Test dataset is available here. Please send us your trained models, test codes and the radio map estimates by November 22, 2024.
Join our challenge discord server for announcements and Q&A.
Preliminary evaluations and rankings for Task 1
Preliminary evaluations and rankings for Task 2
Preliminary evaluations and rankings for Task 3
Call for Participation
In wireless communications, the pathloss (or large scale fading coefficient) quantifies the loss of signal strength between a transmitter (Tx) and a receiver (Rx) due to large scale effects, such as free-space propagation loss, and interactions of the radio waves with the objects and structures in the propagation environment, such as penetration, reflection and diffraction.
Many present or envisioned applications in wireless communications explicitly rely on the knowledge of the pathloss function, and thus, estimating pathloss is a crucial task. Some example use cases include: User-cell site association, fingerprint-based localization, physical-layer security, optimal power control, path planning, and activity detection.
To evaluate PL or other wireless channel characteristics one can use radio propagation models. Existing channel modeling techniques exhibit a notorious trade-off: deterministic models, e.g. ray tracing, are highly accurate, yet computationally demanding, while the opposite holds for empirical and stochastic models. Recently, substantial effort has been made to develop data-driven methods, which can be trained to yield commensurate accuracy with deterministic propagation models, accompanied by impressive computational efficiency due to the native graph processing unit (GPU) parallelization of deep neural networks. However, while previous research has focused mostly on deep learning (DL)-based PL inference for outdoor urban environments and isotropic transmission, due to the wide deployment of indoor wireless networks in fifth-generation and beyond (5G/B5G) networks, it's necessary to develop models tailored for indoor environments. In such cases, the refracted electromagnetic field components through obstacles play a more significant role in radio signal propagation, as opposed to the outdoor scenarios that is dominated by reflected field components. Therefore, accurate indoor radio map estimation requires accounting for the larger variety of construction materials and their electromagnetic properties. Similarly, the antenna radiation pattern of the transmitters needs to be appropriately considered.
To advocate further research in this direction and facilitate fair comparisons in the development of DL-based radio propagation models in the less explored case of directional radio signal emissions in indoor propagation environments, and motivated by the success of the First Pathloss Radio Map Prediction Challenge at ICASSP 2023 , we share an indoor PL radio map dataset generated through ray tracing simulations, launching the First Indoor Pathloss Radio Map Prediction Challenge at ICASSP 2025. The Challenge comprises three tasks, aiming at testing the generalizability of DL-based models in different indoor environments, frequency bands, and antenna radiation patterns.
The top 5 ranked teams will be invited to submit a 2-page paper and present it at ICASSP 2025, and the accepted papers will be published in the ICASSP proceedings. Additionally, the top teams will be invited to submit an extended version of their work as a full paper to the IEEE Open Journal of Signal Processing.
The deadline for submission of the trained models, test codes and the radio map estimates is November 22, 2024.
Support on the dataset and the instructions will be provided by the organizing team.
IMPORTANT NOTE: The intellectual property (IP) of the shared/submitted material (e.g. code) will not be transferred to the challenge organizers. When such material is made publicly available by a participant, an appropriate license should accompany.