A Comprehensive Survey of Machine Learning Based Localization with Wireless Signals

Abstract

The last few decades have witnessed a growing interest in location-based services. Using localization systems based on Radio Frequency (RF) signals has proven its efficacy for both indoor and outdoor applications. However, challenges remain with respect to both complexity and accuracy of such systems. Machine Learning (ML) is one of the most promising methods for mitigating these problems, as ML (especially deep learning) offers powerful practical data-driven tools that can be integrated into localization systems. In this paper, we provide a comprehensive survey of ML-based localization solutions that use RF signals. The survey spans different aspects, ranging from the system architectures, to the input features, the ML methods, and the datasets.A main point of the paper is the interaction between the domain knowledge arising from the physics of localization systems, and the various ML approaches. Besides the ML methods, the utilized input features play a major role in shaping the localization solution; we present a detailed discussion of the different features and what could influence them, be it the underlying wireless technology or standards or the preprocessing techniques. A detailed discussion is dedicated to the different ML methods that have been applied to localization problems, discussing the underlying problem and the solution structure. Furthermore, we summarize the different ways the datasets were acquired, and then list the publicly available ones. Overall, the survey categorizes and partly summarizes insights from almost 400 papers in this field.This survey is self-contained, as we provide a concise review of the main ML and wireless propagation concepts, which shall help the researchers in either field navigate through the surveyed solutions, and suggested open problems.

A Tutorial on Terahertz-Band Localization for 6G Communication Systems

“…However, this category can only perform data preprocessing, and location information cannot be obtained. [38]- [40], [42], [209], [210]. In summary, despite the channel at THz frequencies being more deterministic than at lower frequencies, which suits geometry-based methods well, we argue that learning-based methods still have advantages in two aspects.…”

“…In terms of localization, a number of surveys exist and share the localization basics and performance metrics in common. However, their goals are totally different and their main focuses can be categorized based on the environment (indoor [32], [35], [39], [41], outdoor [2], [33], [44] or both), techniques (SLAM [2], MDS [36], machine learning (ML) [38]- [40], etc. ), and signal types (radio signal [5], [6], visible light [9], RFID [41], etc.).…”

“…3) Practical ML algorithms: Machine learning algorithms are usually classified into supervised learning (used for solving classification and regression problems) and unsupervised learning (used for data clustering) [38]. Other approaches, including semi-supervised learning, reinforcement learning, transfer learning, and federated learning, are designed to solve the issues faced by the supervised and unsupervised learning algorithms, which will be discussed as follows.…”

A Survey of Indoor Localization Systems in Multi-Floor Environments

“…Note that KNN can be considered as an ML classifier too. For more details about using ML in localization, the reader can refer to [28], [29].…”

Artificial intelligence enabled radio propagation for communications-Part II: Scenario identification and channel modeling

“…But with the development of multiantenna techniques and the computing power of the devices, more and more channel information can be obtained during communications. The contribution of the channel information to the localization/positioning has been extensively investigated [127]. According to the existing studies, the propagation environment impacts on the localization/positioning accuracy significantly.…”