Filipe Meneses

Combining similarity functions and majority rules for multi-building, multi-floor, WiFi positioning

Fingerprint is one of the most widely used methods for locating devices in indoor wireless environments and we have witnessed the emergence of several positioning systems aimed for indoor environments based on this approach. However, additional efforts are required in order to improve the performance of these systems so that applications that are highly dependent on user location can provide better services to its users. In this work we discuss some improvements to the positioning accuracy of the fingerprint-based systems. Our algorithm ranks the information about the location in a hierarchical way by identifying the building, the floor, the room and the geometric position. The proposed fingerprint method uses a previously stored map of the signal strength at several positions and determines the position using similarity functions and majority rules. In particular, we compare different similarity functions to understand their impact on the accuracy of the positioning system. The experimental results confirm the possibility of correctly determining the building, the floor and the room where the persons or the objects are at with high rates, and with an average error around 3 meters. Moreover, detailed statistics about the errors are provided, showing that the average error metric, often used by many authors, hides many aspects on the system performance.

An open architecture for developing mobile location-based applications over the Internet

The mobile Internet is enabling a broad range of new applications that dynamically obtain information that is relevant to their current location. This type of application would greatly benefit from generic mechanisms for supporting the association between network resources and physical space, but existing systems are typically based on vertical approaches valid only for narrow application scenarios. This paper argues that a comprehensive solution to this issue should address the important challenges of heterogeneity and openness, and proposes an approach based on the concept of location-based service, i.e. a service whose usage is associated with physical space, as a generic abstraction to support the development of location-dependent systems. The paper describes a model for associating location scopes with services, an architecture to support the discovery of location-based services on the Internet, and a prototype infrastructure in which several services and applications have been developed for validating the architecture.

Using GSM CellID Positioning for Place Discovering

Mobile phones can be used not only for voice and data communications but also as a computing device running context-aware applications. In this paper we present a model that, based on GSM cell identification, identifies places visited by a user and provides a user familiarity level for each of these places. This information can be used by context-aware applications to adapt their behaviour accordingly to the knowledge its user has about the current location. The achieved results are assessed by overlapping the discovered places with manual collected data, showing that GSM cellID positioning data can be used to identify places that are closer to each other than the average cell radius

The Smartphone-Based Offline Indoor Location Competition at IPIN 2016: Analysis and Future Work

This paper presents the analysis and discussion of the off-site localization competition track, which took place during the Seventh International Conference on Indoor Positioning and Indoor Navigation (IPIN 2016). Five international teams proposed different strategies for smartphone-based indoor positioning using the same reference data. The competitors were provided with several smartphone-collected signal datasets, some of which were used for training (known trajectories), and others for evaluating (unknown trajectories). The competition permits a coherent evaluation method of the competitors’ estimations, where inside information to fine-tune their systems is not offered, and thus provides, in our opinion, a good starting point to introduce a fair comparison between the smartphone-based systems found in the literature. The methodology, experience, feedback from competitors and future working lines are described.

Removing useless APs and fingerprints from WiFi indoor positioning radio maps

Maintaining consistent radio maps for WiFi fingerprinting-based indoor positioning systems is an essential step to improve the performance of the positioning engines. The radio maps consist of WiFi fingerprints collected at a predefined set of positions/places within a positioning area. Each fingerprint consists of the identification and radio signal level of the surrounding Access Points (APs). Due to the wide proliferation of WiFi networks, it is very common to observe 10 to 20 APs at a single position and more than 50 APs across a single building. However, in practical, not all of the detected APs are useful for the position estimation process. Some of them might have weak signals at certain positions or might have less significance for a positions fingerprint. Thus, those useless APs will add additional computational overheads during the position estimation, and consequently they will reduce the overall performance of the positioning engines. A similar phenomenon also occurs with some of the collected fingerprints. While it is widely accepted that the larger and more detailed the radio map is, the better is the accuracy of the positioning system, we found that some of the fingerprint samples on the radio maps do not contribute significantly to the estimation process. In this paper, we propose two methods for filtering the positioning radio maps: APs filtering and Fingerprints filtering. Then we report on the results of a set of experiments that have been done to evaluate the performance of a WiFipositioning radio map before and after applying the filtering approaches. The results show that there is possibility to simplify the radio maps of the positioning engines without significant degradation on the positioning precision and accuracy, and therefore to reduce the processing time for estimating the position of a tracked WiFi tag. This result has an important impact on increasing the number of tags a single instance of a WiFi positioning engine can handle at a time.

Wi-Fi fingerprinting in the real world - RTLS@UM at the EvAAL competition

Research and development around indoor positioning and navigation is capturing the attention of an increasing number of research groups and labs around the world. Among the several techniques being proposed for indoor positioning, solutions based on Wi-Fi fingerprinting are the most popular since they exploit existing WLAN infrastructures to support software-only positioning, tracking and navigation applications. Despite the enormous research efforts in this domain, and despite the existence of some commercial products based on Wi-Fi fingerprinting, it is still difficult to compare the performance, in the real world, of the several existing solutions. The EvAAL competition, hosted by the IPIN 2015 conference, contributed to fill this gap. This paper describes the experience of the RTLS@UM team in participating in track 3 of that competition.

A realistic evaluation of indoor positioning systems based on Wi-Fi fingerprinting: The 2015 EvAAL–ETRI competition

This paper presents results from comparing different Wi-Fi fingerprinting algorithms on the same private dataset. The algorithms where realized by independent teams in the frame of the off-site track of the EvAAL-ETRI Indoor Localization Competition which was part of the Sixth International Conference on Indoor Positioning and Indoor Navigation (IPIN 2015). Competitors designed and validated their algorithms against the publicly available UJIIndoorLoc database which contains a huge referenceand validation data set. All competing systems were evaluated using the mean error in positioning, with penalties, using a private test dataset. The authors believe that this is the first work in which Wi-Fi fingerprinting algorithm results delivered by several independent and competing teams are fairly compared under the same evaluation conditions. The analysis also comprises a combined approach: Results indicate that the competing systems where complementary, since an ensemble that combines three competing methods reported the overall best results.

Large scale movement analysis from WiFi based location data

Understanding and modeling the way humans move in urban contexts is beneficial for many applications. The recent advances on positioning technologies, namely those based on the ubiquity of wireless networks, is facilitating the observation of people for human motion analysis. In this paper we present the result of a large scale work conducted to study the human mobility in a Universitys campuses. The study was conducted along several months, using data collected from thousands of users that freely moved inside the numerous buildings existent in two University campuses and a few other buildings in the city center. A Wi-Fi infrastructure of more than 550 access points provides Internet access to the academic community. We tracked the user movements by logging the devices connected to each access point. Based on that data, an analysis process that highlights the relationships between space features and human motion has been developed. In this paper we introduce the concepts of “place connectivity” and “flow across a boundary” to model these relationships. Results show the mobility patterns detected, which are the attraction places along the day, and what places are more strongly connected. This paper also includes an analysis of the short and long term movements between places. With this study we extended our understanding of the life in the campus, enabling us to feel the campus “pulse”.

Navigation based on symbolic space models

Existing navigation systems are very appropriate for car navigation, but lack support for convenient pedestrian navigation and cannot be used indoors due to GPS limitations. In addition, the creation and the maintenance of the required models are costly and time consuming, and are usually based on proprietary data structures. In this paper we describe a navigation system based on a human inspired symbolic space model. We argue that symbolic space models are much easier to create and to maintain, and that they can support routing applications based on self-locating through the recognition of nearby features. Our symbolic space model is supported by a federation of servers where the spatial descriptions are stored, and which provide interfaces for feeding and querying the model. Local models residing in different servers may be connected between them, thus contributing to the system scalability.

A flexible location-context representation

Ubiquitous computing and the development of context-aware applications have been limited by the lack of open and generic solutions. We propose a flexible location-context representation which supports data acquired through multiple sensors represented in different space models.