Resource Allocation For Coding Multicast In OFDM Systems

With the rapid development of information technology and services, it is becoming the trend to develop multimedia multicast service in wireless systems. Compared with the unicast, conventional multicast (CM) has many benefits, but different users often experience different channel conditions, performance of CM is limited by the least channel gain (LCG) user, which is a great challenge to enhance the system performance. Coding multicast, that is, the multicast data is coded before data transmission. The quality of coded data has scalability where the more the data receive, the higher the recovery of data quality. Therefore, the performance of coding multicast is not restricted by the LCG users any more, and the base station (BS) can realize flexible multicast based on the users’channel conditions. Coding multicast provides a new way to solve the problem that the CM is limited by the LCG user.In addition, to solve the increasingly serious problem of spectrum scarcity, OFDM has been widely adopted in many wireless communication systems owing to its high spectrum efficiency. On the other hand, cognitive radio (CR) technology can sense the spectrum hole and allow the secondary users to exploit the existence opportunities of spectrum hole while guaranteeing the normal communication of the authorized users.This dissertation focuses on the resource allocation problem for the multicast in OFDM systems. Firstly, the OFDM, the cognitive radio and the OFDM-based cognitive radio system are introduced, then the multicast is provided and the resource allocation for multicast in OFDM systems is summarized.Furthermore, this dissertation investigates the resource allocation problem for the cluster-based cooperative multicast. In the clustering phase, users with good channel conditions are selected as cluster heads, while others decide to which cluster they belong. When the clusters are organized, it turns to a two-stage data transmission phase:in stage one, the BS transmits data to the cluster heads; in stage two, the cluster heads forward the received data to their cluster members. Simulation results show that the proposed scheme through cooperation improves the system performance.Then this dissertation investigates the resource allocation problem for the multiple description coding (MDC) multicast. It provides the model of MDC and proposes a two-step resource allocation algorithm: under the assumption of equal power allocation, subcarriers are firstly assigned; then combining with the convex optimization theory, the water filling algorithm is used to maximize the system throughput. Simulation results show that the proposed algorithm outperforms the CM. In addition, when the transmit power of cluster heads is large enough, performance of the cluster-based cooperative multicast is better than MDC multicast, but at the cost of consuming extra power.Finally, considering the practical systems, different MGs often have different traffic for reception. MGs with bad conditions may have quite a lot of data to receive, while others with good conditions may have fewer. Therefore, this dissertation will introduce "saturation rate" to set up the MDC multicast with non-full buffer traffic multicast model, and propose a low-complexity resource allocation algorithm based on dual decomposition. Simulation results show that the proposed algorithm significantly outperforms the CM, and it can get better balance in the resource allocation among users.