A mobile EMF sensor for appliance identification

This paper explores the feasibility of using a mobile electromagnetic field (EMF) sensor for appliance identification. This sensor uses the EMF radiated by appliances as a signature for appliance identification. It has been prototyped and implemented for data collection. Our mobile EMF sensor will be used for occupant-specific energy measurement in the later phase of the project. This phase, however, focuses on signal extraction and appliance identification using the EMF radiated by the appliances. Using data from the sensor, we compared the performances of k-Nearest Neighbors (k-NN), Support Vector Machines (SVM), and Naïve Bayes classifiers for signature extraction and appliance identification. These classifiers were trained and tested using spike train data obtained from the Fast Fourier Transform (FFT) of the EMF radiated by appliances. We obtained classification accuracies of 96.47% with k-NN and 94.86% with an SVM classifier.