Secure communication routing and attack detection in UAV networks using Gannet Walruses optimization algorithm and Sheppard Convolutional Spinal Network

Unmanned Aerial vehicles (UAV) are high-speed moving machines that attained rapid growth in various activities and are considered an integral component in the Satellite-Air -Ground-Sea (SAGS) incorporated network. However, UAVs are affected by communication delays and malicious attacks. Therefore, a...

Full description

Saved in:
Bibliographic Details
Published in:Peer-to-peer networking and applications Vol. 17; no. 5; pp. 3269 - 3285
Main Authors: Renu, Yuvaraj, Sarveshwaran, Velliangiri
Format: Journal Article
Language:English
Published: New York Springer US 01.09.2024
Springer Nature B.V
Subjects:
Algorithms
Artificial neural networks
Communication
Communication networks
Communications Engineering
Computer Communication Networks
Data communication
Decision making
Engineering
Information Systems and Communication Service
Multiple objective analysis
Networks
Neural networks
Nodes
Optimization
Optimization algorithms
Packets (communication)
Recurrent neural networks
Routing (telecommunications)
Signal,Image and Speech Processing
Special Issue on 2 - Track on Security and Privacy
Unmanned aerial vehicles
Agent-based security
Unmanned aerial vehicles attacks
Sheppard Convolutional Spinal Network
Gannet Walruses Optimization
Secure routing
Deep Recurrent Neural Network
ISSN:1936-6442, 1936-6450
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Unmanned Aerial vehicles (UAV) are high-speed moving machines that attained rapid growth in various activities and are considered an integral component in the Satellite-Air -Ground-Sea (SAGS) incorporated network. However, UAVs are affected by communication delays and malicious attacks. Therefore, an adequate and secure communication routing and attack detection model is necessary for UAV communication networks. This research described a novel approach for initiating secure communication in UAV networks namely Gannet Walruses Optimization Algorithm + Sheppard Convolutional Spinal Network (GWOA + ShCSpinalNet). Initially, the UAV network is simulated, and the data packets are transmitted among the nodes using optimal routing paths. An optimal routing path is computed using the Gannet Walruses Optimization Algorithm (GWOA) by considering some multi-objective functions through the Deep Recurrent Neural Network (DRNN). The developed GWAO integrates Gannet Optimization (GOA) and Walruses Optimization (WaOA). The data communication is done through monitoring agents. The newly devised Sheppard Convolutional Spinal Network ( ShCSpinalNet) is utilized as a decision-making agent for malicious attack detection. The attributes considered for decision-making are round trip time, packet delivery ratio, the strength of the signal, the size of the packet, and the number of incoming packets. Once the SpinalNet categorizes the normal and attacked nodes the defense agent is implemented for attack migration. The ShCSpinalNet is devised by the combination of the Sheppard Convolutional Neural Network and Spinal Network. The GWOA + ShCSpinalNet accomplishes a diminished delay of 0.614 s, an increased detection rate of 0.930%, an energy of 0.439 J, and a Packet Delivery Ratio (PDR) of 0.749.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1936-6442
1936-6450
DOI:10.1007/s12083-024-01753-4