This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Providing proper secure communications is challenging in Mobile ad hoc networks because of unreliable wireless media, mobility of the nodes and lack of infrastructure. Usually, wired and wireless networks use cryptographic techniques for secure communications. Symmetric and asymmetric cryptography have been extensively used in Adhoc networks and have both advantages and disadvantages. Any cryptography becomes weak if it has a weak key management which also forms the main aspect for security in MANETs. Thus, the network is susceptible to attacks by malicious nodes and packets are dropped in attacks like grayhole attack. In this paper, it is proposed to evaluate the performance of a network under the impact of malicious nodes. Grayhole attack was simulated as it is difficult to identify them in the network and their behaviour is also highly unpredictable. Simulations are conducted using DSR routing protocol to evaluate the performance degradation of MANET due to malicious node activity.
Keywords: Mobile ad hoc networks (MANETs), Dynamic Source Routing (DSR), Malicious Nodes, and Performance Degradation.
A Mobile ad hoc network (MANET) consists of wireless mobile nodes that communicate with each other without network infrastructure/centralized administration. Mobile hosts are free from any centralized control like base stations/mobile switching centres. Though providing unrestricted mobility and connectivity to users, the onus of network management is entirely on network nodes . Due to wireless network interfaces limited transmission range, multiple hops are required to exchange data with another across the network. In such networks, each node operates both as host and router, forwarding packets to other network nodes not within wireless transmission range of each other. All nodes participate in ad hoc routing protocols enabling discovery of multi hop paths to other network nodes. MANETs are also called infrastructure-less networking, as mobile nodes establish routing among themselves to form networks on the fly. The latter are formed instantaneously using multi-hop routing for information transformation. MANET technology provides flexible method to establish communication where geographical/terrestrial constraints need a distributed network without a fixed base station like battlefields, military applications, and emergency/disaster situations. A sensor network of numerous small low-powered nodes with sensing capabilities is one of MANET's applications.
Research reveals that wireless MANET has a bigger security problem when compared to traditional wired and wireless networks [2, 3], though most features make MANETs popular. To begin with in MANETs all signals are routed through bandwidth-constrained wireless links making is liable to many security threats as compared to fixed landline networks and include passive eavesdropping to active interference. Improperly protected mobile nodes can be captured, compromised, and hijacked. Also an attacker can listen in and modify traffic on wireless communication channel. There are chances that attempts might be made to masquerade as a participant. Authentication - based on public key cryptography and certification authorities - could be hard to accomplish in MANETs because of the lack of infrastructure.
Secondly, as nodes roam freely in any direction security solutions with static configuration are in adequate for a dynamically changing topology. In MANET routing protocols, nodes exchange information about network topology to ensure establishment of routes between sources and destinations. As messages are transmitted over the air an intruder can maliciously update information incorrectly by pretending to be legitimate. An instance is denial of service (DoS) being launched when a network is flooded with counterfeit routing messages by malicious nodes. Such message could be forwarded by other innocent nodes.
Third, decentralized decision making in the MANET relies on the cooperative participation of all nodes. The malicious node could simply block or modifies the traffic traversing it by refusing cooperation to break the cooperative algorithms. This property makes some centralized intrusion detection schemes fail.
Finally, some/all MANET nodes rely on batteries for their energy. A new type of DoS attack can be created by forcing a node to replay packets to exhaust energy. Due to nodes limited network capacity and battery power, disconnections are frequent in MANETs making identification of anomalies harder.
Generally, wireless MANETs are vulnerable because of their fundamental characteristics which include open medium, dynamic topology, absence of central authority, distributed cooperation and constrained capability. Present security solutions meant for wired networks are inapplicable for wireless MANETs.
Network attacks can be either External attacks or internal attacks. In external attack, the attacker aims to cause congestion, spread false routing information or disturb nodes from providing services. External attacks are attacks launched by challengers who are not initially official to participate in the network operations. These attacks usually aim to cause network congestion, deny access to specific network function or to interrupt the whole network operations. External attackers are mainly outside the networks who want to get access to the network, once they get access to the network they start sending fake packets, denial of service in order to interrupt the performance of the whole network.
In internal attack the attacker wants to have usual access to the network as well as contribute in the normal activities of the network. Internal attack is more strict attacks then external attacks. Internal attacks, in which the challenger wants to gain the normal access to the network and contribute the network activities, either by some malicious impersonation to get the access to the network as a new node, or by directly compromising a existing node and using it as a basis to conduct its malicious behaviours. Internal nodes might misbehave to save their limited resources, such as the battery powers, the processing capabilities, and the communication bandwidth. Attacks that are caused by the misbehaving internal nodes are hard to detect because to distinguish between normal network failures and misbehaviour activities in the ad hoc networks is not an easy task.
Black hole attack is one of the security threat in which the traffic is redirect to such a node that actually does not exist in the network. In these attacks, black hole attack is that kind of attack which occurs in Mobile Ad-Hoc networks (MANET). In black hole attack, a malicious node uses its routing protocol in order to endorse itself for having the shortest path to the destination node or to the packet it wants to interrupt. This destructive node advertises its availability of new routes irrespective of checking its routing table. In this way attacker node will always have the availability in replying to the route request and thus intercept the data packet and retain it. Grey hole is a node that can
switch from behaving acceptably to behaving like a black hole.
It has to be understood that security implies identification of potential attacks, threats and vulnerabilities in a system. Karpijoki  and Lundberg  discussed selected types of attacks possible against a MANET and they could be classified as passive and active. The former does not disrupt routing protocol operation and only tries to discover information listening to routing traffic and hence is hard to detect. An active attack tries to modify data, gain authentication, or procure authorization through insertion of false packets into data streams/modifying packets transition through networks. Active attack are further categorised into external and internal attacks. The former is caused by nodes strange to a network while the latter is from compromised/hijacked nodes within the network.
Key management is most crucial one among security issues in MANETs, because it is the assumption of many security services. Secure routing protocols like ARAN  and SRP , assume that private and public keys and a Trusted Third Party (TTP) signed certificate are assigned to nodes. Research work currently in key management [7, 8] is capable of handling only limited nodes. When this number increases, most are either inefficient or insecure. Also since MANET has no clearly defined lines of defence, nodes cannot be classified - based on risks - in advance due to MANETs dynamic property. Hence flexibility and adaptability should be considered when planning a key management scheme for MANETs. A major difference between MANETs and wired networks is that in the former, nodes have limited power supplies making redundant any protocol requiring high computation.
Though several security schemes for MANETs were proposed, MANET's security research is still in its infancy. Transmitting routing information in distributed key management services  is through a redundant way so that when a route fails or when limited nodes are compromised, it does not affect the network much. Share refreshing is used to frustrate attacks attempting to discover the certificate authority's secret key within limited time as it is felt that the shared signature of private key of key management services should not be disclosed to an adversary
To counter threats MANETs use mechanisms like IP Security (IPsec), to ensure security for transmitted data. But before using IPsec nodes should form Security Associations (SAs). During this process, two nodes authenticate one another using certificates, a primary way to verify identities. Key Management Systems create, distribute, and manage such certificates and hence it is the heart of a network's defences.
In this paper, it is proposed to evaluate the performance of a network under the impact of malicious nodes. Simulations are conducted using DSR to evaluate the performance degradation of MANET due to malicious node activity. Section 1 dealt with the basics of the wireless network, section 2 reviews some previous works available in the literature. Section 3 details the methods used for evaluation, section 4 gives the simulation result and discusses the same. Section 5 concludes the paper.
2. Related Works
To form impulsively huge network in MANET using mobile nodes, which is bigger than that of the radio range where the routing supports the communication among each other. Adjih, et al.,  investigated the issues related to security of MANET and proposed an architecture comprising multiple securing mechanisms is described in detail. OLSR, one of the routing protocols for these types of MANET networks is the main focus in this paper. The proposed architecture mitigates the attacks. Information regarding algorithms, protocols, methods and accomplishment information are provided.
In the current information technology mainly in wireless and mobile environments such as MANETs, key management plays key role in the security. The dynamic nature of network leads to more concentration on key management as its implementation is very complicated. Based on PKI and identity-based public key cryptography (ID-PKC) are the classical key management approaches that experiences the key escrow problem and more cost for computation for certificate verification. Lu Li et al.,  introduced a new distributed key management approach, which is a combination of certificateless public key cryptography (CL-PKC) and threshold cryptography that retards the single point of failure in addition to the requirement of certificate-based public key distribution and the key escrow problem.
Hadjichristofi et al.,  proposed a new framework in MANETs for key management which offers robustness and redundancy for the purpose of Security Association (SA) establishment among pairs of nodes. A modified hierarchical trust Public Key Infrastructure (PKI) model is utilized in the proposed KMS where the management roles are assumed by nodes dynamically. The advantages of using the proposed KMS based on the network environment is it maximizes service availability for every nodes, maximizes the accommodating of novel nodes flexibility, reduces pre-configuration, and is able to reconfigure itself dynamically.
Bo Zhu et al.,  proposed a new hierarchical approach based on threshold cryptography to deal the issue of key management and certification service in MANET considering both security and effectiveness. The contributions of the proposed key management approach comprises: 1) the flexibility to select suitable security configurations in relation to the risks faced is afforded to different parts of MANET, 2) for rapidly-modifying environments, the adaptivity is offered to cope with, 3) MANETs consisting many nodes are handled, 4) with various levels of assurance the certificates are issued. In ad hoc networks to protect certification services from active attacks two algorithms are additionally proposed that can be employed autonomously from the hierarchical structure. The results obtained by simulating reveals that the 1024 bits key length is around six to eight times faster in the process of renewing or generating a certificate and around 20-80 times faster in the partial certificates generation process. The results obtained by simulating also demonstrate that in a hostile environment where the present methods show weak performance but the proposed two algorithms achieves best performance.
Bing Wu et al.,  proposed a secure and efficient key management (SEKM) framework. Using a secret sharing scheme and using an underlying multi-cast server groups SEKM builds a public key infrastructure (PKI). Detailed theoretical information in relation to developing and maintaining the server groups is provided. The entire server group forms a view regarding the certificate authority (CA) and offers certificate update service for every node comprising the servers themselves in SEKM. For the purpose of effective certificate service a ticket scheme is introduced. The proposal introduces an additional efficient server group updating scheme.
3. Experimental Setup and Results
Experiments were conducted with 30 mobile nodes, spread over an area of 2 km by 2 km. The nodes communicate over UPD/IP network. The data rate is uniformly maintained at 11 Mbps for all nodes. The transmission power of 0.005 watts and reception power threshold set at -95dBm is maintained. Simulations are conducted for 5 minutes. Figure 1 shows the scenario without gray hole attack with each node sending data randomly to other nodes in the network with an exponential packet inter arrival time. The size of the packets sent is set at exponential (1024).
Figure 1: The network scenario without any malicious nodes.
In the second scenario five nodes were simulated to perform grayhole attack. The location of the nodes is shown in figure 2.
Figure 2: The network with malicious nodes.
The malicious nodes are located at the centre of the network for maximum damage. The malicious nodes are designed to randomly drop packets irrespective of the source or destination address. Experiments were conducted to simulate the Dynamic source protocol (DSR) with all the nodes cooperating and with 15% (5 nodes) of the nodes being malicious. The attack simulated is grayhole attack. Figure 3 to Figure 5 show the network performance in terms of acknowledgment received, Utilization of routes from cache and throughput respectively. All the outputs plotted are in time average format. Figure 3 shows the total percentage of acknowledgements received within the network for successful data sent.
Figure 3: Successful acknowledgements received.
It is seen from figure 3 that acknowledgments received when all nodes are cooperating in a network is considerably more when compared to network with malicious nodes. Acknowledgment received is less in network with malicious nodes as packets are dropped by the malicious nodes during the grayhole attack. Figure 4 shows the average cache utilization for discovering routes.
Figure 4: Utilization of routes from cache
From figure 4 it can be seen that routes become stale faster in network consisting of malicious nodes. This is due to the constant packet dropping leading to new route discovery making the route cache stale. Figure 5 shows the throughput of the system.
Figure 5: Throughput in bits/sec for cooperating and malicious network
Due to the presence of 15% malicious nodes the throughput which is the average rate of successful message delivery over the wireless medium of the network decreases by 54.76%. The network degradation is extremely high leading to poor Quality of Service and insecure transmission of data.
MANETs are susceptible to attacks by malicious nodes and packets are dropped in attacks from malicious nodes. In this paper it was proposed to evaluate the performance of a Ad hoc network under the impact of malicious nodes. Simulations were conducted using DSR routing protocol to study the performance degradation of network due grayhole attacks. Simulation was conducted using 30 nodes with 15% of the nodes being malicious and network without any malicious nodes. Simulation results show that the acknowledgments sent and the various types of route replies are decreased in a network under grayhole attack due to the packet drops. Further investigations to identify malicious node and ways to mitigate them are critical.