Physical intrusion detection and home-office automation systems using harvested radio frequency energy

Abstract

With the rapid growth of wireless network sensor (WSN) technology, the improvement of low data rate, low cost, low power consumption and long battery life of ZigBee wireless sensor networks has been reported. The use of wireless sensor technology has proliferated in various fields namely; military security, environmental monitoring, medical and home automation. There is an emerging application for ZigBee sensor technology for indoor perimeter security and physical intrusion detection. Hence, the scopes of this research work is to develop a ZigBee-based system that can double as an alarm for detection of physical presence of an individual in a confined indoor environment or for automation in office environment. The developed systems work in two phases; an offline learning phase and an online active phase, while utilizing freely available Radio Frequency for Wi-Fi and ZigBee. A statistical profiling is used to identify a purposely emptied room and then using a time-widow statistical analysis, the system monitors the indoor environment to detect physical intrusion. Variance, standard deviation and kurtosis are found to be excellent candidates to indicate the slightest changes in the RF field. These received signal strength indicator (RSSI) fluctuations are used also to switch ON/OFF appliances, lighting and air conditioning in a room, office, and classroom and laboratory environment. The antenna orientation, separation distance between transmitter and receiver, vertical positioning of sensors and radio signal irregularities are studied to refine and improve the accuracy of the developed system. Results achieved for the alarm indicate a physical intrusion detection accuracy of 100% for a separation distance of less than 5 meters. Further separation severely degrades the accuracy performance and limits the flexibility of placement of sensor nodes. However, when doubling as a control switch for electrical appliances, the system performed well for a large room with more than 50 meters separation distance between transmitter and receiver utilizing the existing Wi-Fi signals around campus.