To fully exploit the performance benefits of relay station （RS）, in the two-hop cellular networks covering hotspots, when the number of RSs is predetermined, both RS deployment and frequency reuse scheme are jointly...To fully exploit the performance benefits of relay station （RS）, in the two-hop cellular networks covering hotspots, when the number of RSs is predetermined, both RS deployment and frequency reuse scheme are jointly optimized for the purpose of maximizing the system capacity based on the constraints of system demand of capacity and the maximum number of outage demand nodes （MNDN）. Further, considering the overhead of increasing RSs, it is desired to use minimum number of RSs. The joint RS deployment and frequency reuse scheme （JRDFR） problem is formulated into a mixed integer nonlinear programming, which is non-deterministic polynomial-time hard in general. A heuristic approach based on genetic algorithm is proposed to tackle the JRDFR problem. The computational experiment of the heuristic approach is achieved and optimized RS deployment and frequency reuse scheme is obtained. Finally, we discuss the impacts of MNDN and the number of RSs on the system performance.展开更多
文摘To fully exploit the performance benefits of relay station （RS）, in the two-hop cellular networks covering hotspots, when the number of RSs is predetermined, both RS deployment and frequency reuse scheme are jointly optimized for the purpose of maximizing the system capacity based on the constraints of system demand of capacity and the maximum number of outage demand nodes （MNDN）. Further, considering the overhead of increasing RSs, it is desired to use minimum number of RSs. The joint RS deployment and frequency reuse scheme （JRDFR） problem is formulated into a mixed integer nonlinear programming, which is non-deterministic polynomial-time hard in general. A heuristic approach based on genetic algorithm is proposed to tackle the JRDFR problem. The computational experiment of the heuristic approach is achieved and optimized RS deployment and frequency reuse scheme is obtained. Finally, we discuss the impacts of MNDN and the number of RSs on the system performance.
