Intelligent Task Scheduling Approach for Internet of Health Things with Integrated Cyber-Physical Systems

Abstract

The Internet of Health Things (IoHT) based Cyber-Physical Systems (CPS) have greatly improved the intelligent provision of healthcare services, but secure, reliable, and latency-sensitive task scheduling is still far from being realized in the face of dynamic network environments, resource heterogeneity, and growing cyber threats. In order to overcome these drawbacks, this paper introduces an Intelligent Egret Swarm Optimization Algorithm (ESOA) for efficient task scheduling in IoHT-enabled CPS, which aims to optimize communication performance, security and service reliability simultaneously. The proposed ESOA is applied and tested in the OMNeT++ simulation environment with ECU-IoHT dataset, which leads to better latency, packet delivery, encryption robustness, and system resilience through adaptive task prioritization, intelligent resource allocation, and optimized scheduling strategies. Latency, packet delivery ratio, data encryption strength, system availability, and response time are employed as metrics to compare the performance of ESOA with Moth Flame Optimization (MFO), Deep Reinforcement Learning (DRL) and Biomedical-CPS (BioCPS). The experimental results show a consistent improvement of the ESOA over the current approaches, obtaining an average reduction of 21.4% in latency, an increase of 11.8% in packet delivery ratio, an improvement of 18.6% in data encryption strength, an improvement of 9.7% in system availability, and a reduction of 23.1% in response time compared with the competing approaches. The results demonstrate the effectiveness of the proposed intelligent scheduling framework in secure and real-time communications in healthcare that can be integrated with IoHT, providing better reliability and QoS.