Power Optimization by Detection and Monitoring of Sensor Event in Smart Home

Sensor event monitoring and detection facilitate to recognize the activities of the human lives in a smart home daily. The major aim of this technology is to attain optimal performance in activity recognition. The symbol of the daily activity recognition involves the usage of feature inputs which have a noteworthy consequence on the output. However, commonly used representations of features dependent on daily activity have limited performance on the recognition activity. Based on the dynamic nature of the sensor events caused by daily activities, this paper presents a statistical processing approach for time series of sensor events. First time, the statistical values are extracted from sensor events dependent on time series. Subsequently, different categories of statistic formulae are proposed to solve daily activity features. To evaluate the proposed approach, two distinct datasets are adopted for activity recognition based on artificial neural network (ANN), fuzzy logic (FL), and hybrid neuro-fuzzy logic (HNFL). The experimental results reveal that the proposed HNFL approach provides power optimization.