The most common machine-learning methods solve supervised and unsupervised problems based on datasets where the problem’s features belong to a numerical space. However, many problems often include data where numerical and categorical data coexist, which represents a challenge to manage them. To transform categorical data into a numeric form, preprocessing tasks are compulsory. Methods such as one-hot and feature-hashing have been the most widely used encoding approaches at the expense of a significant increase in the dimensionality of the dataset. This effect introduces unexpected challenges to deal with the overabundance of variables and/or noisy data. In this regard, in this paper we propose a novel encoding approach that maps mixed-type data into an information space using Shannon’s Theory to model the amount of information contained in the original data. We evaluated our proposal with ten mixed-type datasets from the UCI repository and two datasets representing real-world problems obtaining promising results. For demonstrating the performance of our proposal, this was applied for preparing these datasets for classification, regression, and clustering tasks. We demonstrate that our encoding proposal is remarkably superior to one-hot and feature-hashing encoding in terms of memory efficiency. Our proposal can preserve the information conveyed by the original data.

Wireless body area networks (WBANs) can enable e-health applications under Internet of Things (IoT) scenarios. However, to use WBAN technologies in practical applications, sensitive data collected by wireless sensors must be protected when transmitted across a network and until accessed by authorized applications or end-users. Specifically, it is necessary to provide confidentiality, integrity, authentication and access control in WBANs. This paper presents a security approach to provide these security services in a layered WBAN system using lightweight cryptography. Layer 1 consists of the communication between the sensor nodes and the base station (data acquisition); Layer 2 involves the communication between the base station and a data repository (data storage); and Layer 3 deals with the communication of end-users to the repository (data access). In the past, security has focused only on Layer 1 and for limited security levels. In this paper, security concerns in the three layers of a WBAN system are studied and addressed. As primary contributions, the design details of a secure WBAN system prototype and the impact of lightweight cryptographic engines on the performance of the primary use cases in the WBAN system are highlighted from data acquisition until data use. We present a novel WBAN system prototype that ensures most of the required security services for standard security levels.

The tracking of frequently visited places, also known as stay points, is a critical feature in location-aware mobile applications as a way to adapt the information and services provided to smartphones users according to their moving patterns. Location based applications usually employ the GPS receiver along with Wi-Fi hot-spots and cellular cell tower mechanisms for estimating user location. Typically, fine-grained GPS location data are collected by the smartphone and transferred to dedicated servers for trajectory analysis and stay points detection. Such Mobile Cloud Computing approach has been successfully employed for extending smartphone’s battery lifetime by exchanging computation costs, assuming that on-device stay points detection is prohibitive. In this article, we propose and validate the feasibility of having an alternative event-driven mechanism for stay points detection that is executed fully on-device, and that provides higher energy savings by avoiding communication costs. Our solution is encapsulated in a sensing middleware for Android smartphones, where a stream of GPS location updates is collected in the background, supporting duty cycling schemes, and incrementally analyzed following an event-driven paradigm for stay points detection. To evaluate the performance of the proposed middleware, real world experiments were conducted under different stress levels, validating its power efficiency when compared against a Mobile Cloud Computing oriented solution.

Vehicular ad-hoc Networks (VANETs) are recognized as a cornerstone of Intelligent Transportation Systems (ITS) to enable the exchange of information among vehicles, which is crucial for the provision of safety-related and entertainment applications. However, practical useful realizations of VANETs are still missing, mainly because of the elevated costs and the lack of a final standardization. In this regard, the feasibility of using smartphones as nodes in VANETs has been explored focusing on small-scale deployments to mainly validate single-hop communication capabilities. Moreover, existing smartphone-based platforms do not consider two crucial requirements in VANETs, namely, multi-hop communication and the provision of security services in the message dissemination process. Furthermore, the problem of securing message dissemination in VANETs is generally analyzed through simulation tools, while performance evaluations on smart devices have not been reported so far. In this paper, we aim to fill this void by designing a fully on-device platform for secure multi-hop message dissemination. We address the multi-hop nature of message dissemination in VANETs by integrating a location-based protocol that enables the selection of relay nodes and retransmissions criteria. As a main distinction, the platform incorporates a novel certificateless cryptographic scheme for ensuring data integrity and nodes’ authentication, suitable for VANETs lacking of infrastructure.

Mobile Edge Computing (MEC) relates to the deployment of decision-making processes at the network edge or mobile devices rather than in a centralized network entity like the cloud. This paradigm shift is acknowledged as one key pillar to enable autonomous operation and self-awareness in mobile devices in IoT. Under this paradigm, we focus on mobility-based services (MBSs), where mobile devices are expected to perform energy-efficient GPS data acquisition while also providing location accuracy. We rely on a fully on-device Cognitive Dynamic Systems (CDS) platform to propose and evaluate a cognitive controller aimed at both tackling the presence of uncertainties and exploiting the mobility information learned by such CDS toward energy-efficient and accurate location tracking via mobility-aware sampling policies. We performed a set of experiments and validated that the proposed control strategy outperformed similar approaches in terms of energy savings and spatio-temporal accuracy in LBS and MBS for smartphone devices.

In this paper we investigate the impact of data uncertainty on the prediction of vital signs abnormalities. We analyze the evolution of mean arterial blood pressure (MAP) and heart rate (HR) vital signs time series over a given forecasting window, whose values are classified according to a model of normality that models patient’s physiological deterioration. To this end, we developed a generic prediction framework that enables the deployment of different methods for data preparation, training, prediction, and classification so as to perform a set of experiments. To address our purpose, we selected a general regression neural network model (to predict vital sign trends) and a support vector machine (to recognize abnormalities) as the appropriate methods for a baseline analysis. However, other algorithm choices could be tailored in the defined prediction framework. We extract from the Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC II) waveform database a set of 302,940 vital sign samples (organized as 459 records) to define training and testing datasets. From our experiments, we obtained evidence demonstrating the incidence of null values in the MAP/HR input time series on forecasting tasks. This was performed under different conditions in terms of null values in the observation window, and varying other factors like labeling methods and forecast window sizes. In our experiments, we concentrate on analyzing the extent of prediction performance under such conditions, and identify that a tolerable level of null values is around 25% where it is possible to still derive forecast windows with an accuracy target ≥0.9 for both MAP and HR vital signs. The presented evaluation contributes to acknowledge the importance of uncertainty conditions on the prediction performance of vital signs, and paves the way for future extensions involving other prediction approaches.