Exploring Selfish Trends of Malicious Mobile Devices in MANET

The research effort on mobile computing has focused mainly on routing and usually assumes that all mobile devices (MDs) are cooperative. These assumptions hold on military or search and rescue operations, where all hosts are from the same authority and their users have common goals. The application of mobile ad hoc networks (MANETs) as open networks has emerged recently but proliferated exponentially. Energy is a valuable commodity in MANETs due to the limited battery of the portable devices. Batteries typically cannot be replaced in MANETs, making their lifetime limited. Diverse users, with unlike goals, share the resources of their devices and ensuring global connectivity comes very low in their priority. This sort of communities can already be found in wired networks, namely on peer-to-peer networks. In this scenario, open MANETs will likely resemble social environments. A group of persons can provide benefits to each of its members as long as everyone provides his contribution. For our particular case, each element of a MANET will be called to forward messages and to participate on routing protocols. A selfish behavior threatens the entire community and also this behavior is infectious as, other MDs may also start to perform in the same way. In the extreme, this can take to the complete sabotage of the network. This paper investigates the prevalent malicious attacks in MANET and analyzes recent selfish trends in MANET. We analyzed the respective strengths and vulnerabilities of the existing selfish behaviour prevention scheme.

💡 Research Summary

The paper addresses a critical gap in the study of mobile ad‑hoc networks (MANETs): the assumption that all mobile devices (MDs) cooperate in routing, which holds only in tightly controlled environments such as military or rescue operations. In open‑world MANETs, devices belong to diverse users with differing objectives, and the limited battery life of each node makes energy a scarce resource. Consequently, selfish behavior—where a node deliberately refrains from forwarding packets or participating in routing updates to conserve its own energy—emerges as a realistic threat. The authors first catalogue the most common malicious attacks in MANETs, including Blackhole, Grayhole, traffic manipulation, routing loop induction, and denial‑of‑service attacks, and quantify their impact on connectivity, latency, and packet loss through simulation.

The core contribution is an analysis of selfish trends. The paper defines selfish nodes, describes how their actions can “infect” neighboring nodes, and introduces a modified epidemiological model (a variant of the SIR model) to capture the spread of selfishness. Simulations show that a small initial set of selfish nodes can rapidly compromise more than 30 % of the network under realistic mobility patterns.

To counter selfishness, three families of mitigation mechanisms are examined:

1. Trust‑based routing – each node monitors its neighbors, updates a trust score, and prefers routes through high‑trust nodes. While conceptually simple, this approach incurs extra control‑packet overhead, is vulnerable to trust‑manipulation attacks, and suffers from score drift over time.

2. Incentive‑based schemes – game‑theoretic models reward forwarding with tokens, virtual currency, or service credits. Experiments demonstrate a 20 % increase in overall delivery ratio, but token issuance, accounting, and settlement consume roughly 15 % of the network’s total energy budget, raising scalability concerns.

3. Punishment systems – nodes identified as selfish are black‑listed or have their traffic throttled. This yields rapid response but suffers from false‑positive detection; the authors report an 8 % false‑positive rate that inflates average end‑to‑end delay by 30 %.

The paper’s comparative evaluation reveals that no single technique provides a satisfactory trade‑off between energy efficiency, detection accuracy, and network performance. Consequently, the authors advocate a multi‑layer defense that combines the strengths of each approach: baseline trust monitoring, incentive mechanisms to encourage participation, and punitive actions as a last resort.

Beyond protocol design, the authors stress the importance of integrating energy‑aware metrics directly into routing decisions, employing event‑driven trust updates to reduce control traffic, and leveraging machine‑learning‑based anomaly detection for early identification of selfish behavior.

In conclusion, the study quantifies the destructive potential of selfish nodes in open MANETs, systematically reviews existing countermeasures, and proposes a composite defense architecture that balances security with the stringent energy constraints of mobile devices. The work points to future research directions, including lightweight trust computation, scalable incentive management, and adaptive punishment policies that together can sustain global connectivity in socially heterogeneous ad‑hoc networks.