Thermal signal analysis in smart home environment for detecting a human presence

The Elderly’s Independent Living in Smart Homes: A Characterization of Activities and Sensing Infrastructure Survey to Facilitate Services Development

“…They also use a “first order hold” (FOH) filter to remove redundancy in the dataset collected from presence sensors placed in different areas in a smart home. Guettari et al [ 113 ] also used a median filter to avoid abnormal measurements of passive presence sensors used to detect human presence within a smart home.…”

On analyzing user location discovery methods in smart homes: A taxonomy and survey

“…User privacy, installation [40] Simplicity, easy to use needing to have data set, user privacy, error prone (noise in depth image information, camera motion estimation inaccuracy) [6] Multiple users, reliability (cameras and sensory floor) User privacy, installation [7] reliability (cameras, microphones), accuracy, multiple users User privacy, complexity, installation [41] scalability, user privacy, low complexity, robustness, fault tolerance accuracy depends on number of sensors, installation [42] Simplicity, easy to use, user privacy low accuracy, limitation to only one person [24] User privacy, multiple users installation, scalability [44] easy to use and simplicity, user privacy effected by the clutter (multiple-input multipleoutput radar system reduced this effect) [31] User privacy, easy to use relative accuracy (Increasing the number of reference nodes can improve accuracy), signal strength is effected by environmental parameters such as humidity or by interference from other RF sources [43] User privacy, easy to use and comfortable Scalability, only one user [49] Cost, comfortable, privacy Complexity, accuracy, need to have recorded observations [54] Cheap, easy to use, energy efficient, dealing with noise, recovery from the failure Accuracy depends on number of sensors, limited to single person [57] Based on cheap and low cost PIR sensors, privacy, user comfort…”

“…Guettari et al [49] proposed a ULD method using thermopile sensors. The method considers two states of occupied and unoccupied for each room and detects person presence in a room using the two time series generated by a thermopile sensor (First time series corresponds to temperature measured and the second one representing the sensor's temperature).…”

“…[34] Video camera, access Point, mobile phone 2 video cameras , 6 access points High [35] Inertial measurement unit (IMU) sensors, insole devices, wireless receiver 8 IMU sensors ,cameras ,Insole Devices, 1 Wireless Reciver Very high [37] Server camera (32 bit RISC CPU), client camera 2 Server cameras, 6 client cameras High [38] Camera 2 Ceiling security camera Very high [40] Kinect camera 2 Kinect camers Very high [41] wireless pyroelectric infrared sensors, 802.15. [49] Wireless thermopile PIR 8 wireless PIRs Very high [54] ZigBee devices equipped with PIRs, Router 4 ZigBee devices,1 Router Very high [57] PIR sensors 50 PIR sensors Very high…”