Ilaria Sergi

A survey on indoor positioning systems

This paper aims to provide the reader with a review of the main technologies explored in the literature to solve the indoor localization issue. Furthermore, some systems that use these enabling technologies in real-world scenarios are presented and discussed. This could deliver a better understanding of the state-of-the-art and motivate new research efforts in this promising field. Finally, focusing on one of the major challenges in the indoor localization field, i.e., the indoor animal tracking, existing indoor tracking systems have been reviewed and compared by analyzing advantages and drawbacks.

Smart RFID Antenna System for Indoor Tracking and Behavior Analysis of Small Animals in Colony Cages

Radio frequency identification (RFID) technology is more and more adopted in a wide range of applicative scenarios. Nevertheless, in many applications, commercial and general-purpose solutions can be unsuitable as in the case of the tracking of small-size living animals for the behavior analysis. In such a case, the whole RFID hardware, as well as the control software, should be opportunely tailored for the particular application. In this paper, a novel RFID-based approach enabling an effective localization and tracking of small-sized laboratory animals is proposed. It is mainly based on a near-field (NF) RFID multiantenna system working in the UHF bandwidth, to be placed below the animals cage, and able to rigorously identify the NF RFID tags implanted in laboratory animals. Once the requirements of the reader antenna have been individuated, an antenna system composed of a matrix of specifically designed segmented loops has been prototyped. Moreover, to improve the effectiveness of the whole tracking system, a properly algorithm based on the measured received signal strength indication has been developed and integrated. It aims at removing potential minor ambiguities of the data captured by the multiantenna system. The whole animal tracking system has been then largely tested on phantom mice to verify its ability to precisely localize each subject and to reconstruct its path. Additionally, a first test performed on living mice has been presented. The achieved and discussed results demonstrate the effectiveness of the proposed approach.

Near Field UHF RFID Antenna System Enabling the Tracking of Small Laboratory Animals

Radio frequency identification (RFID) technology is more and more adopted in a wide range of applicative scenarios. In many cases, such as the tracking of small-size living animals for behaviour analysis purposes, the straightforward use of commercial solutions does not ensure adequate performance. Consequently, both RFID hardware and the control software should be tailored for the particular application. In this work, a novel RFID-based approach enabling an effective localization and tracking of small-sized laboratory animals is proposed. It is mainly based on a UHF Near Field RFID multiantenna system, to be placed under the animals’ cage, and able to rigorously identify the NF RFID tags implanted in laboratory animals (e.g., mice). Once the requirements of the reader antenna have been individuated, the antenna system has been designed and realized. Moreover, an algorithm based on the measured Received Signal Strength Indication (RSSI) aiming at removing potential ambiguities in data captured by the multiantenna system has been developed and integrated. The animal tracking system has been largely tested on phantom mice in order to verify its ability to precisely localize each subject and to reconstruct its path. The achieved and discussed results demonstrate the effectiveness of the proposed tracking system.

Web of Topics: An IoT-aware model-driven designing approach

In the Internet of Things, the extreme heterogeneity of sensors, actuators and user devices calls for new tools and design models able to translate the users needs in machine-understandable scenarios. The scientific community has proposed different solution for such issue, e.g., the MQTT (MQ Telemetry Transport) protocol introduced the topic concept as “the key that identifies the information channel to which payload data is published”. This study extends the topic approach by proposing the Web of Topics (WoX), a conceptual model for the IoT. A WoX Topic is identified by two coordinates: (i) a discrete semantic feature of interest (e.g. temperature, humidity), and (ii) a URI-based location. An IoT entity defines its role within a Topic by specifying its technological and collaborative dimensions. By this approach, it is easier to define an IoT entity as a set of couples Topic-Role. In order to prove the effectiveness of the WoX approach, we developed the WoX APIs on top of an EPCglobal implementation. Then, 10 developers were asked to build a WoX-based application supporting a physics lab scenario at school. They also filled out an ex-ante and an ex-post questionnaire. A set of qualitative and quantitative metrics allowed measuring the models outcome.

Performance evaluation of end-to-end security protocols in an Internet of Things

Wireless Sensor Networks are destined to play a fundamental role in the next-generation Internet, which will be characterized by the Machine-to-Machine paradigm, according to which, embedded devices will actively exchange information, thus enabling the development of innovative applications. It will contribute to assert the concept of Internet of Things, where end-to-end security represents a key issue. In such context, it is very important to understand which protocols are able to provide the right level of security without burdening the limited resources of constrained networks. This paper presents a performance comparison between two of the most widely used security protocols: IPSec and DTLS. We provide the analysis of their impact on the resources of embedded devices. For this purpose, we have modified existing implementations of both protocols to make them properly run on our hardware platforms, and we have performed an extensive experimental evaluation study. The achieved results are not a consequence of a classical simulation campaign, but they have been obtained in a real scenario that uses software and hardware typical of the current technological developments. Therefore, they can help network designers to identify the most appropriate secure mechanism for end-to-end IP communications involving constrained devices.

An ambient assisted living system for elderly assistance applications

In last few years, the number of elderly people is more and more increasing. Consequently, various efforts were made in order to improve the quality of lifestyle for the elderly. Related to this issue, an Ambient Assisted Living (AAL) works to create better living conditions for older or disable people. AAL systems are able to continuously monitor the health status of the elderly through data coming from heterogeneous sensors. In this paper, we propose an AAL architecture for the monitoring of elderly people, able to guarantee the collection of heterogeneous sensor data as well as the detection of critical events such as the older adult fall. A remote reasoning system processes this data with the aim of generating appropriate events. In order to demonstrate the feasibility of the proposed architecture, a proof of concept was used and validation functional tests were carried out.

An animal tracking system for behavior analysis using radio frequency identification.

Evaluating the behavior of mice and rats has substantially contributed to the progress of research in many scientific fields. Researchers commonly observe recorded video of animal behavior and manually record their observations for later analysis, but this approach has several limitations. The authors developed an automated system for tracking and analyzing the behavior of rodents that is based on radio frequency identification (RFID) in an ultra-high-frequency bandwidth. They provide an overview of the systems hardware and software components as well as describe their technique for surgically implanting passive RFID tags in mice. Finally, the authors present the findings of two validation studies to compare the accuracy of the RFID system versus commonly used approaches for evaluating the locomotor activity and object exploration of mice.

An RFID tracking system supporting the behavior analysis of colonial laboratory animals

Animal tracking and animal behavior analysis have a crucial impact in biomedical disciplines to study new pathologies and effects of new drugs. There are several solutions, based on different technologies such as GPS, radar, and vision, designed to obtain animals tracking systems, but they are effective mainly in presence of large size animals and outdoor environment. Unfortunately, they show poor performance when groups of small laboratory animals have to be monitored in indoor environments. In such a context, the adoption of passive Near Field (NF) Ultra High Frequency (UHF) Radio Frequency Identification (RFID) technology seems to be a winning approach, even though the straightforward use of commercial solutions does not guarantee satisfactory performance. Specifically, customized hardware and software solutions are then required. The main goal of this work is to present the development and then to validate a reliable and effective system for the automatic tracking of laboratory mice, based on suited NF UHF RFID hardware capturing system combined with an ad hoc software system able to guarantee hardware control, data processing, and reporting. In particular, the validation phase has been carried out by selecting the most appropriate RFID tags and by surgically implanting them into laboratory mice. Experimental results have demonstrated the efficiency of the proposed solution, which is able to gather data on the animal movements, allowing their subsequent processing for a satisfactory behavioral analysis.

A Complex Event Processing based smart aid system for fire and danger management

When hazardous events occurs in buildings or in large environments with different access points and with a large number of users, rescue workers (firefighters, first aid workers, civil protection teams, etc.) need to intervene in a timely manner, where there is a certainty that there are users to help. Topically such events require avoiding waste of resources in environments where there are no people at the time of the disaster or where the damage is of low magnitude. To guide rescuers at the points of the building where there are users to help, we modeled and built an Internet of Things-based framework that monitors data and environmental parameters of interest and, if certain thresholds are exceeded, alerts the rescuers through a telephone call to emergency numbers. The hardware infrastructure is driven by a complex, flexible and adaptive software layer that behaves depending on a Complex Event Processing engine and a reflective middleware according to the rule based engine that manages data from the sensors and reasoning mechanisms of a knowledge base that models the given domain.

An innovative approach for monitoring elderly behavior by detecting home appliance's usage

Detecting how elderly people interact with their surrounding environment, especially with domestic electrical appliances, is an important parameter to assess Mild Cognitive Impairments and frailty issues. This paper proposes an innovative approach for monitoring elderly behavior by detecting home appliances usage. It is based on an unobtrusive smart meter that periodically measures the global power consumption in the house, associated with some smart plugs for punctually monitoring specific electrical devices. This infrastructure has been implemented and validated within the Personal Data Capturing System of the City4Age Platform, where, joined with other provided monitoring systems, can feed risk detection algorithms with more accurate data. Summarizing, implemented system, although simple and at low cost, is able to combine data provided by designed power meter with those of smart meter plugs and, by means of implemented algorithms, to detect unusual elder behavior, moreover resulting reliable and accurate.