Department of CS and IT, University of Engineering and Technology, Peshawar, KPK, Pakistan.
Department of SE, University of Science and Technology Bannu, Bannu, KPK, Pakistan.
PLoS One. 2024 Sep 12;19(9):e0309920. doi: 10.1371/journal.pone.0309920. eCollection 2024.
Vehicular Adhoc Network (VANET) suffers from the loss of perilous data packets and disruption of links due to the fast movement of vehicles and dynamic network topology. Moreover, the reliability of the vehicular network is also threatened by malicious vehicles and messages. The malicious vehicle can promulgate fake messages to the node to misguide it, which may result in the loss of precious lives. In this situation, maintaining efficient, reliable, and secure communication among automobiles is of extreme importance, especially for a densely populated network. One of the remedies is vehicular clustering, which can effectively perform in a high-density network. However, secure cluster formation and cluster optimization are important factors to consider during the clustering process because non-optimal clusters may incur high end-to-end communication delays and produce overhead on the network. In addition, malicious nodes and packets reduce passenger and driver safety, increase road accidents, and waste passenger and driver time. To this end, we employ Arithmetic Optimization Algorithm (AOA) to design a secure intelligent clustering named AOACNET. AOA is used to achieve optimality of vehicular clusters. During cluster formation, the algorithm prevents unauthentic nodes from becoming cluster members by taking into consideration the performance value of each automobile. The vehicle's performance value is based on the record of data transmission. If a vehicle transmits a fake message, it will receive a penalty of (-1), and in the case of transmitting a legitimate message, a reward of (+1) will be assigned to the vehicle. Initially, all the vehicles have equal performance value which either increase or decrease based on communication with their peers. The vehicles will become cluster members only if their performance value is greater than the threshold value (0). AOACNET is tested in MATLAB using various evaluation metrics (i.e., number of clusters, load balancing, computational time, network overhead and delay). The simulation results show that the proposed algorithm performs up to 25% better than the similar contenders in terms of designated optimization objectives.
车联网（VANET）由于车辆的高速移动和动态网络拓扑结构，会导致危险数据包丢失和链路中断。此外，车辆网络的可靠性也受到恶意车辆和消息的威胁。恶意车辆可以向节点传播虚假消息来误导它，这可能导致宝贵生命的丧失。在这种情况下，保持汽车之间高效、可靠和安全的通信是至关重要的，尤其是在人口密集的网络中。一种解决方案是车载聚类，它可以在高密度网络中有效地执行。然而，在聚类过程中，安全的集群形成和集群优化是需要考虑的重要因素，因为非最优的集群可能会导致高端到端通信延迟，并在网络上产生开销。此外，恶意节点和数据包会降低乘客和驾驶员的安全性，增加道路交通事故，浪费乘客和驾驶员的时间。为此，我们采用算术优化算法（AOA）来设计一种名为 AOACNET 的安全智能聚类。AOA 用于实现车辆集群的最优性。在集群形成过程中，算法通过考虑每个汽车的性能值来防止非可信节点成为集群成员。车辆的性能值基于数据传输记录。如果车辆发送虚假消息，它将收到（-1）的惩罚，如果发送合法消息，将向车辆分配（+1）的奖励。最初，所有车辆的性能值都相等，并且根据与同行的通信，性能值会增加或减少。只有当车辆的性能值大于阈值（0）时，车辆才会成为集群成员。AOACNET 在 MATLAB 中使用各种评估指标（即集群数量、负载均衡、计算时间、网络开销和延迟）进行测试。仿真结果表明，与类似的竞争者相比，所提出的算法在指定的优化目标方面性能提高了 25%。
