Polysomnography (PSG) is a fundamental diagnostical method for the detection of Obstructive Sleep Apnea Syndrome (OSAS). Historically, trained physicians have been manually identifying OSAS episodes in individuals based on PSG recordings. Such a task is highly important for stroke patients, since in such cases OSAS is linked to higher mortality and worse neurological deficits. Unfortunately, the number of strokes per day vastly outnumbers the availability of polysomnographs and dedicated healthcare professionals. The data in this work pertains to 30 patients that were admitted to the stroke unit of the Udine University Hospital, Italy. Unlike previous studies, exclusion criteria are minimal. As a result, data are strongly affected by noise, and individuals may suffer from several comorbidities. Each patient instance is composed of overnight vital signs data deriving from multi-channel ECG, photoplethysmography and polysomnography, and related domain expert’s OSAS annotations. The dataset aims to support the development of automated methods for the detection of OSAS events based on just routinely monitored vital signs, and capable of working in a real-world scenario.

Positioning is a task of increasing importance in today’s life, with applications in many different contexts. The fact that people spend most of their time inside buildings makes indoor positioning a fundamental research topic. A variety of approaches to indoor positioning have been proposed over the years; however, most often it is carried out through WiFi fingerprinting, by reason of its ease of deployment. WiFi fingerprints are (arrays of) signal strengths of access points observed at given locations. A fingerprint can thus be viewed both as a point in the real-world 2D/3D (geometrical) space and as a point in the high dimensional (fingerprint) space determined by access points. In this paper, we explore the relationships between these two spaces. To the best of our knowledge, such an investigation has never been systematically carried out in the literature, that mainly focuses on position estimation. We first highlight some connections between the geometrical and the fingerprint space. Then, we show that, by making use of conventional approaches, it is unfeasible to reconstruct real-world spatial knowledge on the basis of fingerprint data only. All findings are supported by a thorough analysis that takes into consideration various metrics, fingerprint representations (i.e., normalization functions), and granularity levels (e.g., a single building rather than a single floor), and makes use of 15 public well-recognized indoor datasets. Some general considerations on the outcomes of the analysis conclude the paper.

In many application domains, position information is of fundamental importance. However, unlike the case of outdoor positioning, producing an accurate position estimation in the indoor setting turns out to be quite difficult. One of the most common localisation strategies makes use of fingerprinting. Research in this area has been faced with a number of challenges, leading to the proposal of a number of localisation algorithms, sampling strategies, benchmark datasets, and representations of building information. This proliferation made the modeling of the indoor positioning domain quite hard from both a theoretical and a practical point of view. In this paper, we propose a general and extensible framework, based on a relational database, that pairs fingerprints with building information.We show how the proposed system successfully deals with a number of problems that affect indoor positioning, supporting a large set of relevant tasks. The source code of the framework is available online, as well as an implementation of it, that provides an interactive open repository of indoor positioning data.