Publications

Vulnerable Code Detection Using Software Metrics and Machine Learning.

Medeiros N., Ivaki N., Costa P., Vieira M.

IEEE Access

2020

Journal

Abstract:

Software metrics are widely-used indicators of software quality and several studies have shown that such metrics can be used to estimate the presence of vulnerabilities in the code. In this paper, we present a comprehensive experiment to study how effective software metrics can be to distinguish the vulnerable code units from the non-vulnerable ones. To this end, we use several machine learning algorithms (Random Forest, Extreme Boosting, Decision Tree, SVM Linear, and SVM Radial) to extract vulnerability-related knowledge from software metrics collected from the source code of several representative software projects developed in C/C++ (Mozilla Firefox, Linux Kernel, Apache HTTPd, Xen, and Glibc). We consider different combinations of software metrics and diverse application scenarios with different security concerns (e.g., highly critical or non-critical systems). This experiment contributes to understanding whether software metrics can effectively be used to distinguish vulnerable code units in different application scenarios, and how can machine learning algorithms help in this regard. The main observation is that using machine learning algorithms on top of software metrics helps to indicate vulnerable code units with a relatively high level of confidence for security-critical software systems (where the focus is on detecting the maximum number of vulnerabilities, even if false positives are reported), but they are not helpful for low-critical or non-critical systems due to the high number of false positives (that bring an additional development cost frequently not affordable).

Water Based Magnification of Capacitive Proximity Sensors: Water Containers as Passive Human Detectors

Rocha R. , Almeida A., Tavakoli M.

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

2020

Article

Abstract:

Sensors that detect human presence received an increasing attention due to the recent advances in smart homes, collaborative fabrication cells, and human robot interaction. These sensors can be used in collaborative robot cells and mobile robots, in order to increase the robot awareness about the presence of humans, in order to increase safety during their operation. Among proximity detection systems, capacitive sensors are interesting, since they are low cost and simple human proximity detectors, however their detection range is limited. In this article, we show that the proximity detection range of a capacitive sensor can be enhanced, when the sensor is placed near a water container. In addition, the signal can pass trough several adjacent water containers, even if they are separated by a few centimeters. This phenomenon has an important implication in establishing low cost sensor networks. For instance, a limited number of active capacitive sensor nodes can be linked with several simple passive nodes, i.e. water containers, to detect human or animal proximity in a large area such as a farm, a factory or home. Analysis on the change of the maximum proximity range with sensor dimension, container size and liquid filler was performed in order to study this effect. Examples of application are also demonstrated.

Waves in a thin and periodically oscillating medium

Ferreira R., Mascarenhas M.L.

Comptes Rendus Mathematique

2008

Article

Abstract:

We study the asymptotic behavior of the spectrum of an elliptic operator with periodically oscillating coefficients, in a thin domain, with vanishing Dirichlet conditions. Two cases are treated: the case where the periodicity of the oscillations and the thickness of the domain have the same order of magnitude and the case where the oscillations have a frequency much greater than the thickness of the domain. A physical motivation can be to understand the behavior of the probability density associated to the wave function of a particle confined to a very thin domain, with periodically varying characteristics

Weakly Supervised Multi-task Learning for Concept-based Explainability

Belém C., Balayan V., Saleiro P., Bizarro P.

ICLR'2021

2021

Conference Paper

Abstract:

Cell detection and segmentation is fundamental for all downstream analysis of digital pathology images. However, obtaining the pixel-level ground truth for single cell segmentation is extremely labor intensive. To overcome this challenge, we developed an end-to-end deep learning algorithm to perform both single cell detection and segmentation using only point labels. This is achieved through the combination of different task orientated point label encoding methods and a multi-task scheduler for training. We apply and validate our algorithm on PMS2 stained colon rectal cancer and tonsil tissue images. Compared to the state-of-the-art, our algorithm shows significant improvement in cell detection and segmentation without increasing the annotation efforts.

Wearable and comfortable e-textile headband for long-term acquisition of forehead EEG signals

Carneiro M.; Almeida A.; Tavakoli M.

IEEE Sensors Journal

2020

Journal

Abstract:

Electroencephalography (EEG) has a wide range of applications in medical diagnosis, and novel form of Human Machine Interfaces (HMI) for controlling prosthetic implants, wheelchairs, and home appliances in various forms of paralysis. However, the current EEG setups are composed of many wires hanging down from the system, and individual electrodes that must be set manually, which is time-consuming. As a result, the overall system is neither comfortable, nor aesthetically appealing. Here, we introduce for the first time, a comfortable textile-based EEG headband system that is soft, conformal to the skin, and comfortable. We present materials and methods for fabrication of multi-layer stretchable e-textile, that interfaces the human epidermis from one side through printed electrodes, and interfaces a rigid PCB island on the second layer. We as well demonstrate a method that allows creation of VIAs (vertical interconnect access) between the layers, using a CO2 laser. All Electrodes are integrated into the headband, and thus there is no need for individual electrode placement, and individual wiring. By screen printing a home-made conductive stretchable ink, patient-specific EEG headbands can be tailor made considering the optimal positioning of the electrodes for each patient. We show that these printed electrodes benefit from a very low skin-electrode impedance, comparable to gold standard Ag/AgCl, or gold cup electrodes, thanks to the high surface area silver flakes used in this work. The e-textile headband interfaces with an EEG acquisition device that captures, amplifies, and transmits the data to an external mobile phone or a PC. Furthermore, the integrated amplification in the textile and the use of an EMF rejection layer on top of the electrodes were shown to reduce the unwanted EM noise that is picked up by the system. We as well show application of the developed headband for usage in Human Machine Interfaces and Sleep Data Acquisition. Altogether, this device is step toward wider use of EEG acquisition devices for daily-use applications.

Weight optimization for consensus algorithms with correlated switching topology

Jakovetic D., Xavier J., Moura J.M.F.

IEEE Transactions on Signal Processing

2010

Article

Abstract:

We design the weights in consensus algorithms for spatially correlated random topologies. These arise with 1) networks with spatially correlated random link failures and 2) networks with randomized averaging protocols. We show that the weight optimization problem is convex for both symmetric and asymmetric random graphs. With symmetric random networks, we choose the consensus mean-square error (MSE) convergence rate as the optimization criterion and explicitly express this rate as a function of the link formation probabilities, the link formation spatial correlations, and the consensus weights. We prove that the MSE convergence rate is a convex, nonsmooth function of the weights, enabling global optimization of the weights for arbitrary link formation probabilities and link correlation structures. We extend our results to the case of asymmetric random links. We adopt as optimization criterion the mean-square deviation (MSdev) of the nodes’ states from the current average state. We prove that MSdev is a convex function of the weights. Simulations show that significant performance gain is achieved with our weight design method when compared with other methods available in the literature.

What day-ahead reserves are needed in electric grids with high levels of wind power?

Mauch B., Apt J., Carvalho P.M.S., Jaramillo P.

Environmental Research Letters

2013

Article

What is the best achievable QoS for unicast routing in VANET?

Boban M., Misek G., Tonguz O.K.

2008 IEEE Globecom Workshops, GLOBECOM 2008

2008

Conference Paper

Abstract:

Significant efforts and studies were recently reported for enabling active safety, traffic management, and commercial applications in Vehicular Ad Hoc Networks (VANET), since these applications are the drivers of the recent surge in VANET research and development. However, very few research efforts considered analyzing the Quality of Service (QoS) metrics that will be available to these applications in VANET. Furthermore, although there are many proposed solutions for routing in VANET, it is still unclear as to what specific characteristics VANET routing protocols should possess, since none of the proposed solutions achieves optimum performance in both urban and highway, as well as sparse and dense environment. To shed light on these issues, in this paper we analyze some of the most important QoS metrics in VANET. Namely, we determine the upper performance bound for connection duration, packet delivery ratio, end-to-end delay, and jitter for unicast communication in typical highway and urban VANET environments. According to our results, delay and jitter in VANET would be adequate for most of the envisioned unicast-based applications, whereas the packet delivery ratio and connection duration might not meet the requirements for most unicast-based applications.

When Backscatter Communication Meets Vehicular Networks: Boosting Crosswalk Awareness

Pereira F., Sampaio H., Chaves R., Correia R., Luís M., Sargento S., Jordão M., Almeida L., Senna C., Oliveira A.S.R., Carvalho N.B.

IEEE Access Journal

2020

Article

Abstract:

The research of safety applications in vehicular networks has been a popular research topic in an effort to reduce the number of road victims. Advances on vehicular communications are facilitating information sharing through real time communications, critical for the development of driving assistance systems. However, the communication by itself is not enough to reach the most desired target as we need to know which safety-related information should be disseminated. In this work, we bring passive sensors and backscatter communication to the vehicular network world. The idea is to increase the driver (or vehicle) awareness regarding the presence of pedestrians in a crosswalk. Passive sensors and backscatter communication technologies are used for the pedestrians’ detection phase, while the vehicular network is used during the dissemination of the detection information to surrounding vehicles. The proposed solution was validated through end-to-end experimentation, with real hardware and in a real crosswalk with real pedestrians and vehicles, demonstrating its applicability.