Nuno M. Garcia

Enhanced Just-in-Time: A New Resource Reservation Protocol for Optical Burst Switching Networks

This paper proposes a new resource reservation protocol for optical burst switching networks, called enhanced just-in-time (E-JIT). E-JIT intends to improve and optimize the traditional JIT protocol, keeping all the advantages of its simplicity in terms of implementation. E-JIT aims to improve data channel scheduling, reducing the period of time in which the data channel remains in reserved status, optimizing channel utilization and reducing the burst loss probability. A performance assessment of E-JIT is presented for mesh optical burst switching networks with 16 nodes. It is shown that E-JIT may lead to a better performance than JIT for small burst loss probabilities.

Object-oriented modeling and simulation of optical burst switching networks

Optical burst switching (OBS) is becoming an interesting technology for the optical Internet, since it does not need optical buffers like optical packet switching (OPS), and is capable of a better performance than optical circuit switching (OCS). Although OBS has been recently an object of intense research, it still raises a number of important questions. Due to the high costs of an OBS network infrastructure, simulators are a good choice for predicting the behavior of this kind of networks. In this paper, we describe the proposal, implementation and validation of a simulator for OBS networks. The simulator, named OBSim, mimics the behavior of OBS networks in an object-oriented approach.

On the Performance of Shortest Path Routing Algorithms for Modeling and Simulation of Static Source Routed Networks -- an Extension to the Dijkstra Algorithm

Shortest path routing algorithms, such as Dijkstras algorithm present an overload problem when used to define routes for ring topologies in networks that implement source routing. This paper presents the effects of Dijkstras shortest path routing in the simulation and modeling of static source routed networks, in particular we evaluated the effect of this routing scheme in the performance of Optical Burst Switched (OBS) networks. A new static shortest path algorithm is presented and its performance compared with the standard shortest path algorithm, using two new metrics. We propose the use of this routing algorithm in network simulators instead of standard Dijkstra, as it produces more symmetric and balanced routes over the network links, thus producing results that are closer to real networks which implement a more dynamic routing.

The Ethernet Frame Payload Size and Its Effect on IPv4 and IPv6 Traffic

The IPv4 traffic characteristics are deeply related to the 1500 byte limit, a consequence of the payload capacity of the Ethernet frame. This paper describes how this limit may impact future IPv6 traffic and tests some aspects of traffic behavior for new values. To assess this, real IPv4 traffic was transformed to IPv6 traffic, with several simple assumptions. It is showed that in some circumstances, the change from IPv4 to IPv6 causes an overhead in terms of number of transmitted bytes that is irrelevant when compared to the overhead of the number of transmitted packets. Conclusions point to the possible existence of increased routing and switching effort resulting from the shift to the IPv6 protocol. The paper also suggests that a wider limit could decrease significantly the number of generated packets.

Improving Activity Recognition Accuracy in Ambient-Assisted Living Systems by Automated Feature Engineering

Ambient-assisted living (AAL) is promising to become a supplement of the current care models, providing enhanced living experience to people within context-aware homes and smart environments. Activity recognition based on sensory data in AAL systems is an important task because 1) it can be used for estimation of levels of physical activity, 2) it can lead to detecting changes of daily patterns that may indicate an emerging medical condition, or 3) it can be used for detection of accidents and emergencies. To be accepted, AAL systems must be affordable while providing reliable performance. These two factors hugely depend on optimizing the number of utilized sensors and extracting robust features from them. This paper proposes a generic feature engineering method for selecting robust features from a variety of sensors, which can be used for generating reliable classification models. From the originally recorded time series and some newly generated time series [i.e., magnitudes, first derivatives, delta series, and fast Fourier transformation (FFT)-based series], a variety of time and frequency domain features are extracted. Then, using two-phase feature selection, the number of generated features is greatly reduced. Finally, different classification models are trained and evaluated on an independent test set. The proposed method was evaluated on five publicly available data sets, and on all of them, it yielded better accuracy than when using hand-tailored features. The benefits of the proposed systematic feature engineering method are quickly discovering good feature sets for any given task than manually finding ones suitable for a particular task, selecting a small feature set that outperforms manually determined features in both execution time and accuracy, and identification of relevant sensor types and body locations automatically. Ultimately, the proposed method could reduce the cost of AAL systems by facilitating execution of algorithms on devices with limited resources and by using as few sensors as possible.

A Roadmap to the Design of a Personal Digital Life Coach

The timely and exact detection of abnormal events on the life of someone is one of the cornerstone of for successful Ambient Assisted Living strategies, either because these events need to be avoided, or because it is necessary that these events occur as a mean to implement a program for training a particular lifestyle. Technology has given us a plethora of sensors and actuators that can be used to deploy an infrastructure that would allow a successful detection of daily living activities, but often this would imply to use technology intensive, privacy invading solutions that despite its efficiency would render them non-adoptable by users. In this paper, we will explore the opportunities and challenges that AAL poses in the field of identification of daily activities, and how such a solution can be designed as to be an useradoptable part of a monitoring and training of lifestyles.

A review of thermal methods and technologies for diabetic foot assessment

Temperature analysis has been considered as a complementary method in medical evaluation and diagnosis. Several studies demonstrated that monitoring the temperature variations of the feet of diabetic patients can be helpful in the early identification of diabetic foot manifestations, and also in changing behaviors, which may contribute to reducing its incidence. In this review, several and most used techniques for assessing the temperature of the feet are presented, along with original published work on specific applications in diabetic foot complications. A review of solutions and equipment that operate according to the temperature assessment techniques is also presented. Finally, a comparison between the various technologies is presented, and the authors share their perspective on what will be the state of affairs in 5 years.

Identification of Activities of Daily Living Using Sensors Available in off-the-shelf Mobile Devices: Research and Hypothesis

This paper presents a PhD project related to the identification of a set of Activities of Daily Living (ADLs) using different techniques applied to the sensors available in off-the-shelf mobile devices. This project consists on the creation of new methodologies, to identify ADLs, and to present some concepts, such as definition of the set of ADLs relevant to be identified, the mobile device as a multi-sensor system, review of the best techniques for data acquisition, data processing, data validation, data imputation, and data fusion processes, and creation of the methods for the identification of ADLs with data mining, pattern recognition and/or machine learning techniques. However, mobile devices present several limitations, therefore techniques at each stage have to be adapted. As result of this study, we presented a brief review of the state-of-the-art related to the several parts of a mobile-system for the identification of the ADLs. Currently, the main focus consists on the study for the creation of a new method based on the analysis of audio fingerprinting samples in some Ambient Assisted Living (AAL) scenarios.

Burst assembly with real IPv4 data – performance assessement of three assembly algorithms

This paper describes the generation of data bursts using real IPv4 data as input and compares the performance of the three aggregation algorithms. The profiles of bursts are studied, in particular the mean packet delay per burst and burst inter-arrival time followed by mean burst size and mean number of packets per burst. Observations are made regarding the identification of relations between bursts and packets for the studied data traces and assessed the performance of the aggregation algorithms under study. The conclusions are generalized to IPv6 traffic.

TICE.Healthy: A perspective on medical information integration

This paper describes a perspective on medical information integration as implemented by the Portuguese Program TICE.Healthy. The eight subprojects that integrate TICE.Healthy share a common database for medical data, and exchange data using standard formats. Because the subprojects deal with data of different nature, strategies were developed to allow the effective integration of this data. This paper describes the strategies adopted by the consortium, its discovered best practices and eventual pitfalls.