In the recent light of the 4th Industrial Revolution particularly in medical industries, the use of medical robot is an adaptable alternative to reduce the burden in medical services. A mobile robot is normally developed and embedded with numerous sensors especially for its indoor positioning system. This matter has hindered the use of such robots as it increases the development cost. One solution is to employ the readily available sensors such as Wi-Fi. Wi-Fi is widely deployed, and it does not have additional hardware to configure unlike Bluetooth. Despite the potential of Wi-Fi, some adjustment needs to be executed to provide a better accuracy. The Wi-Fi signal has their own disadvantage such as the multipath and shadowing effect making it stand-alone unreliable. Hence, the fingerprinting technique is proposed to solve this issue. The fingerprinting technique mainly used one device in collecting the fingerprinting database. Nonetheless, if the device is broken or new software needed to deploy, the device needs to be substituted. The contribution of this paper is to investigate the behavior of different devices in the fingerprinting method and evaluate the positioning result.
