| ABSTRACT |
Among the challenges in deploying IEEE 802.15.4 WSN in mission-critical systems such as in healthcare, security and forest health monitoring is that it operates in the ISM band. The overlapping frequency with the IEEE 802.11 proves to be of concern, resulting in a problem that is asymmetric in nature, nonetheless the output power of IEEE 802.15.4 devices is typically low at 0 dBm or less, whereas the output power of 802.11 devices is 20 dBm or above. Moreover, the existence of other Internet of Things (IoT) applications namely microwave oven, wireless USB, smart TV and others operating in the same band gave significant impact to the Quality of Service (QoS) of WSN applications. Available method such as Time-division multiple access (TDMA), Carrier sense multiple access (CSMA) and Time slotted channel hopping (TSCH) employed in the existing off the shelf products proves to improve the wireless network such as reduction of data loss. Nevertheless, above methods are unable to guarantee a good QoS for mission critical applications such as healthcare application. Hence, alternatives to the existing solution such as TDMA and CSMA must be established in order to mitigate the interference issues. This project proposes to intelligently scan and select an optimized channel for data transfer between WSNs motes together with in-house motes targeted for mission critical applications including but not limited to healthcare application. Based on the predetermined parameters, a machine learning algorithm is used to access the condition of the channel and select the best channel that ensure a good QoS for data transfer. Unlike normal WSN monitoring applications, mission critical applications such as in healthcare applications requires bounded end-to-end delay and high Quality of Service (QoS) sensitivity. One packet drop in healthcare application will result on very high probability of failure. For example, during a monitoring of a heartbeat of a patient having a heart attack, a packet drop to report this situation will result on the patient losing his life. In summary, this product consists of in-house WSNs motes for healthcare applications that intelligently perform channel interference mitigation by selecting the best channel for data transmission based on machine learning algorithm. The block diagram below describes the functionality of the said invention.
|