient Distributed Signal Detection

An existing approach to energy-efficient signal detection in sensor networks is to order the sensor transmissions such that highly informative sensors enjoy higher priority for transmission. Then, when sufficient evidence is collected at the fusion center (FC) for decision making, the transmissions are stopped. This scheme incurs the same error probability as the optimum unconstrained energy approach, but with fewer sensor transmissions leading to significant energy savings. However, it is assumed that each sensor uses an orthogonal channel to transmit to the FC. In this paper, the ordering approach is revisited by considering the shared medium between sensors and the FC. After the $N$ sensors make observations they wish to transmit to the FC, time is divided into frames and within each frame, sensors with similar likelihood ratios contend for transmission. The contention for transmission is resolved using Slotted ALOHA. Under certain conditions, this new scheme requires less than half the sensors on an average to transmit, with no loss in error probability for the mean-shift binary hypothesis testing problem.