QoS-aware 6G-enabled ultra low latency edge-assisted Internet of Drone Things for real-time stride analysis. (October 2021)

• Record Type: Journal Article
• Authors: Mukherjee, Amartya; Mukherjee, Prateeti; De, Debashis; Dey, Nilanjan
• Is Part Of: Computers & electrical engineering. Volume 95(2021)
• Publication Date: 2021-10
• Subjects: Internet of Drone Things -- Gait -- Randomized search -- Software Defined Networks -- 6G -- Time-series analysis; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2021.107438
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Internet of Things (IoT) concepts constitute a predominant area of research in e-healthcare applications, owing to the plethora of opportunities in medical diagnosis. In this work, a ubiquitous computing and communication architecture is proposed through the amalgamation of Internet of Healthcare and Internet of Drone things by leveraging a 5G/6G communication framework. Gait information is aggregated via a smart shoe and the processing is carried out on a set of edge-enabled Unmanned Aerial Vehicles (UAVs). To transfer the data within the edge and cloud layers, a Software Defined Network (SDN) is modeled. Further, a classifier is designed to analyze the records and make predictions on possible neurological disorders at the edge level. Experimental results suggest a 98% classification accuracy for abnormal gait diagnosis at 20% CPU utilization. The findings further reveal a latency of 335 ms. at QoS 2, and 50 msg/s bandwidth utilization with a Connected Client Ratio and SDN Coverage Ratio of 0.99 and 0.95, respectively. Highlights: The edge-assisted ultra-low latency IoT ecosystem for Parkinson disease assessment technique is proposed. Leverages the Internet of Drone Thing paradigm for step and gait data collection and aggregation. The Latency and the bandwidth performance of the Enhanced MQTT protocols under SDN slices have been analyzed for QoS 0, 1, and 2. An analysis technique and suitable classifier have been proposed to analyze time-series records, which operate in an edge layer. Results show 98% accuracy in abnormal gait diagnosis and disease prediction with a 20% of CPU utilization and 99% of client connection service and 95% of coverage by SDN slices.