Path Mechanism to reduce packet data loss in Wireless Mesh Networks

A wireless mesh network (WMN) is a communication network made up of radio nodes organized in a mesh topology. Wireless mesh network often consists of mesh clients, mesh routers and gateways. A wireless Mesh network uses multi-hop communication. Due to multi-hop architecture and wireless nature, Mesh networks are vulnerable to various types of Denial of Services attack. It suffers from Packet dropping at Routing layer. Client nodes are unable to get services from gateway nodes, hence network gets down. The Paper emphasis on the developing of a path protocol when the minimun possible packet dropp occurs in wireless mesh networks. Due to packet droping occurrences the network performance degrades. In the work, we have evaluated the Performance of WMN under packet dropping on the basis of their throughput and Data packet loss.


Introduction
The proliferation of laptop computers and other mobile devices (PDAs and cell phones) created an obvious application level demand for wireless local area networking. In recent years, a wide variety of mobile computing devices has emerged, including portables, palmtops, and personal digital assistants. At present, most 802.11 Wireless Local Area Networks (WLANs) operate in the infrastructure Basic Service Set (BSS) mode. In it, all stations communicate via single wireless hop with a central entity denoted as Access Point (AP). An AP collocated with portal bridges the 802.11 with a non-802.11 network. Objective of 802.11 is to create the wireless local area network forming BSS (basic service set), BSS's together form the ESS (extended service set). In BSS, user can be mobile user instead of fixed which is the basic objective of 802.11. Also, users can have transition from one BSS to another BSS without the interruption of services called roaming. 802.11 WLAN can be used to provide services at airports, universities, home networking, office networking and outdoor wireless and much more, where the wired infrastructure is not desirable.
To become independent of backbone networks leading to cheap deployments, the traditional single-hop approach needs to be replaced by Wireless Mesh Networks (WMNs). Mesh BSS provides connectivity over multiple wireless hops. Path selection and forwarding operate transparently within the MAC.
WMNs are undergoing rapid commercialization in many application scenarios such as Broadband home networking, Community networking, Building automation, High-speed MAN, Intelligent transport system networks, Enterprise networking and much more.
Wireless mesh networks are becoming popular among Internet Service Providers due to cheap deployment and self healing properties.

LITERATURE REVIEW
A. Pirzada et al, 2004 described a model of building trust relationship between nodes in an ad hoc network. The nodes passively monitor the packets received and forwarded by other nodes and compute the trust values for their neighbours. The trust values are used for computing the trustworthiness of links. For routing, links with high trust values are chosen so as to avoid the malicious and selfish nodes [7].
A. Patwardhan et al, 2006 has proposed a trust-based data management scheme in which mobile nodes access distributed information, storage, and sensory resources available in pervasive computing environment. The authors have taken a holistic approach that considers data, trust, security, and privacy and utilizes a collaborative mechanism that provides trustworthy data management platform in an ad hoc network for secure authentication, data communication, data access and certificate and key management [8].
B-J.Chang at al, 2008 has proposed a trust-based scheme for multicast communication in a MANET. In a multicast MANET, a sender node sends packets to several receiving nodes in a multicast session. Since the membership in a multicast group in a MANET changes frequently, the issues of supporting secure authentication and authorization in a multicast MANET are very critical. The proposed scheme involves a two-step secure authentication method. First, an ergodic continuous Markov chain is used to determine the trust value of each one-hop neighbor. Second, a node with the highest trust value is selected as the certificate authority (CA) server. For the sake of reliability, the node with the second highest trust value is selected as the backup CA server. The analytical trust value of each mobile node is found to be very close to that observed in the simulation under various w w w i j c t o n l i n e . c o m scenarios. The speed of the convergence of the analytical trust value shows that the analytical results are independent of the initial values and the trust classes [9].
G. Vigna et al, 2004 proposed an approach to detect intrusions in AODV that works by stateful signature-based analysis of the observed traffic. Sensors are placed on selected nodes for promiscuous sensing of radio channels. Each sensor has database of attack signatures and looks for a signature match in the traffic. A match triggers a response, usually an alert [10].
H. Yang et al, 2006 have proposed the SCAN protocol that addresses two issues simultaneously; (i) routing (control packets) misbehavior, and (ii) forwarding (data packets) misbehavior . Each node monitors its neighbors independently and the nodes in a neighborhood collaborate with each other through a distributed consensus protocol [11].
J. S. Baras et al, 2005 proposed a trust management scheme for self-organized ad hoc networks, where the nodes share trust information only with their neighbors . For establishing and maintaining trust among the neighbors authors have proposed a voting mechanism [12].
Jay dip, 2011 has proposed mechanism relies on local observation of each node in a WMN. Based on the local information in each node and using a finite state machine model of the AODV protocol, a robust statistical theory of estimation is applied to identify selfish nodes in the network. Using statistical estimation technique, analysis of variance and some additional fields in the headers of the AODV packets, this protocol is able to achieve a higher detection rate with a very low rate of false positives [13]. Jaydip Sen et al, 2011 proposed a self-organized trust establishment scheme for nodes in a large-scale MANET in which a trust initiator is introduced during the network bootstrapping phase . It has been proven theoretically and shown by simulation that the new nodes joining the network have high probability of successful authentication even when a large proportion of the existing nodes leave the network at any instant of time. A distributed detection mechanism of malicious packet dropping attack in MANETs has been proposed in , where local anomaly detection is utilized to make a more accurate network-wide (i.e. global) detection using a cooperative detection algorithm [14].
L.Santhanam et al, 2006 presented a mechanism to judge a node's behavior based on observed traffic reports submitted to local sink agents, dispersed throughout the network .The sink nodes apply a set of forwarding rules to isolate a selfish node based on the number of times it is caught in selfish acts. The scheme is independent of the routing protocol or network architecture, and is suitable for multi-channel wireless mesh network [15].
M. Conti et al, 2006 has proposed a scheme in which a node exploits its local knowledge to estimate the reliability of a path. Unlike the conventional method of denying selfish users, it provides a degraded service to these nodes by selective slow packet forwarding [16]. N.A Benjamin et al, 2010 proposed that WMNs can be used to transmit vital information arising from the wireless body sensor network (WBSN) to a backbone network. The integration of WBSN and WMN technologies results in wireless sensor mesh network (WSMN) and this type of network can be utilized for remote health monitoring of patients. The battery-powered, memory-constrained sensors transmit the sensed information to their nearest mesh nodes and the mesh nodes, in turn, use multi-hop routing to transmit the information to the backbone network devices like PDA or the servers for health monitoring applications. The authors have investigated performance of such a WSMN for patient health monitoring applications, in terms of parameters like delay, and throughput under varying number of patients and doctors [17].
R. Mahajan et al, 2005 illustrated a mechanism named CATCH, which consists of two modules: (i) anonymous challenge message (ACM), and (ii) anonymous neighbor verification (ANV). In the security scheme, first an ACM message from an unknown sender is sent to all its neighbors. As the sender is unknown, all the nodes further broadcast the ACM message.In the ANV phase, a tester node sends cryptographic hash of a random token for rebroadcast and also records other hashes sent by other nodes. The tester node releases the secret token to another node which successfully authenticates itself [18].

Conclusion Future Work
Client nodes are unable to get services from gateway nodes, hence network gets down. The Paper emphasis on the developing of a path protocol when the minimun possible packet dropp occurs in wireless mesh networks. Due to packet droping occurrences the network performance degrades. In the work, we have evaluated the Performance of WMN under packet dropping on the basis of their throughput and Data packet loss.
In the future directions, this work can be extended by using hundreds of nodes and we need to develop the Intrusion detection System (IDS) that also chooses the monitor by considering battery life parameter. It is important to consider congestion conditions of the nodes using information obtained from other layers before determining the nodes to be malicious. Also, detecting intrusions at different layers increases the information about the malicious nodes thus identifying these nodes more accurately.