Interference Alignment Based Link Scheduling Algorithm under MIMO-pipe Model

Although interference alignment (IA) is capable of enhancing the spectrum utility from the aspect of information theory, its performance adopted in link scheduling was limited by interference model. Compared with the commonly used interference model such as protocol interference model and physical interference model, the newly proposed ‘MIMO-pipe’ model was developed to describe the tradeoff between multiplexing gain and diversity gain in MIMO communication systems. For the purpose of considering the effect of IA on the performance of link scheduling with MIMO-pipe model, the equivalent form of MIMO-pipe model under the IA circumstance was derived. A set of pairs between feasible transmission rate and the requirement of signal to interference and noise ratio (SINR) for each MIMO link was extracted, and a link scheduling optimization problem was formulated. In order to solve such an optimization problem, a novel distributed algorithm based on the equivalent model was proposed. The theoretic analysis and simulation results demonstrate that the equivalent MIMO-pipe model was propitious to reveal the characteristic of IA. What’s more, the total number of scheduling time slots obtained by the distributed link scheduling algorithm with IA is obviously less than that without IA while the throughput decreases a little.