Minimum-length scheduling in wireless networks with multi-user detection and interference cancellation: Optimization and performance evaluation

Minimum-length scheduling has been a subject of extensive research in performance engineering of wireless ad hoc networks. We consider the problem under the unconventional setting of multi-user detection receivers that can perform decoding successively, potentially admitting interference cancellation. Thus, in contrast to classical spatial reuse, strong interference is not necessarily harmful. The paper studies fundamental aspects of assessing the achievable performance by successive decoding and interference cancellation versus that of the conventional setup. Computationally, the task is very challenging - the optimal decoding pattern is intertwined with the composition of time slots. We provide structural results of the problem, and develop integer programming formulations enabling to effectively determine the optimal schedule. Computational experiments are provided to shed light on the achievable performance gain of successive decoding with interference cancellation for minimum-length scheduling.