An efficient self‐diagnosis protocol for hierarchical wireless mesh networks

Abstract

With the development of wireless mesh networks (WMNs), fault diagnosis in WMNs is becoming a very challenging task. In this paper, a two-level scheme for fault diagnosis in WMNs is presented. We partition the network into a two-level topology architecture where level 1 is composed of mesh clients and level 2 consists of mesh routers. A new comparison approach is introduced to diagnose the two levels. On the basis of the new comparison approach, every node in WMNs can be diagnosed either as fault-free or faulty. Our protocol assumes that the WMN's topology may change during the testing phase and utilize the shortest path spanning tree, which is constructed along with the process of fault diagnosis, to disseminate local messages and global messages in WMNs. The proposed model is only for static fault circumstances. We provide the analysis of correctness, communication complexity, and time complexity of our protocol, and the comparison between our protocol and others through both theoretical proof and practical simulation. The analysis shows that our model has significant advantages over other existing models. unable to cover but does not require seamless voice and mobility capabilities of cellular systems. Consequently, designing dependable WMNs is at the heart of the research of wireless networks.An important problem in designing dependable WMNs that are subject to the failure of nodes is the distributed self-diagnosis problem. Fault nodes affect the reliable transmission of messages across the network. To support the high QoS, it is important for WMNs to be capable of automatically detecting and identifying faults of the network system. Fault diagnosis is one of the main building blocks of many dependable protocols. Actually, if one unit is diagnosed as faulty, the remaining units can identify it and eliminate it from the network or repair it. Preparata et al. first proposed the diagnostic model in wired networks, known as the Preparata-Metze-Chien (PMC) model [2]. This is followed by several comparison models including the broadcast comparison model [3] and the generalized comparison model [4]. However, these models were introduced for wired networks, and until now there are few papers that introduce fault diagnosis in wireless mesh networks. Most diagnostic protocols that have been proposed till now are appropriate for wired networks [5,6]. The seminal work of Chessa and Santi developed the comparison approach to the wireless environment [7]. They proposed a static distributed self-diagnosis protocol (static-DSDP) to complete the diagnosis problem. Chessa and Santi's model exploits the comparison approach, which is the promising practical diagnosis approach. In their approach, a node generates a task, and the task is assigned to a pair of different nodes whose outputs are compared. If the outcome of such a comparison test is 0, it indicates agreement. Otherwise, it indicates disagreement.Recently, Elhadef et al. extended Chessa and Santi's diagnostic model. They presented different models to complete fault diag...

Advances in parallel, distributed, embedded, and ubiquitous systems

“…With the development of Wireless Mesh Networks, fault diagnosis in Wireless Mesh Networks is becoming a challenging task. In the fourth paper, ‘An Efficient Self‐Diagnosis Protocol for Hierarchical Wireless Mesh Networks’ , Xu et al . propose a new protocol that has a comparison approach to diagnose two‐level topology architecture.…”