Cross Layer Scheduling Algorithms for Downlink Multi-Antenna CDMA Systems

In today´s wireless communication systems, the design of efficient packet scheduling algorithms at the MAC layer is proven to have significant impact on their overall performances. In the light of this fact, we propose and compare in this paper different scheduling techniques which aim at satisfying the users´ requirements in terms of both rates and delays. The considered system is a downlink multi antenna code division multiple access (MIMO-CDMA) system which assumes both traffic arrival and users´ mobility. First, the MIMO-CDMA system is modeled as a weighted graph. The weight of each vertex is then updated at each time slot according to a specified scheduling rule. Finally, we solve heuristically a graph coloring problem in order to find a near- optimal scheduling decision. We evaluate through simulations the performance of the proposed algorithms and show that a cross layer design taking the benefits of both MIMO and scheduling may be efficient to address the tradeoff between system capacity and users´ quality of service requirements.