Internet of Ambience: An IoT Based Context Aware Monitoring Strategy for Ambient Assisted Living (original) (raw)

— In this paper we enhance the innovative architectural model for context-aware monitoring, BDCaM that uses cloud computing platforms. Every generated context of Ambient Assisted Living (AAL) systems is sent to the cloud. A number of distributed servers in the cloud store and process those contexts to extract required information for decision-making using this novel technique. We develop a 2-step learning methodology. In the first step, the system identifies the correlations between context attributes and the threshold values of vital signs. Using Map Reduce Apriori algorithm, over a long term context data of a particular patient, the system generates a set of association rules that are specific to that patient. In t he second step, the system uses supervised learning over a new large set of context data generated using the rules discovered in the first step. In this way, the system becomes more robust to accurately predict any patient situation. We demonstrate the performance and efficiency of BDCaM model in situation classification by implementing a case study. Our system refines patient-specific rules from big data and simplifies the job of healthcare professionals by providing early detection of anomalous situations with good accuracy. The big data producers of BDCaM model are a large number of AAL systems. The low level setup of each system varies according to the requirements of the patient. The sensors, devices and software services of each AAL system produce raw data that contain low level information of a patient's health status location, activities, surrounding ambient conditions, device status, etc. This paper would promote a lot of research in the area of application of IoT in Ambient Assisted Living.