Greedy spectrum sharing for cognitive MIMO networks

Cognitive radio is regarded as an emerging technique developed to enhance the efficiency of spectrum usage. In this paper, we consider a point-to-multipoint secondary network sharing frequency bands with several primary transmitters and receivers. Assuming a cognitive base station equipped with multiple transmit antennas and single antenna receivers, we propose a greedy spectrum sharing algorithm in order to approach the maximum achievable sum rate of the secondary network. First, we model the overall studied system as a weighted graph, and then we formulate the spectrum sharing as a vertex coloring problem. Second, we show that the two problems are similar and we design a greedy algorithm to solve both problems using two selection criteria. We show through computer simulations that the proposed algorithm is able to achieve near-optimal performances with very low computational complexity.