Feature Engineering for Grid-based Multi-Floor Indoor Localisation using Machine Learning

Nowadays, the need for indoor localization is increasing as it has many possible implementations in many sectors, e.g. navigation, health care, etc. In order to obtain an accurate indoor location, fingerprinting is the most commonly used method. The problem with fingerprinting approach is the high variation of RSSI values, resulting in erroneous location estimation. Machine learning approach is a new alternative to fingerprinting approach that aims to solve this problem. In this study, we adopt grid-based approach to train the Random Forest models with an end-to-end pipleline to autotune the models hyperparameter. In addition, we explore the impact of features by studying the default RSSI values for undetected AP as well as the feature importance. It highlights the significance of selecting the data pre-processing and the AP's RSSI combinations. RSSI data are collected from WiFi access points (APs) as WiFi access points are widely available and are at fixed location, which does not requires specific hardware support. The evaluation are benchmarked against the self-collected data in the university, called N4, and the publicly available dataset UJIndoorLoc (UJI). The results show that random forest classifier is the overall best performance for classification-based indoor localization, with up to 88.21% and 86.34% for N4 and UJI dataset respectively. By eliminating unimportant APs, the accuracy is maintained at 80% accuracy with 20% of features using PCA. It has shown that our proposed pipeline/framework provides a resilient constructed model in localizing the grid-base location when some of the APs are not detected.