In this work, we present a wireless localization method that operates on self-supervised and unlabeled channel estimates. Our self-supervising method learns general-purpose channel features robust to fading and system impairments. Learned representations are easily transferable to new environments and ready to use for other wireless downstream tasks. To the best of our knowledge, the proposed method is the first joint-embedding self-supervised approach to forsake the dependency on contrastive channel estimates. Our approach outperforms fully-supervised techniques in small data regimes under fine-tuning and, in some cases, linear evaluation. We assess the performance in centralized and distributed massive multiple-input multiple-output (MIMO) systems for multiple datasets. Moreover, our method works indoors and outdoors without additional assumptions or design changes.

The ALEX²-test (MacroArray Diagnostics, Vienna, Austria) is a diagnostic multiplex IgE-test for the simultaneous detection of IgE to 178 allergens and 117 extracts, in addition to total IgE. Test results from more than 90 countries are stored on a GDPR-compliant cloud server for backup, customer support, and continuous postmarket surveillance. To improve the coverage of exposomes on a global scale and to further increase the sensitivity of the test, the allergen panel was updated from ALEX² to ALEX³. By mid-2023, when ALEX³ was designed, almost 400,000 real-world ALEX² test results were available. Analysing prevalences and average sIgE-levels of individual allergen preparations, coverage of extracts by components, and co-reactivity of members of the same allergen family provided a rationale for updating the array. In parallel, based on the scientific literature and clinical studies, new allergens were selected. On ALEX³, 218 allergens and 82 extracts will be represented, including 52 new allergens. Allergen preparations with low prevalence and clinical relevance, as well as redundant allergens and extracts, were discontinued. New allergens encompass, e.g., cyclophilins, alpha-gal, and additional markers from respiratory and food allergen sources. Using a large dataset of ALEX² test results exemplifies the targeted, data-driven improvement of a diagnostic IgE-macroarray.

In this paper, we systematically study the electro-magnetic (EM) and communication aspects of a RIS through EM simulations, system-level and ray tracing simulations, and finally measurements. We simulate a nearly perfect, lossless RIS, and a realistic lossy anomalous reflector (AR) in a ray tracer and analyze the large-scale fading of simple RIS-assisted links. We also compare the results with continuous and quantized unit cell reflection phases with one to four-bit resolutions. Finally, we perform over-the-air communication link measurements in an indoor setting with a manufactured sample of a wide-angle AR. The EM, system-level, and ray tracing simulation results show good agreement with the measurement results. It is proved that the introduced macroscopic model of a RIS from the EM aspects is consistent with our proposed communication models, both for an ideal RIS and a realistic AR. The verified system-level simulator and ray tracer for a RIS could be tailored to, e.g., the wireless communication system engineers in the cellular network planning business, providing tools to optimize network performance.

In this chapter, we provide an overview of several data-driven techniques for wireless localization. We initially discuss shallow dimensionality reduction (DR) approaches and investigate a supervised learning method. Subsequently, we transition into deep metric learning and then place particular emphasis on a transformer-based model and self-supervised learning. We highlight a new research direction of employing designed pretext tasks to train AI models, enabling them to learn compressed channel features useful for wireless localization. We use datasets obtained in massive multiple-input multiple-output (MIMO) systems indoors and outdoors to investigate the performance of the discussed approaches.