Aggregate throughput maximization in a hierarchical cooperation scheme under consideration of packet arrival rate, control and data overhead

The conventional hierarchical cooperation (HC) scheme proposed by Ozgur et al. [1] is a novel communication scheme that can achieve linear capacity scaling, O(N) in wireless ad-hoc network from the viewpoint of information theory without considering control and data overhead. In this paper, we investigate the effects of practical parameters such as packet arrival rate, control overhead and data overhead on the HC scheme. We formulate an aggregate throughput optimization problem in terms of cluster size m and solve it by well-known algorithms. Through the performance evaluations, we observe three main effects of the practical parameters. First, large data overhead ratio degrades the aggregate throughput. 31% performance degradation occurs when we consider data overhead ratio α = 0.45. Second, large control overhead also degrades the aggregate throughput and affects to determine the cluster size m for maximum aggregate throughput. Third, the packet arrival rate λ plays a key factor for system stability and we obtain a sufficient condition for system stability.