Dynamic intrusion detection for internet of drones using glowworm swarm optimization

The Internet of Drones (IoD) is spreading at excessive speeds that have posed a challenge in operation in that should be both secure and viable in terms of identifying the threats of intrusion and avoiding them in a dynamic and resource limited setting. In the given paper, a novel technique called DIGS-RTD is presented and this is a fresh framework using the bio-inspired Glowworm Swarm Optimization (GSO) algorithm to dynamically pick features. The luminescent communication based GSO approach can be applied to multimodal optimization problems, with its approach to the optimization of multimodal components appropriately applicable to locate a variety of intrusion patterns, as well as very complex intrusion patterns. The proposed framework has an active feature selection mechanism that aims to reduce the size of the feature subset that is utilized in intrusion detection to minimize the number of computational load without compromising the accuracy of detection.An efficient classification model has been integrated with the selected features to detect the intrusions in real-time. The system is compared with state-of-the-art bio-inspired approaches, namely Sea Turtle Foraging Algorithm (STFA), against a wide range of performance metrics, including detection accuracy, false positives, false negatives, processing time and memory usage. The scalability and adaptability of the framework in handling the increasing diversity of IoD with the data flow rate and network size are also evaluated. Through experimental evaluation, the approach GSOSTFA proves of having superior trade-off between computational cost and detection performance by decreasing false positives and processing time against baseline methods. The dynamic nature of the IoD and the framework’s ability to adapt to changing environments highlights the potential of this as a powerful method of securing IoD operations against the constantly evolving landscape of cyber threats. This represents an important advance in the field of IoD security, with efficient and scalable real-time capabilities for intrusion detection in next-generation drone networks.