Abstract
Conference Title: 2014 27th Biennial Symposium on Communications (QBSC) Conference Start Date: 2014, June 1 Conference End Date: 2014, June 4 Conference Location: Kingston, ON, Canada Balancing between fairness and throughput remains a challenging problem in the design of future wireless networks. This paper addresses this challenge and proposes a practical power optimization method, which has the potential of balancing between fairness and data throughput. The paper assumes a non-line of sight (NLOS) wireless backhaul network, especially introduced to serve areas with dense data traffic. The network comprises several transmitters, known as access-nodes (AN), each serving several remote terminals (RT) via wireless links. To reach a feasible, yet a simple, proportional fairness solution, this paper maximizes an approximated network-wide utility function, where the individual terms are based on long-term channel gains, utilized to compute the geometry carrier to interference and noise ratio (CINR). The paper then solves the power optimization problem by introducing a scalar bias to balance fairness with throughput. Through Monte Carlo simulations, performance improvement of the proposed method is illustrated and compared against existing classical methods. [PUBLICATION ABSTRACT]