Assessing Network Communication Metrics in Cloud-Based BCI Performance for Remote Wheelchair Training of Novice Users

Cloud robotics and Brain-Computer Interface (BCI) applications are revolutionizing the field of rehabilitation and e-Health by enabling remote operation of assistive devices without interruption. This work presents a cloud-based BCI system designed to train individuals to control a wheelchair using brain signals. The work focuses on evaluating the Quality of Service (QoS) for the cloud-based BCI by studying the impact of communication parameters, Message Loss Rate (MLR), latency, and jitter, on system performance. The results show that a higher MLR significantly reduces Accuracy (ACC) and Information Transfer Rate (ITR) while increasing CPU and RAM usage. Latency also negatively affects ACC and ITR, with a slight increase in resource consumption, whereas jitter has minimal impact on these metrics. The proposed system demonstrates the importance of cloud robotics in providing continuous support and monitoring, significantly improving rehabilitation and independence for individuals with severe motor impairments. This research is crucial for developing efficient and user-friendly BCI systems for e-Health and cloud robotics applications.