Resource Allocation Techniques In Cooperative Wireless Communication

Abstract

Wireless communication is becoming popular day by day as it provides mobility, flexibility and wide spectrum of applications. In wireless communication the demand of higher data rate is increasing day by day and fading channel becomes the bottleneck. MIMO systems is one of the solutions to improve the performance of the wireless systems but it is difficult to put more than one antenna and supporting RF sections in handheld devices from size, weight and power consumption perspective. Cooperative communication in wireless network is possible due to broadcast nature of the wireless channel. In cooperative communication one user’s antenna helps another user to repeat the transmission of the signals and thus forming the virtual MIMO like system. In this network one user act as relay which help another users to transmit their signals and provides benefit due to diversity combining at the destination. If more than one user is supported by the relay, then allocation the resources among them is one of the key problem. The issue of resource allocation in centrally controlled variable data rate cooperative network is considered in this thesis. The resources determine the data rate s of the users. Several criterions for enhancing the performance of cooperative network through appropriate resource allocation are presented. One way is to allocate equal resources to all users. In this case users which have good channel condition can have higher data rate and user which have poor channel can have data rate comparatively lower than first user. Other way is to distribute resources among user to maximize system data rate, in which the user with good channel can get favor to contribute in maximizing sum total data rate of the entire network. There are two perspective- efficiency and fairness which are very difficult to achieve simultaneously. We have presented a novel concept of utility for resource allocation in cooperative network which can optimize data rate and at the same time promise reasonable fairness. Different utility functions like log, power, iso-elastic and exponential are found suitable for efficiency-fairness trade-off. With the help of simulation results, we have demonstrated that it possible to achieve fairness in the range close to 0.99 with just 2% loss in the efficiency with the help of utility based resource allocation technique.