Design Space Exploration of a Multi-Model AI-Based Indoor Localization System

Abstract

In this paper, we present the results of a performance evaluation and optimization process of an indoor positioning system (IPS) designed to operate on portable as well as miniaturized embedded systems. The proposed method uses the Received Signal Strength Indicator (RSSI) values from multiple Bluetooth Low-Energy (BLE) beacons scattered around interior spaces. The beacon signals were received from the user devices and processed through an RSSI filter and a group of machine learning (ML) models, in an arrangement of one model per detected node. Finally, a multilateration problem was solved using as an input the inferred distances from the advertising nodes and returning the final position approximation. In this work, we first presented the evaluation of different ML models for inferring the distance between the devices and the installed beacons by applying different optimization algorithms. Then, we presented model reduction methods to implement the optimized algorithm on the embedded system by appropriately adapting it to its constraint resources and compared the results, demonstrating the efficiency of the proposed method.

Power Efficient Machine Learning Models Deployment on Edge IoT Devices

“…For the results to have any meaningful impact and for our measurements to provide real-world results, the selection of models had to follow a group of criteria as follows: Each model is required to belong to a group of well-defined and well-known ML models. Specifically, all selected models should be popular models whose behavior is well known (their features and parameters are understood) or models that have been used in other scientific research in microcontroller-based systems [ 13 , 14 , 15 ]. This criterion will help to make our results meaningful and easier to understand.…”

Deep Learning-Based Indoor Localization Using Multi-View BLE Signal

“…In contrast to fingerprinting approaches, deep neural networks may generalize better when the measurement environment undergoes changes with the introduction of additional or different multipath propagation components. Most of the existing approaches that employ neural networks (NNs) for indoor localization, either estimate the position directly [ 20 , 21 ] or via a combination of distance estimation and trilateration [ 22 ]. It has to be noted that AoA-based neural network localization approaches exist [ 23 ] but have so far been scarce.…”