| Heterogeneous networks can enhance the capacity of the communication system, which have improved the system performance of the conventional cellular network. And with low economic cost and more flexible of deployment, it is becoming a major network structure of the communication network. Because OFDMA technology can provide high rate of data transmission and high spectrum efficiency, the study of OFDMA has become a focus. The channel allocation severely affects the reliability of system, an important indicator to measure the communication system, which has been the focus of research. In order to improve the reliability of OFDMA-based heterogeneous network, the channel allocation is studied and some channel allocation algorithms are designed. The main work of this thesis:Firstly, the channel allocation of one-cell heterogeneous network is studied. According to the Greedy algorithm and the WSA algorithm, a new allocation algorithm, Global Greedy algorithm, is proposed. The system model is introduced in detail and mathematical model for the channel allocation is built. And then three traditional channel allocation algorithms are described in detail, which include traditional Greedy algorithms, WSA algorithm and the Hungarian algorithm. Global Greedy algorithm is then proposed and described in detail. Finally, the algorithm is verified by matlab in different network scenarios. Simulation results show that the performance of the Global Greedy algorithms is better than the traditional Greedy algorithm and WSA algorithm and that its BER performance curve is closer to that of the Hungarian algorithm.Secondly, the channel allocation of multi-cell heterogeneous network based on OFDMA is studied. And channel allocation algorithms are proposed, described and validated. The co-channel deployment of different cells, especially between low power base stations and Macrocells, causes severe interference which affects the performance of the communication network. Some channel allocation algorithms dealing with interference are proposed, which include interference-based Greedy algorithm, interference-based WSA algorithm and interference-based Hungarian algorithm. The BER performance of these algorithms is verified by matlab in different network scenarios. Simulation results show that the performance of the interference-based Hungarian is best and then that of the interference-based WSA algorithm. However, the BER performance of theinterference-based Greedy algorithm is worst. In addition, the reliability of these algorithms is also analyzed, compared and summarized. |
