Abstract
This paper presents the TEC4SEA research infrastructure created in Portugal to support research, development, and validation of marine technologies. It is a multidisciplinary open platform, capable of supporting research, development, and test of marine robotics, telecommunications, and sensing technologies for monitoring and operating in the ocean environment. Due to the installed research facilities and its privileged geographic location, it allows fast access to deep sea, and can support multidisciplinary research, enabling full validation and evaluation of technological solutions designed for the ocean environment. It is a vertically integrated infrastructure, in the sense that it possesses a set of skills and resources which range from pure conceptual research to field deployment missions, with strong industrial and logistic capacities in the middle tier of prototype production. TEC4SEA is open to the entire scientific and enterprise community, with a free access policy for researchers affiliated with the research units that ensure its maintenance and sustainability. The paper describes the infrastructure in detail, and discusses associated research programs, providing a strategic vision for deep sea research initiatives, within the context of both the Portuguese National Ocean Strategy and European Strategy frameworks.

Abstract
INESC TEC is strongly committed to become a center of excellence in maritime technology and, in particular, deep sea technology. The STRONGMAR project aims at creating solid and productive links in the global field of marine science and technology between INESC TEC and established leading research European institutions, capable of enhancing the scientific and technological capacity of INESC TEC and linked institutions, helping raising its staff's research profile and its recognition as a European maritime research center of excellence. The STRONGMAR project seeks complementarity to the TEC4SEA research infrastructure: on the one hand, TEC4SEA promotes the establishment of a unique infrastructure of research and technological development, and on the other, the STRONGMAR project intends to develop the scientific expertise of the research team of INESC TEC.

Abstract
When operating low cost autonomous vehicles, we are often faced with the need to apply integro-differential operators to numerical sequences. The need may arise in many different contexts, from vehicle navigation to data collection/estimation, and is strongly reinforced by constraints to the number of deployable sensors. Integration and, particularly, differentiation of discrete data sequences are, however, error prone operations. Even in the absence of noise, the traditional approaches introduce distortions and artifacts in the output data, mostly due to the mismatch between their underlying polynomial model and the spectral contents of the collected data. This article presents an alternative way to apply integro-differential operators to discrete data streams. The operators are applied in a strictly band-limited way, in both time and frequency, and no extraneous artifacts are introduced in the data. No assumptions or models are used, other than assuming that the original data stream was correctly sampled. As such, the procedure can be safely applied to data streams sampled at rates close to Nyquist, without the usual performance degradation.

Abstract

Abstract
The complexity of real-life transients, coupled with the incomplete (or absent) knowledge of their statistical structure or defining features has motivated the interest on the use of blind, data driven detection schemes. One such scheme, proposed by Jones and Sayeed, uses Time-Frequency distributions to implement sub-optimal quadratic detectors which, under certain conditions, approach the performance of optimal quadratic detectors. However, their use of Fisher's discriminants to obtain class separation has some drawbacks, which we solve by using a simple perceptron to obtain the discriminant. Also, more often than not, we will have a multiclass situation, implying the use of different Time-Frequency Distributions, each one of them tuned for a given class of transients. The different nature of these distributions (bias, type of cross-terms, time-frequency resolution, etc.) will hamper the performance of the algorithm, forcing the need for experimental validation of its heuristical aspects. These are the issues we will address. The algorithm will be applied to real data, and its performance investigated.

Abstract
Is there a limit to the maximum resolution one can achieve when representing the signal's energy in the Time-Frequency plane? Some authors sustain that such a limit exists, and ignoring it is the cause of the known difficulties with some joint Time-Frequency distributions; others maintain that there is no such limit. In this article, we propose to analyze the merits and demerits of the several existing approaches, and suggest further arguments one might wish to consider. This will take us to the conclusion that, both from a tool-specific and from a general information-theoretic point of view, there is, indeed, a lower limit on the achievable resolution, even though the expression for that limit can not be given by the traditional Heisenberg-Gabor relations.