Resource On Mbms Resource Allocation Algorithms In LTE

With the rapid development of mobile Internet, the proportion of multimedia services in the mobile communications services is increasing rapidly. Efficient use of multimedia broadcast multicast technology to transmit multimedia services becomes a hotspot. LTE is one of the most widely used mobile communication systems, the MBMS technology of LTE makes sharing of resources possible, which eases the growing tension in the radio resource requirements. The study of MBMS resource allocation algorithms can further improve resource utilization efficiency and improve service quality. MBMS system in LTE has two different transmission modes, MBSFN and single cell, this thesis performed research in single cell mode.This thesis introduces the LTE downlink system network architecture and physical resources related technologies, based on this highlighted in LTE MBMS network architecture, channel support, protocol stacks and the transmission mode and business processes. Then this thesis introduces the LTE downlink resource allocation framework and classic unicast resource allocation algorithms. Compared to unicast resource allocation algorithms, this thesis briefly analyzes the classic single-rate and multi-rate multicast resource allocation algorithms.This thesis presents a dynamic MCS selection method for MBMS resource allocation algorithm, this method can effectively improve throughput of MBMS system at a given packet loss rate threshold. The dynamic MCS selection method adjusts MCS selection policy according to the average packet loss rate of users within a multicast group including Min, Max and TC. This thesis proposes three algorithms for MBMS resource allocation, including D-BKM, D-IKM and D-MaxKM. D-BKM use Kuhn-Munkres algorithm to guarantee rate requirement of MBMS services at first step, and then allocate RBs to services at the maximum throughput principle. D-IKM significantly reduces numbers of run of Kuhn-Munkres algorithms by estimating the number of resource blocks. D-MaxKM is the reverse version of D-BKM, which allocates RBs at the principle of maximum throughput at first step, and then use Kuhn-Munkres algorithm to guarantee rate requirement of MBMS services.This thesis also presents a priority-based layered multi-rate MBMS resource allocation algorithm PLRA, PLRA is performed in two stages, base layer allocation and enhancement layers allocation. In the base layer allocation stage, PLRA tends to allocate resources to the multicast group which requires less number of RBs. In the enhancement layers allocation stage, RLRA chooses multicast group to allocate RBs according to the priorities, which takes into account a variety of factors spectral efficiency, throughput, and fairness and so on.Finally, this thesis designs and implements a MBMS system level simulation platform based on Python, which can perform different simulations of different scenarios. This thesis evaluates algorithm performance in throughput, packet loss rate, fairness and the satisfaction rate. Simulation results show that the proposed algorithms in the third chapter can guarantee rate requirement of MBMS services at the cost of limited packet loss rate exaltation. Besides, D-MaxKM achieved higher throughput than other algorithms, and the performance of D-IKM and D-BKM is roughly equal. The algorithm PLRA proposed in the fourth chapter can effectively improve the multi-rate multicast throughput and fairness of the system at the expense of a small amount of spectrum efficiency...