In wireless communications, propagation modeling is an essential task that allows engineers to understand and predict the behavior of emitted radio waves in a propagation environment of interest. Often, propagation models are evaluated on a dense grid in the area of interest to produce so-called radio maps.
Deterministic simulation methods such as ray tracing (based on a high-frequency approximation of Maxwell's equations) are known to provide highly accurate estimates of real-world electromagnetic propagation. However, such methods suffer from high computational complexity and their accuracy depends on the availability of detailed and reliable knowledge of the propagation environment, i.e. the shape and material of the objects (such as buildings or vegetation) in the propagation environment.
The generation of pathloss (or large scale fading coefficient) radio maps by such models has been of particular interest due to its ubiquitous appearance in wireless system design problem formulations, such as in user-cell site association, activity detection, fingerprint-based localization, physical layer security, optimal power control, and path planning. Specifically, the pathloss is the quantity of the loss in signal strength between a transmitter (Tx) and a receiver (Rx) due to large scale effects such as free-space propagation loss, and interactions (e.g. penetration, reflection and diffraction) of radio waves with obstacles (e.g. buildings, vehicles, pedestrians) that block the line-of-sight (LOS).
Recently, many research groups have shown that well-designed deep neural networks can approximate highly complex propagation models very accurately and with orders of magnitude less computation time, providing accurate propagation estimates for the intended applications.
The aim of the workshop is to provide an overview of the latest developments in the field of radio propagation modeling, in particular learning-based methods. The applications of propagation models (or the radio maps generated by the models) are also of interest.
In particular, wireless localization is an application that greatly benefits from the availability of accurate and rapidly computed radio maps. Localization based on wireless signal signatures is a very active research area that owes its importance to the failure of Global Navigation Satellite Systems to provide good accuracy in environments where satellite signals are blocked by obstructions such as buildings or trees (or indoor scenarios).
High quality radio maps of the environment of interest allow matching the observed signal characteristics (e.g., pathloss, time of arrival, or other forms of channel state information) with the positions with similar characteristics in the radio map estimates. The design of efficient matching algorithms that yield high accuracy is also an active area of research with a long history. The importance of accurate radio maps was recognized many years ago, but due to the impossibility of generating accurate and fast radio maps at that time, radio map-based localization was considered infeasible.
With the recent advent of efficient machine learning algorithms, in particular deep learning, radio map-based localization, and many other wireless applications that rely on fast and accurate radio map predictions, have attracted a great deal of interest from many wireless system engineers, researchers, and even machine learning practitioners/researchers without a wireless communications background. In addition, the modeling of radio wave propagation has many similarities to the formulation of ray tracing problems in computer graphics. Overall, the topics covered by the workshop are of great importance to many different disciplines due to their many overlapping aspects.
In this regard, we hope that this workshop can be an opportunity for different research communities to interact and share their knowledge and experience.
In addition, we hope that the workshop will also serve as a means of attracting the interest of communities that have been working on very similar problems that have gone unnoticed. An example might be audio (source) localization research, where measured sound wave characteristics are used to localize the source of sound. Somehow, even though wireless and audio localization have a lot in common, the two communities have not made references to each other.
We hope to attract novel contributions on the above points, and also to foster the identification of next research questions through panel discussions and Q/A sessions of the invited/plenary talks and of the regular papers.
Prof. Andreas F. Molisch
Dr. Mate Boban
For papers to be published on IEEE Xplore
11 January 2024, Paper Submission Deadline
31 January 2024, Workshop Paper Acceptance Notification
8 February 2024, Workshop Final Paper Submission Deadline
For non-IEEE Xplore papers
2 March 2024, Paper Submission Deadline
22 March 2024, Workshop Paper Acceptance Notification
30 March 2024, Workshop Final Paper Submission Deadline