RADIO FREQUENCY BASED INPAINTING FOR INDOOR LOCALIZATION USING MEMORYLESS TECHNIQUES AND WIRELESS TECHNOLOGY

Recently, the Internet of Things (IoT) has grown to encompass the surveillance of devices through the utilization of Indoor Positioning Systems (IPS) and Location Based Services (LBS). One commonly used method for developing an Intrusion Prevention System (IPS) is to utilize wireless networks to determine the location of the target. This is achieved by leveraging devices with known positions. Location-based services (LBS) play a vital role in many smart building applications, enabling the creation of efficient and effective work environments. This study examines four memoryless positioning algorithms, namely K-Nearest Neighbour (KNN), Decision tree, Naïve Bayes and Random Forest regressor. The algorithms are compared based on their performance in terms of Mean Square Error, Root Mean Square Error, Mean Absolute Error and R2. A comparative analysis has been conducted to verify the outcomes of different memoryless techniques in Wi-Fi technology. Based on empirical evidence, Naïve Bayes has been determined to be the localization strategy that exhibits the highest level of accuracy. The dataset containing the Received Signal Strength Indicator (RSSI) measurements from all the studies is accessed online.