Malicious Nodes In A Multihop Mobile Computer Science Essay

Published: Last Edited:

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, host mobility and lack of infrastructure. Usually, wired and wireless networks use cryptographic techniques for secure communications. Both Symmetric and asymmetric cryptography have 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 gray hole attack. 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.

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 and centralized administration. Mobile hosts are free from any centralized control like base stations and mobile switching centres. Though providing unrestricted mobility and connectivity to users, the onus of network management is entirely on network nodes [1]. 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 theMANET relies on the cooperative participationof 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 centralizedintrusion 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.

It has to be understood that security implies identification of potential attacks, threats and vulnerabilities in a system. Karpijoki [2] and Lundberg [4] 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 [5] and SRP [6], 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 [9] 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., [10] 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., [11] introduced a new distributed key management approach, which is a combination of certificate less 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., [12] 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., [13] 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., [14] 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 TCP/IP or 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. 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 gray hole attack. Figure 1 to Figure 6 show the network performance in terms of acknowledgment sent, cached replies sent, route replies sent, and throughput respectively. All the outputs plotted are in time average format.


Figure 1: Total Acknowledgments sent in cooperating and malicious network

C:\Users\omshree\Desktop\MATERIAL\108\RESULTS\acknowledgement sent.jpg

Figure 2: Total Acknowledgments request sent in cooperating and malicious network

It is seen from the Figures 1 and 2 that the acknowledgments sent when all nodes are cooperating in a network is considerably more when compared to network with malicious nodes. Acknowledgment sent is less in network with malicious nodes as packets are dropped by the malicious nodes during the gray hole attack.


Figure 3: Total Cached Replies sent


Figure 4: Total Replies sent from destination nodes


Figure 5: Total Route Replies sent

The total route replies, cached replies and the replies sent from the destination node is less in networks under attack due to packet dropping.


Figure 6: Throughput in bits/sec for cooperating and malicious network

It is obvious from the above Figures that the networks performance degrades significantly in the presence of malicious nodes. The throughput is severelyaffectedin presence of malicious nodes.

4. Conclusion

MANETs are susceptible to attacks by malicious nodes and packets are dropped in attacks like gray hole attack. It was proposed to evaluate the performance of a network under the impact of malicious nodes. Simulations were conducted using DSR tostudy the performance degradation of network due to malicious node activity.Simulation was conducted using 30 nodes with 15% of the nodes being malicious. Simulation results show that the acknowledgments sent and the various types of route replies are decreased in a network under gray hole attack. Further investigations to study the impact of malicious nodes, different type of attacks needs to conduct. Propose mechanisms based on distributed key management to improve the throughput of the network.