Optimal Scheduling and Antenna Configuration

My dissertation examines algorithms for optimal spatial reuse TDMA scheduling with reconfigurable directional antennas. I present and solve the joint beam steering and scheduling problem for spatial reuse TDMA and describe a prototype implemented system based on the algorithms developed. The current prototype achieves up to a 600% speedup over TDMA with a mean of 234% in the experiments to date. I use an optimization decomposition approach to arrive at a working distributed protocol which is provably equivalent to the original problem statement while also producing optimal solutions in an amount of time that is at worst linear in the size of the input. This is, to the best of my knowledge, the first actually implemented wireless scheduling system based on dual decomposition. Scheduling for spatial reuse (whether explicit or implicit) is fundamental to the efficient use of finite radio spectrum across a wide range of communication systems.