Flexible and efficient aggregation framework for antifragile wireless mesh networks

Abstract

Wireless mesh networks (WMNs) have found an interesting alternative application in public safety and disaster recovery as they are enabled with key features such as fault tolerance, broadband support, and interoperability. However, their sparsely distributed wireless nodes need to frequently share the control packets among each other for successful data transfer. These packets give rise to a considerable amount of control overhead, especially for multimedia traffic, which is not bearable in jeopardy situations of network disaster such as 9/11 and Hurricane Katrina. Hence, to avoid the huge cost of control overhead, aggregation is supposed to be one of the handy solutions for building a new object from one or more existing objects of network traffic. Generally, aggregation has three types to be executed on ‘packets’, ‘frames’ and the ‘links’ of a network. The network decision that when and which type of aggregation is suitable in a given scenario is a complex balancing act. Because, it is bounded by various live statistics of the communication link (e.g., buffer size, link quality, bandwidth, maximum transmission unit, and delay etc.) and thus, can considerably affect the network efficiency. If such statistics do not support aggregation or its certain type, network performance may cause worse affects. This paper proposes a decision-oriented dynamic solution, namely AADM, which is an adaptive aggregation-based decision model. Based on the link quality and its live statistics, AADM dynamically takes judicious decisions about aggregation and its specific type to achieve the desired outcome. Despite network scalability, quality of service, power optimization, and network efficiency, it also reduces control traffic in WMNs. OMNET++ simulations are used to verify AADM. Simulation results have shown that AADM outperforms existing static approaches in terms of packet loss, throughput and delay.