Energy Efficient Wireless Sensor Network for Smart Building

Abstract

A significant amount of energy is being used in various requirements of buildings, especially HVAC (Heating, Ventilation, Air-conditioning and lighting) system. Smart building concept is to minimize energy consumption of buildings while maintaining or improving the living comfort of buildings. For this purpose, various parameters related to building, such as temperature, humidity, CO2 percentage, are being monitored continuously. The monitoring of these parameters can be done efficiently by wireless sensor networks. One of the major challenges of Wireless Sensor Network (WSN) is its limited battery lifetime. Hence WSN should be designed in such a way that network lifetime is maximized. This thesis explores and implements various methods to enhance the network lifetime of wireless sensor network designed for the building. A clustering algorithm is proposed to generate the sleep-wakeup scheduling of sensor nodes. The clustering algorithm designed for this purpose is a novel effort, which takes into account similarity patterns of temperature values and geographic location of sensor nodes as well. PEGASIS, an existing routing protocol for multi-hop routing, is modified for further improvement in network lifetime. The proposed approach is evaluated in an office building of size 35m x 35m which is designed with all parameters, such as outside temperature, HVAC set points, percentage of windows and shading around building, activity schedules of building, in Design-Builder tool. Then the temperature profile of the building for a time period of one year is generated using Energy Plus tool. Network Simulator (NS2) is used, to implement wireless sensor network to sense temperature values of various zones inside the above mentioned virtual building. The results show significant improvement in network lifetime with not so much compromising the quality of sensed data.