A Fair Scheduling using Spectrum Load Smoothing Algorithm for Mesh Networks

Fair distributed scheduling is always a big challenge issue in wireless mesh networks. In this work we focus on the IEEE 802.11s proposal by the Mesh Network Alliance (MNA) which aims to enhance the legacy 802.11 medium access protocol to enable efficient mesh operation. We propose an Enhanced Distributed Reservation Protocol (EDRP) to achieve fair scheduling in wireless mesh networks. Our protocol calculates fair shares for all links in the local network in a distributed manner. Each Mesh Point (MP) in the network learns the traffic requirements of all MPs in the local network by analyzing received beacons from its neighbors. With the usage of the acquired knowledge of all neighboring links, MPs are able to calculate the shares of the resources that they should occupy by performing Spectrum Load Smoothing (SLS) Algorithm. We implement this fair scheduling schemes in the WARP2 simulator and compare its performance with which using the Distributed Reservation Protocol (DRP), that executes medium occupation with First-Come-First-Served (FCFS) scheduling discipline. The simulation results show that our solution is fair and collision-free.