Abstract
We study the conditions for the formation of multiple regions of high activity or "bumps" in a one-dimensional, homogeneous neural field with localized inputs. Stable multi-bump solutions of the integro-differential equation have been proposed as a model of a neural population representation of remembered external stimuli. We apply a class of oscillatory coupling functions and first derive criteria to the input width and distance, which relate to the synaptic couplings that guarantee the existence and stability of one and two regions of high activity. These input-induced patterns are attracted by the corresponding stable one-bump and two-bump solutions when the input is removed. We then extend our analytical and numerical investigation to N-bump solutions showing that the constraints on the input shape derived for the two-bump case can be exploited to generate a memory of N > 2 localized inputs. We discuss the pattern formation process when either the conditions on the input shape are violated or when the spatial ranges of the excitatory and inhibitory connections are changed. An important aspect for applications is that the theoretical findings allow us to determine for a given coupling function the maximum number of localized inputs that can be stored in a given finite interval.

Abstract
Wireless sensor networks are notoriously difficult to program and debug. This fact not only stems from the nature of the hardware, but also from the current approaches for developing programming languages and runtime systems for these platforms. In particular, current systems do not place enough stress on providing formal descriptions of the language and its runtime system, and on proving static properties, like type-safety and soundness. In this paper, we present the design, specification, and implementation of a programming language and a runtime system for wireless sensor networks that are safe by design. We say this in the sense that we can statically detect a large set of would-be runtime errors, and that the runtime system will not incorrectly execute an application, once the latter is deployed. We have a full prototype implementation of the system that supports SunSPOT devices, the simulation tool VisualSense, and local computer networks for fast deployment and testing of applications. Development is supported by an IDE implemented on top of the Eclipse tool that embeds both the compiler and the virtual machine seamlessly, and is used to produce software releases.

Abstract

Abstract
This work describes a preliminary methodological framework for the assessment of coastal volcanic rocky platforms in Pico Island (Azores). This study also deals with the importance of GIS-based mapping and unmanned aerial vehicles (UAVs) whichwere tested in two studied areas (Lajes and Madalena sites). This research gives an overview about the UAVs work flow and its application for photogrammetric assessment of coastal volcanic rocky platforms. The main purpose of this study is to explore the influence of basaltic lava coastal platforms and associated boulder strewn, which are well differentiated in Lajes and Madalena sites. In general, this research describes the image acquisition process and the UAV LiDAR technology and the aerial surveys to acquire the geodatabase which will enable the design of geoscience and geotechnical maps on coastal volcanic rock platforms. The images analysis will allow: (i) boulder geotechnical description and evaluation; (ii) shore and coastal platform assessment; (iii) propose monitoring coastal plans (short to long-term); (iv) rock boulders (basaltic blocks) mobility, movement trend, imbrication, indicating flow and source direction; (v) obtain DTM and contour lines to generate 3D models of the coastal area. All these data is crucial to understand the coastal dynamics of the sites and develop an applied mapping which couples GIS technologies and UAV based spatial platform. It was also developed a GNSS (Global Navigation Satellite System) datasheet which incorporates the entire database of ground control points, using high resolution GPS system for the georeferencing and differential correction process. This research presents the preliminary results of the coastal geotechnics mapping for volcanic environments. Furthermore, high resolution image acquisition, georeferencing and differential correction of the ground control points using high accuracy GPS, were also thorough analysed to improve the general methodology presented herein. Finally, it was proposed a preliminary integrated coastal engineering study for the rocky platform zoning and short to long-term monitoring in selected sites on Pico Island. The study was partially financed by FEDEREU COMPETE Funds and FCT (GeoBioTec|UA: UID/GEO/04035/2013) and LABCARGA|ISEP reequipment program: IPP-ISEP|PAD’2007/08. © 2016 Taylor & Francis Group, London.

Abstract
We present and evaluate two different methods for building spatio-temporal features: a propositional method and a method based on propositionalisation of relational clauses. Our motivating application, a regression problem, requires the prediction of the fraction of each Portuguese parish burnt yearly by wildfires - a problem with a strong socio-economic and environmental impact in the country. We evaluate and compare how these methods perform individually and combined together. We successfully use under-sampling to deal with the high skew in the data set. We find that combining the approaches significantly improves the similar results obtained by each method individually.

Abstract
A considerable number of classic and modern control applications in distribution systems are developed aiming at make the most of the existing infrastructure. Under this perspective, investments postponement in grid upgrade appears as a good expectable result. The advent of smart grids enhances this expectation since it promises ubiquitous system monitoring, secure and trustworthy communication technologies and advanced control schemes. These features will allow the implementation of innovative algorithms for better operation and management of system assets. New possibilities arise for DSM, microgeneration control, electric vehicle charging and intelligent operation of energy storage devices. The scientific literature has countless examples that illustrate the benefits of such tools. A frequent outcome of these studies highlights the advantage of investments deferral in network reinforcement. This paper analyzes the combined effects of these control actions with the investments required for network upgrade, namely in lines and transformers reinforcement. Contrary to other research papers, our results demonstrate that investments deferral could lead frequently to higher costs in a long-term perspective.