A way to reduce the falling risk in the area where closed-circuit television (CCTV) is not available has been studied systematically using ultrasonic sensors for detection. The ultrasonic sensors, behaving a receiver and a transmitter, connect to Arduino microcontroller in order to send the signal via WiFi to computer. The sensors are positioned to array on top and wall of room. Then, the threshold signal is analyzed to find the action such as stand, sit and fall by comparison between top and side signal. The various side signals can indicate the failure. In addition, the signal processing can achieve the temperature around the simple leading to classify an object or a human. The accuracy to indicate the fall is 92 percentages.