Distributed Medium Access and Opportunistic Scheduling for Ad-Hoc Networks: An Analysis of the Constant Access Time Problem

In this paper, we study the problem of medium access and distributed opportunistic scheduling to exploit channel fluctuations for wireless ad-hoc networks. This work focuses on the Constant Access Time (CAT) problem, where the total time duration of channel probing and data transmission is slotted into fixed block length. In particular, we consider the dependence between channel rates at different time instances during the channel probing phase and its impact on the overall system throughput. We first analyze the system performance of the CAT problem under independent channel rate assumption and compare our result to the existing work on the Constant Data Time (CDT) problem. We then propose a protocol to reduce the channel probing costs based on this analysis. We show analytical results of the proposed protocol for systems with a sufficiently large number of users. We compare the system performance of the proposed protocol with that of the original protocol under independent channel rate assumption. We mathematically prove that the proposed protocol improves the system performance.