An intelligent hybrid MAC protocol for a sensor-based personalized healthcare system

Abstract

Sensors based on personalized healthcare systems have been widely used in the medical field. However, energy limitations have greatly hindered the further development of medical sensors. For the traditional Medium Access Control (MAC) protocol, the duration of low-power listening is fixed because it ignores that the available energy of sensors is different in some situations, which leads to a high delay and low energy utilization. In this paper, a Maximum Listening Length MAC (MLL-MAC) protocol is proposed to fully utilize the energy in the sensor-based systems. The MLL-MAC protocol is an improvement of the Receiver-Initiated (RI) MAC protocol. The main advance is that the sensor node performs the following additional operations: (1) The sender sends a beacon when it wakes up and sends data, thus establishing a communication link with the receiver in the listening state; (2) The receiver keeps listening as long as possible to reduce the delay when it wakes up and listens to the channel, which is different from the previous strategy in which the node turns into a sleep state immediately without receiving data. Furthermore, the sensor node can dynamically determine whether to send beacons and prolong listening duration according to its available energy level. The MLL-MAC protocol is evaluated through theoretical analysis and experimental results. The results show that, compared with the RI-MAC protocol, the MLL-MAC protocol can reduce the average end-to-end delay by 41.4% and improve the energy utilization by 15.1%.