Resource Allocation And Physical Network Coding In Cooperative Relay Communication Systems

In LTE-B and future 5G networks, how to extend the coverage of a base station is always a hot topic for communication industry. Cooperative relay networks can effectively expand the coverage of wireless communication systems, improve the communication quality of wireless links. So it is considered to be one of the core technologies in 5G mobile communication systems. With the microcell or picocell deployment in 5G networks, cell size is shrinking. Because of smaller transmission power, flexible deployment and other features, relay is well suited for 5G networks. Therefore, cooperative relaying technology will play a pivotal role in the future 5G networks.Current relay research is mainly focussed on to optimize system capacity. The related researches on net-work architecture and resource allocation algorithm based on network energy efficiency has not been through. While considering to maximize the whole network energy efficiency, the individual user may have to sacrifice their own utilities. So we must take into account the network performance and user fairness at the same time. Multi-objective optimization and game theory are effective tools to solve selfish user collaboration problems. In this paper, we will carry out the researches on cooperative relay resource allocation algorithm and physical network coding under the premise of ensuring the QoS by using multi-objective optimization, game theory and other mathematical methods.For asymmetric cooperative relay network power allocation, the multi-objective evolutionary algorithm is introduced, and a multi-source multi-relay cooperative power allocation strategy is proposed. The multi-objective free search algorithm is proposed by extending free search algorithm to multi-objective optimization. The performance analysis and numerical experiments show that the accuracy and efficiency of our algorithm for multi-objective optimization problem. The objective function is to maximize the effective SNR for each source under the limited relay power. The utility functions and coding scheme are designed for multi-objective free search algorithm. So we can get the effective relay power distribution solution by MOFS. Simulation results verify the effectiveness of the strategy. Meanwhile, the proposed strategy can obtain the different power allocation solutions for the different application requires, by properly selecting Pareto optimal solution.It is a promising way that physical network coding theory can further improve the transmission capacity of cooperative relay networks. In this paper, compute and forward physical network coding strategy is studied based on Lattice coding in two way relay channel, and an efficient decoding coefficient vector search algorithm is proposed. According to the characteristics of compute and forward strategy, the relay decoding coefficient optimization problem is modeled as a quadratic integer programming model with quadratic constraints. A lifting and relaxation cutting plane algorithm is proposed to solve the quadratic integer programming problem. By lifting, convex relaxation, and generating the cutting plane, the original optimization problem is expressed into new easier linearly programming problem. And the optimal solution of the original problem is obtained by solving the new relaxation problem. Simulation results show that our proposed algorithm can effectively obtain the optimal relay decoding coefficient vector.Based on the two way relay channel compute and forward coding scheme, the in-depth study of the multiple access relay channel physical network coding scheme is carried out. From the perspective of multi-objective optimization, a relay selection and relay decoding coefficient vector search strategy is proposed for MARC compute and forward coding network. On the basis of free search algorithm, the exploration strategy is improved, and a mixed integer MOFS algorithm is proposed. The standard multi-objective traveling salesman problem is used to verify the effectiveness of the proposed algorithm. We investigate the case when relays excess sources, take all possible cooperative relays into account, and optimize overall network performance at system-level by using mixed integer MOFS. Under the premise of the coefficient matrix of taken relays satisfying full rank, the objective function is the maximization of the sum rate and the minimum computation rate. The proposed strategy completes the relay selection and the optimal coefficient vector search. Simulation results demonstrate the effectiveness of the proposed algorithm.Existing research on cooperative relaying network is mostly the case for asymmetric cooperative relay. For nodes selfishness characteristics in symmetrical cooperative relay network, a non-cooperative game the-ory bandwidth resource allocation algorithm is proposed. A symmetric cooperative relay game is established with the bidding mechanism. By designing an effective incentive mechanism, the selfish nodes can effectively participate in cooperative relay to improve the overall system performance, and all cooperative relays also can get a fair return. To determine the Nash equilibrium allocation solution, a distributed search algorithm is used, and the convergence condition is analyzed. Simulation results show that our presented game theory can be reasonably and effectively to solve the symmetric cooperative relay network bandwidth allocation problem.