Walter Balzano

V2V-EN - Vehicle-2-Vehicle Elastic Network

Vehicle2Vehicle (V2V) is an emerging and fruitful area of research widely investigated in IoT and in the driverless setting, in particular. It allows cars to establish ad-hoc networks to cooperate in order to share information and therefore improve the safety of the drivers. A fundamental aspect in V2V is the positioning accuracy. While in several outdoor scenarios modern Global Navigation Satellite Systems (GNSS) are sufficiently precise, they are less reliable in indoor situations. This is undesirable as often cars move from outdoor to indoor scenarios and vice versa (think for example to cars leaving a box, or entering a tunnel). For this reason large research effort has been recently devoted to integrate GNSS systems in order to improve indoor positioning accuracy.In this paper we present Vehicle-2-Vehicle Elastic Network, a novel metodology that allows accurate vehicular indoor positioning based on a dead-reckoning technique via an Indoor Inertial Navigation System, aided by an elastic graph generated by a V2V-WiFi system. The received signal strength (RSS) from nearby cars is detected, and is used to improve the location via RSS-to distance information crowdsourcing.

WiFACT -- Wireless Fingerprinting Automated Continuous Training

The increasing importance in ubiquitous computing and context-dependent information has led in the last years to a growing interest in location-based applications and services. A considerable market demand concentrates on indoor localization tasks. In this setting, WiFi fingerprinting is currently one of the most popular and widespread techniques as it provides reasonable positioning accuracy while being able to exploit, at the same time, existing wireless infrastructures. WiFi-fingerprinting systems mainly operate though two distinct phases: one initial, named training, in which signals are collected and one subsequent, named usage, in which the recorded data are used to localize users. While the usage phase is fast and effective, the training phase is time consuming. Moreover, to maintain a localization accuracy, the training needs to be repeated anytime the network structure changes. The latter may occur, for example, if an access point goes off-line or it is removed. In this paper, we propose a novel framework that allows for an automatic and continuous training in WiFi-fingerprinting systems, which is based on an opportune deployment of a WSN(Wireless Sensor Network). Precisely, the solution we propose allows for an efficient real-time updating of the database collecting the signals, without any human intervention.

Genomic comparison using Data Mining techniques based on a possibilistic fuzzy sets model

Current copiousness of genomic information stored in biological databases [Mar Albà, M., Lee, M., Pearl, D., Shepherd, F.M.G., Martin, A.J., Orengo, N., Kellam, C.A., 2001. P. VIDA: a virus database system for the organisation of virus genome open reading frames. Nuleic Acids Res. 133-136] makes ultimately feasible the proposal for an application of knowledge management aimed to discover general rules in subcellular phenomena. The goal of this work is primarily to discover relationships between genes by microarray analysis. The tools exploited come from clustering techniques and are mainly based on Knowledge Discovery in Databases (KDD) concepts [Fayyad, U., Piatetsky-Shapiro, G., Smyth, P., 1996. From data mining to knowledge discovery in databases. AI Magazine 17(3), 37-54]. Starting from a data set, each element can be represented by a characteristic matrix, which sums up all data attributes. In this case data mining is oriented to perform a Pattern Recognition of related sequences, hidden in databases [Hand, D.J., Nicholas, A., 2005. Heard finding groups in gene expression data. J. Biomed. Biotechnol. 215-225]. Following a bottom up approach, the next refinement is to compare retrieved data to gather similar features, by dedicated clustering algorithms [Kaufman, L., Rousseeuw, P.J., 1990. Finding groups in data. An Introduction to Cluster Analysis. John Wiley & Sons, New York; Forman, G., Zhang, B., 2000. Distributed Data clustering can be efficient and exact HP. Laboratories Palo Alto HPL-2000, p. 158], driven by fuzzy logic, allowing us to perceive by intuition a common denominator for various genomic families and to anticipate likely future developments.

A Logic-based Clustering Approach for Cooperative Traffic Control Systems

Cooperative traffic control-systems have helped in the last years to extend the electronic horizon of vehicles in such a way that drivers can be informed quite in advance about any unexpected event occurring along a road. In this area, Vehicular Ad Hoc Networks (VANETs) perform information exchanges between vehicles by exploiting decentralized structures, where each vehicle is equipped with On Board Units (OBU) that allow to broadcast signals. In this paper, we revisit this communication framework by using a centralized approach, i.e., we consider a central unit for the data processing. This approach, on the one hand, facilitates the clustering process, thus obtaining the mitigation of the “broadcast storm” phenomenon (typically affecting VANETs); on the other hand, it allows the collection, management, and storage of the most relevant data concerning roads. Along the paper we evaluate the benefits of using our model, as well as we discuss on a smartphone integration.

DiG-Park: A Smart Parking Availability Searching Method Using V2V/V2I and DGP-Class Problem

Localization systems for outdoor areas using GNSS based systems (like GPS, GLONASS and more) are nowadays widespread and commonly-available in devices like smartphones. For indoor localization, on the other hand, several methods has been proposed in the latest years, like inertial navigation systems (INS) or wireless positioning systems (WPS, like wireless trilateration or fingerprinting methods), but an always-available and reliable technology is still under research. Vehicle-2-vehicle and Vehicle-2-infrastructure technologies allow vehicles to communicate via wireless to issue security warnings and to share traffic information. In this paper we propose an alternative localization technique based on the presence of nearby vehicles and known fixed infrastructures like generic radio hotspots. We aim to improve localization accuracy of the user in indoor parking lots using a combination of wireless radio signal strenght-to-distance evaluation, V2V/V2I, and a DGP algorithm, allowing users to rapidly find a free parking spot. This grade of accuracy is mandatory in indoor car park scenario since the vehicles are very close one to another in comparison with normal road traffic.

PAM-SAD: Ubiquitous Car Parking Availability Model Based on V2V and Smartphone Activity Detection

GNSS based systems (like GPS or GLONASS) for outdoor localization are nowadays widespread in common smartphones. Vehicle-2-vehicle technology allows vehicles to communicate via wireless to share any kind of information and detect mutual distances via RSS-to-distance evaluation. Smartphones can be used to detect when an user switches between car driving, walking and several other kind of activities.

CoTracks: A New Lossy Compression Schema for Tracking Logs Data Based on Multiparametric Segmentation

A massive diffusion of positioning devices and services, transmitting and producing spatio-temporal data, raised space complexity problems and pulled the research focus toward efficient and specific algorithms to compress these huge amount of stored or flowing data. Co Tracks algorithm has been projected for a lossy compression of GPS data, exploiting analogies between all their spatio-temporal features. The original contribution of this algorithm is the consideration of the altitude of the track, an elaboration of 3D data and a dynamic vision of the moving point, because the speed, tightly linked to the time, is supposed to be one of the significant parameters in the uniformity search. Minimum Bounding Box has been the tool employed to group data points and to generate the key points of the approximated trajectory. The compression ratio, resulting also after a further Huffman coding, appears attractively high, suggesting new interesting developments of this new technique.

EENET: Energy Efficient Detection of NETwork Changes Using a Wireless Sensor Network

Contextual services are having increasing importance in nowadays literature due to current availability of always-connected personal devices like smartphones and tablets. It is therefore increasingly important to have an affordable and accurate way for user localization in both indoor and outdoor environments. In outdoor GPS or GLONASS are reliable, but indoor a valid and ubiquitous localization system is still under research. One of the most promising methodology is wireless fingerprinting, which exploits available access points infrastructure. It is composed of two distinct phases: one training phase during which the interesting area is monitored and recorded and a usage phase in which the recorded data is used for localization purposes. The usage phase is reliable and accurate, but the training phase is often time consuming (in particular for large areas) as it must be performed manually and may also need to be repeated in case of structural and environmental changes. In this paper we propose a novel framework which uses an appropriate Wireless Sensors Network allowing continuous training over time in order to achieve real-time updating of the fingerprinting database without any human interaction, while also aiming to reduce power consumption needed for training phase, determining a minimal set of sentinel, which are sensors able to detect network alterations and able to trigger RadioMap rescanning.

DGP Application for Support Traffic Information Systems in Indoor and Outdoor Environments

V2V and V2I systems are networks in which vehicles and roadside units are nodes able to broadcast environmental conditions over large distances. Each vehicle has an on-board unit (OBU) which may have several kind of sensors (satellite-based positioning systems and positional sensors like digital compass, accelerometers and gyroscopes), may be connected to the car control unit, and has a long-range wireless module (able to transmit up to 1000 m away).

LoDGP: A Framework for Support Traffic Information Systems Based on Logic Paradigm

Nowadays, many studies demonstrate that one of the main causes of traffic congestion is the search for available parking lots. In this paper, we want to exploit the VANETs potentialities in order to overcome this problem, with the aim of improving the traffic conditions, by helping drivers to easily identify free parking lots in a urban area. The challenge is try to use information exchange between vehicles, in order to communicate parking information. To this goal, we analyze the well known distance geometry problem (DGP), by adapting it to our situation. This problem cares about how, starting from a graph, we can locate its nodes according to the known distances between some pairs of them. In our situation, the idea is to deal with the DGP, by using an inverse procedure: we start from a network where each vehicle is located by a global positioning system (GPS), and then we want to obtain a corresponding graph which contains useful information about parking lots in a certain area. For this purpose, we divide a huge urban area in some smaller clusters, by exploiting road side units (RSUs) and we separate vehicles in three groups corresponding to three possible states they can assume within a cluster, associating a different color to each of them in the corresponding graph. We make clear this model by means of logic programming, which offers compactness and intuitively, allowing us abstracting from concrete structures and variables concerning a procedural language and to explain the features of this framework in an elegant way. To this end, we provide a Prolog algorithm for the graph configuration and evolution.