Real Time Health Monitoring Based on the Interaction of Social Networks and Body Sensors

Abstract

x newlineABSTRACT newlineRecent technological advances in sensors, low-power integrated circuits, and newlinewireless communications have enabled the design of low-cost, lightweight, miniature newlineand intelligent physiological sensor nodes. These sensor nodes are capable of sensing, newlineprocessing, and communicating one or more vital signs, and can be seamlessly newlineintegrated into social sensor networks for healthcare applications. These networks newlinepromise to revolutionize healthcare by allowing inexpensive, non-invasive, newlinecontinuous, ambulatory health monitoring with almost real time updates of medical newlinerecords through the internet. Though a number of ongoing research efforts are newlinefocusing on various technical, economic and social issues, several technical hurdles newlinestill need to be solved in order to have various technical, economic, and social and newlinepower-efficient social sensor networks suitable for medical and emergency newlineapplications. newlineThe present research work discusses the implementation issues and describes newlinethe social sensor networks for health monitoring. The presented system uses wearable newlinesensor solutions to improve the growth rate, packet delivery ratio, and efficiency and newlineresponse time. Clustering is one approach towards dealing with those sensor network newlinerouting-related problems. However, due to infrastructureless nature of WSN (Wireless newlineSensor Network), it is sometimes difficult for the protocols specialized in clustering newlinetechniques to divide sensor networks into clusters in conjunction of their ClusterHead selection criterion. newlineAnother problem faced with sensor networks is the energy constraints, where newlinesome nodes are dead due to their low battery power, resulting in degradation of the newlinewhole network s performance. The third problem is related to the recommendation newlineservices. In order to overcome all these negative issues, new frameworks were newlineproposed in this study such as, HIVSSC: HIV Sensor Symptom Clustering (HIV newlineassociation assistance and suggestions based on Symptoms Based Sensing newlineClustering), SCRHF (Sensor clustering for Reducing Heart failure), FitbitSurge-AEM newline(Fitbit Surge-Activity Extraction Model), and ASD-EBEM (Autism Spectrum newlineDisorder-Emotions and Behavior extraction model). newlinexi newlineFinally, we have compared our new four protocols to find out which one was newlineperforming better than others. The comparative analysis was conducted by newlineconsidering various parameters namely: Cluster Formation Time, Performance of newlinePacket Delivery Ratio, Efficiency of nodes energy, Jitter, Overhead and End-to-End newlineDelay. The results proved that the FitbitSurge-AEM protocol outperformed other newlineprotocols when considering Cluster Formation Time, Performance of Packet Delivery newlineRatio, Jitter, Overhead and End-to-End Delay parameters. And concerning the newlineefficiency of saving energy of nodes, SCRHF has shown high efficiency when newlinecompared to the other three remaining frameworks. newlineThe simulation results were provided with the OMNET++ simulator by varying the newlinenumber of nodes and by evaluating various parameters namely, routing overhead, newlinepacket delivery ratio, normalized routing load, and the average end-to-end delay for newlinecomparing the performances of the proposed scheme with the existing approaches newlinesuch as, the energy based routing algorithms, clustering algorithms, mobility based newlinealgorithms, routing based protocols algorithms. newlineKeywords: Wireless Sensor Networks, Social Networks, Health Monitoring, newlineClustering, Load balancing, Recommendation System, Routing. newline