Abstract
The number of people afflicted with paralysis is increasing worldwide due to stroke, spinal cord injury, polio, and other related diseases. Exoskeletons have emerged as one of the promising technologies to provide assistance and rehabilitation for the paralyzed people. But most of the exoskeletons are limited by its bulkiness, lack of flexibility and stability, instant control and adaptability. To overcome these issues, this article proposes a novel and efficient software-defined network (SDN)-powered humanoid assistive and rehabilitation system. In the proposed system, the signals acquired by the human sensor module are processed with multiple node MCUs and transmitted via the SDN incorporated with universal software radio peripheral (USRP). Using edge computing, the signal from the USRP is sent to the receiver node MCU and is used for controlling the movements of the humanoid that provides assistance to the paralyzed patients. The experimental setup is done for controlling a humanoid hand, and the results show high quality-of-service (QoS) for hand roll-up and roll-down posture. QoS is also evaluated for different electroencephalogram (EEG) signals, and the results show that the SDN-enabled assistive humanoid system is an efficient method for providing instant control in rehabilitation of the paralyzed patients.