Fairness-Based Resource Allocation Schemes For Mbms

With the rapid development of the wireless communication technology, MBMS (Multimedia Broadcast Multicast Service) is being extensively studied for the reason that it can effectively utilize precious frequency resource with them mode PTM(Point to Multipoint) and improve the frequency efficiency of the wireless communication system. However, due to the wireless channel having a frequency selective fading characteristics, the MBMS system throughput performance is always limited by the user with the worst channel quality. So in order to increase the throughout of the overall system, people always use the resource allocation to overcome this kind of frequency selective fading without considering the fairness of the user or the service group. Therefore, the paper improves existing resource allocation and proposed two resource allocation algorithm considering inter-group rate-proportionality fairness and intra-group user fairness. The main work is as follows:(1) The resource allocation algorithm based on inter-group rate-proportionality fairness considers that the rate proportionality is constant in the in the multi-user scenario. Firstly, it is the bit loading step. In this step, the bits of each subcarrier can be calculated, and then we adjust the bits allocation to ensure the overall power constraint. Secondly, it is the subcarrier allocation step. In this step, every user is assigned to a subcarrier with the most bits subsequently after initialization. Then the user with the minimum normalized rate is assigned one unused subcarrier with the most bits until all subcarriers are allocated. Finally, a few bits are reduced in greedy way to satisfy the total power constraint.(2) The resource allocation algorithm based on intra-group user fairness firstly assume that in a service group the power of every subcarrier is the same, then allocates the subcarriers to the user to meet the most basic QoS demand. Secondly, we allocate the remained subcarriers to the users with the principle maximizing the overall throughout of the system. Thirdly, it is the bit loading step, in this step, we use the greedy way to further enhance the system throughout.