A new dynamic hierarchical reputation evaluation scheme for hybrid wireless mesh networks☆
Graphical abstract
Introduction
Wireless mesh networks have become the focus of research in recent years because of their low maintenance overhead and high data rates [1], [2], [3]. Wireless mesh networks are multi-hop, decentralized and self-organizing; thus, they show great promise for numerous next-generation wireless services. The network structure of wireless mesh networks differs from that of traditional wireless networks for some applications [4], [5], [6], such as mobile ad hoc networks [7], [8]. In general, wireless mesh networks consist of mesh gateways, mesh routers and mesh clients. However, wireless mesh networks do not necessarily include all of the above types of nodes. The node types required for the network depend on the specific configuration of the related networks. There are three basic network structures with different node functions for wireless mesh networks, flat network structure, hierarchical network structure and hybrid network structure [9], [10], [11], [12], [13], [14].
A flat wireless mesh network is one type that is composed of mesh clients with wireless routing functions. It provides direct point-to-point service to the customer. Using flat mesh networks, all client functions are implemented on the same level. Therefore, additional backbone access is not necessary for this type of network. That is, flat mesh networks are similar to ad hoc networks and can provide effective communication support, with the difficulty or without to leverage current networks. They have the same disadvantages as ad hoc networks, such as difficulty accessing the core networks.
Hierarchical wireless mesh networks are more common in daily applications. Generally, they consist of mesh routers used as backbone networks with the ability to self-configure and self-heal. The mesh clients in hierarchical mesh networks have no inherent routing function and can only access the backbone network through the mesh routers. The mesh routers access the wireless service for the clients by connecting to gateway nodes and the Internet. The structure of hierarchical mesh networks can be analyzed in two layers, the client layer and the backbone layer. Using this hierarchical structure, building cost is greatly reduced, while both reliability and the network coverage rate are increased. However, a disadvantage of hierarchical wireless mesh networks is the inability of direct communication between any two nodes.
Hybrid wireless mesh networks are synthesized from the above two mesh network structures. Thus, hybrid wireless mesh networks have some characteristics of each of the two types of networks. The typical hybrid mesh structure is the hierarchical structure, as shown in Fig. 1. In the lowest layer, the mesh clients have forwarding functions to establish direct communication between nodes, similar to flat wireless mesh networks. However, if mesh clients want to access the core networks, they must go through mesh routers and gateways, which compose the upper layer. Therefore, multi-hop wireless communication is implemented more easily with the hybrid wireless mesh network structure. The hybrid mesh networks are not only a type of merged WLAN and ad hoc network but also act as a wireless version of the Internet. Given the features discussed above, we expect applications for hybrid wireless mesh networks to be prolific.
Wireless mesh networks are vulnerable because of some of their natural characteristics, such as open media and dynamic topology. The network environment will be further deteriorated in hybrid wireless mesh networks that have a large number of mobile mesh clients. Even one malicious node in a hybrid wireless mesh network could attack many mesh routers and clients, seriously affecting the network security. Thus, we must propose an effective detection scheme for hybrid wireless mesh networks, with a particular focus on the participating mesh clients.
As a security sub-problem in wireless mesh networks, the reputation evaluation mechanism has been widely studied in recent years [15], [16], [17], [18], [19]. Compared with the static reputation model, the dynamic reputation model, which is evaluated based on the node’s experience, may be more suitable for the security requirement in wireless mesh networks because of the complex environment and the diversified and unpredictable security problems found in wireless mesh networks. Nodes often have components such as watchdogs [17] to avoid defective nodes while forwarding messages, which buffer all packets before transmission and then monitor its neighbors’ transmissions to verify that they are forwarded. Simulations show that application of watchdogs can improve throughput by 17–27%. Work by Liu et al. [18] has mapped this problem by combining familiarity values with subjective opinions. The familiarity value represents a node’s degree of familiarity with another individual node and is used to calculate the weighting factor that determines how much a node’s recommending opinion impacts the reputation computation result. Catania et al. [19] used network coding to build a practical reputation system to improve network performance in wireless mesh networks. The reputation system rewards cooperative behavior in routing and packet forwarding and penalizes non-cooperative behavior.
Nevertheless, there are few approaches that deal with reputation management specifically for hybrid wireless mesh networks. Most of the current reputation evaluation schemes only consider a single node’s behavior and assume every node plays an equal role. Node cooperation and hierarchical construction have not been considered. Therefore, these traditional reputation schemes cannot be applied directly to hybrid wireless mesh networks.
Considering the characteristic security requirements for hybrid wireless mesh networks, we divide the network into virtual cluster structures. Based on the virtual cluster structures, we propose a dynamic hierarchical reputation evaluation scheme (DHRES) to improve the security performance of hybrid wireless mesh networks. The scheme builds a reputation relation network based on the consideration of node behavior and cooperation, increasing the accuracy of the reputation evaluation. The different reputation calculation methods are included, according to the roles of the nodes (i.e., cluster head (CH), cluster member (CM) and gateway (GW)) and their relations. The cooperation between nodes is enhanced and network security is improved accordingly. Simulation results show that compared with traditional reputation evaluation models, DHRES can reflect the security status more rapidly and accurately, which can effectively improve the routing performance under attacks in hybrid wireless mesh networks.
The remainder of this paper is organized as follows. Section 2 describes the DHRES system of the reputation evaluation scheme in detail. Section 3 shows the simulation results and analysis of the reputation evaluation performance. Section 4 presents the conclusions of the study.
Section snippets
Dynamic hierarchical reputation evaluation scheme
In hybrid wireless mesh networks, mesh routers and gateways are the backbone nodes for routing. The mesh router forwards packets either to gateways or mesh clients; the gateway is an intermediate node in the communication path between mesh networks and the Internet. Although the mesh client is not precisely in the route between clusters, if it misbehaves, the effectiveness of the networks can be affected. Therefore, cooperation among nodes in hybrid wireless mesh networks is important. To
Simulation results and analysis
In this study, we performed our experiments with MS VC++ 6.0 software and the NS-2 simulator. In the simulation, various network scenarios were designed and analyzed to verify the performance and characteristics of the reputation evaluation scheme.
Conclusions
As a new wireless networking paradigm, the hybrid wireless mesh network is flexible and combines the benefits of both infrastructure and mesh clients. However, security is a major challenge in this type of network because its natural characteristics leave it vulnerable to attacks. Reputation evaluation is one of the most common and effective approaches to help assess and solve these security concerns. However, traditional reputation schemes are not suitable for direct application in hybrid
Acknowledgements
This work was supported in part by the Fundamental Research Funds for the Central Universities (N120404002, N100304009, N110204001, N120804002), the Specialized Research Fund for the Doctoral Program of Higher Education (20110042120035, 20110042110023), the National Natural Science Foundation of China (61172051), and the Program for New Century Excellent Talents in University (NCET-12-0102).
Yao Yu received the Ph.D. degree from communication and information systems in Northeastern University, Shenyang, China, in 2010. She is currently a Lecture in the same university. She has authored over 20 papers in journals and international conferences. Her research interests include routing and security in wireless networks.
References (22)
- et al.
Optimization model for handoff-aware channel assignment problem for multi-radio wireless mesh networks
Comput Networks
(2012) LSSP: a novel local segment-shared protection for multi-domain optical mesh networks
Comput Commun
(2007)- et al.
RBC-OLSR: reputation-based clustering OLSR protocol for wireless ad hoc networks
Comput Commun
(2012) - et al.
Green survivability in Fiber-Wireless (FiWi) broadband access network
Opt Fiber Technol
(2012) - et al.
Auxiliary graph based protection for survivable Fiber-Wireless (FiWi) access network considering different levels of failures
Opt Fiber Technol
(2012) - et al.
Role based privacy-aware secure routing in WMNs
TrustCom
(2012) - et al.
A hierarchical architecture for detecting selfish behaviour in community wireless mesh networks
Comput Commun
(2011) - et al.
A novel reputation computation model based on subjective logic for mobile ad hoc metworks
Future Generat Comput Syst
(2011) - et al.
Routing security scheme based on reputation evaluation in hierarchical ad hoc networks
Comput Networks
(2010) - et al.
A cooperative channel assignment protocol for multi-channel multi-rate wireless mesh networks
Ad Hoc Networks
(2011)
Optimizing backup optical-network-units selection and backup fibers deployment in survivable hybrid wireless-optical broadband access networks
J Lightwave Technol
Cited by (13)
Enhancing Security and Trust in Internet of Things through Meshtastic Protocol Utilising Low-Range Technology
2024, Electronics (Switzerland)Security in IoT Mesh Networks Based on Trust Similarity
2022, IEEE AccessSecure Trust-Based Group Key Generation Algorithm for Heterogeneous Mobile Wireless Sensor Networks
2021, Lecture Notes in Networks and SystemsA vulnerability analysis method for critical elements based on network connectivity
2020, Kongzhi yu Juece/Control and DecisionSurvey on computational trust and reputation models
2018, ACM Computing Surveys
Yao Yu received the Ph.D. degree from communication and information systems in Northeastern University, Shenyang, China, in 2010. She is currently a Lecture in the same university. She has authored over 20 papers in journals and international conferences. Her research interests include routing and security in wireless networks.
Yuhuai Peng received the Ph.D. degree from communication and information systems in Northeastern University, Shenyang, China, in 2013. He is currently a Postdoctoral Fellow in the same university. His research interests include routing and reliability in wireless mesh networks. He has published over 20 technical papers in the above areas.
Yinpeng Yu received the M.S. degree in communication and information systems at Northeastern University, Shenyang, China, in 2012. He is currently a Ph.D. student in the same university His research interests include power optical communication network and network coding in wireless mesh networks.
Tianyu Rao received the B.S.E degree in communication engineering at Northeastern University, Shenyang, China, in 2012. He is currently working at an IT company.
- ☆
Reviews processed and recommended for publication to Editor-in-Chief by Deputy Editor Dr. Mehdi Shadaram.