Performance of Wireless Mesh Networks with Centralized Multihop Scheduling for Intelligent Transportation Systems

We consider the problem of centralized multihop scheduling in wireless mesh networks for solutions to realize intelligent transportation systems. A scheduling tree construction strategy is investigated, which could use any predetermined selection metrics, while avoiding primary and secondary interferences to a great extent. Two centralized multihop scheduling algorithms are presented to schedule nodes on the constructed scheduling tree. In contrast with FIFO algorithm, the fair algorithm has the property that it guarantees fairness among downlink relay queues of each relay node, while making an efficient utilization of free mini-slots. The feasibility of scheduling tree construction strategy with both minimal hop count and maximal signal strength metrics for a mesh cell is studied through experiment. Simulation results of network throughput and average active link number show that the Fair algorithm utilizes network resources more efficiently than the FIFO algorithm

I. Introduction

The mesh-type fixed broadband wireless access system defined by the IEEE 802.16–2004 standard [1], is an attractive way of providing proprietary commercial solutions. It serves as an access network that employs multihop wireless forwarding by fixed nodes to relay traffic to and from the wired backbone Internet, which is the typical application scenario of intelligent transportation systems (ITS). Although some research results show that the network capacity can be increased by deploying relay nodes [2], [3], scheduling is the principal effect of the introduction of mesh concept in a fixed broadband wireless access network. The scheduling problem in traditional cellular networks now becomes one of routing and scheduling, i.e., first choosing what seqnence of relays packets will go through before reaching the destination node, and then choosing the relay nodes whose packets will be transmitted next.