Radio Resource Allocation Research In Heterogeneous Network

In order to meet the challenge brought by smart terminal and mobile Internet, cellular networks combine with other wireless access networks to increase the system capacity and provide users with better service.Relays can help to forward information and extend the coverage of macrocell. However, relays can’t communicate with mobile station or base station independently. That’s why we access relays to the network through broadband in this thesis. Relays can share the spectrum with macrocell, which make users can communicate with both relay and base station at the same time. In cellular/relay het-network, relays afford high rate service while cellular macrocell provide broad cover. Their cooperation helps to increase system spectrum utilization. As to the network selection problem, we introduce Signal to Interference plus Noise Ratio (SINR) threshold a in the network selection algorithm. After connecting to relay micro base station, we compare the received SINR to a in order to decide whether connect to relay and base station at the same time. Simulation results show that our new proposed network selection algorithm obtains a better performance compared the typical maximum received power scheme and the minimum channel gain scheme.In het-network, service type and the attribution of MS lead to load imbalance. At the same time, radio resource allocation (RRA) in het-network also faces other challenges such as long time delay and high complexity. Therefore we design a stepwise subcarrier allocation scheme considering load balancing. Firstly, subcarriers are allocated to every user to meet the minimum rate requirements, then each remaining idle subcarrier is allocated by greedy algorithm. During the resource allocation process, base station keeps monitoring the real-time loads of different access network nodes. These statistics are used to adjust the subcarrier allocation scheme and network selection scheme. System simulation proves that the proposed algorithm improves the effective throughput of the system, as well as load balancing.