CNN-Driven Resource Optimization for Energy-Efficient and Low-Latency Healthcare Monitoring in Wireless Body Area Networks

The increasing adoption of Wireless Body Area Networks (WBANs) in smart healthcare has created new opportunities for unobtrusive and continuous collection of patients’ physiological data. However, the resource-constrained nature of WBAN devices makes efficient network management particularly challenging. Sensor nodes typically operate with limited battery capacity and communication resources, while healthcare applications require timely and reliable transmission of critical medical information. Poor resource management can therefore result in excessive energy consumption, communication delays, premature node failure, and degradation of overall Quality of Service (QoS).To address these challenges, this study proposes an intelligent resource management approach based on Convolutional Neural Networks (CNNs). The proposed method is designed to determine appropriate communication resource allocation by learning from the interaction between network conditions and performance requirements. Information related to node energy levels, traffic conditions, bandwidth availability, transmission characteristics, and communication delay is incorporated into the learning process. Through this data-driven strategy, the proposed model seeks to achieve a more effective balance between energy utilization and communication performance than traditional resource allocation techniques. The proposed framework is developed and assessed using MATLAB-based simulation experiments. Its performance is examined under different network conditions using several indicators, including residual energy, network lifetime, packet transmission delay, throughput, energy consumption, and the number of inactive sensor nodes. The experimental findings indicate that incorporating CNN-based prediction into the resource management process can improve the overall efficiency and reliability of WBAN communication. In particular, the proposed approach reduces unnecessary energy expenditure and communication delays while supporting more effective data transmission and longer network operation. The results highlight the potential of deep-learning-based resource management as a practical approach for developing reliable, energy-aware, and responsive WBAN architectures for future healthcare applications.

کلیدواژه ها:

Wireless Body Area Networks (WBANs) ، Intelligent Resource Management ، Deep Learning ، Convolutional Neural Networks (CNNs) ، Energy-Aware Communication ، Healthcare Data Transmission ، Network Performance