Rute C Sofia

— We are living in the era of ubiquitous mobile phone development, devices which are already equipped with a variety of sensors. Such sensors are being used to assist the most varied aspects of our daily routines, including physical health conditions and proactive and ubiquitous personal management. In this PhD, we are proposing to conceive, and to validate a non-intrusive framework for inference of social interaction based on personal devices such as but not withstanding to, smartphones.

—Mobile sensing has been gaining ground due to the increasing capabilities of mobile, personal devices that are carried around by citizens, giving access to a large variety of data and services, all based on the way humans interact. Nevertheless, pervasive platforms used to capture and to infer human interaction are still designed in a simplistic way, aspect which prevents them to adequately scale. This paper contributes by bringing awareness into the challenges faced by mobile sensing platforms that capture and perform learning based on human interaction; how current solutions perform activity recognition, which classification models they consider, and which type of behavior inference can be seamlessly provided as of today. The paper contributes to raise awareness to current challenges faced by these platforms, and provides a set of guidelines towards a better functional design.

—Information-centric networking integrates by design a pull-based model which brings in advantages in terms of control as well as of in-network caching strategies. Currently, ICN main areas of action concern content distribution and IoT, both of which are environments that often require support for periodic and even-triggered data transmission. Such environments can benefit from push-based communication to achieve faster data forwarding. This paper describes mechanisms to implement push-based communication in information-centric paradigms, explaining the trade-off associated with the different approaches. The paper provides design guidelines for integrating push communications in information-centric networking, having as example the application of this networking architecture in IoT environments.

Information-centric networking architectural design
make it prone to support environments attaining a high-degree of mobility: ICN is content-oriented, connection-less, and receiver-driven. Still, mobility management support in ICN requires further tuning. Solutions being developed often consider the principles of mobility management in IP-based networks, in an attempt to achieve interoperability. The downside of this is endangering the advantages that are naturally provided by a networking architecture that is content-centric. This paper provides an overview on how ICN paradigms handle mobility management, highlighting current challenges and proposing guidelines to overcome them, thus steering a vision towards a content-centric mobility management approach.

This paper contributes to the topic of mobility management anticipation via the exploitation of preferred network ranking utility functions. The work proposes, implements, and evaluates a set of ranking functions based on Quality of Service and Quality of Experience indicators. Such indicators are obtained via wireless overhearing, or via active network probing. The functions also take into consideration recommendations from neighboring nodes. The implementation of the functions has been integrated into open-source software. The proposed implementation is evaluated via experiments carried out on a wireless testbed in terms of handover time to completion, total handovers successfully achieved, and rejected handovers. Results obtained show that overhearing is relevant to estimate mobility, as it brings in more sensitivity and lowers the cost of active probing.

Activity recognition is an increasingly relevant topic in the context of the most varied end-user services. In outdoor environments , activity recognition based on close-to-real-time information is key in providing awareness to the user about their surroundings in a timely and user-friendly manner, thus allowing to the user to improve its overall use (Quality of Experience). In this context, it is relevant to understand how data extracted from multiple sensors can be fused, interpreted and classified, to best provide feedback to the user. Having as target case Mobile Augmented Reality Systems for outdoor environments, this paper presents a first analysis on how smart data captured via multiple sensors can assist activity recognition and adequate feedback to the user. The paper also debates the existent restrictions imposed by ap-plications' usage in these environments, describing possible use scenarios and presenting results of an experiment for discriminating activities when using common sensors, such as the accelerometer.

This paper contributes to the topic of mobility management anticipation via the exploitation of preferred network ranking utility functions. The work proposes, implements, and evaluates a set of ranking functions based on Quality of Service and Quality of Experience indicators. Such indicators are obtained via wireless overhearing, or via active network probing. The functions also take into consideration recommendations from neighboring nodes. The implementation of the functions has been integrated into open-source software. The proposed implementation is evaluated via experiments carried out on a wireless testbed in terms of handover time to completion, total handovers successfully achieved, and rejected handovers. Results obtained show that overhearing is relevant to estimate mobility, as it brings in more sensitivity and lowers the cost of active probing.

—This technological demonstration goes over a new trust management framework which can assist in achieving secure connections in wireless networks, without necessarily implying the use of strong security associations. The demonstration is based on open-source software, and has been developed to run on OpenWRT as well as on Android.

This paper describes a software-based tool that tracks mobile
node roaming and infers the time-to-handover as well as the preferential
handover target, based on behavior inference solely derived from regular
usage data captured in visited wireless networks. The paper presents
the tool architecture; computational background for mobility estimation;
operational guidelines concerning how the tool is being used to track
several aspects of roaming behavior in the context of wireless networks.
Target selection accuracy is validated having as baseline traces obtained
in realistic scenarios.

This paper provides an interdisciplinary perspective concerning the role of prosumers on future Internet design based on the current trend of Internet user empowerment. The paper debates the prosumer role, and addresses models to develop a symmetric Internet architecture and supply-chain based on the integration of social capital aspects. It has as goal to ignite the discussion concerning a socially-driven Internet architectural design.

Wireless networks are today employed as com-
plementary access technology, implemented on the last hop
towards the Internet end-user. The shared media that wireless
deployments provide and which is relevant to interconnect
multiple users, has a limited technical design, as only one
device can be served per unit of time, design aspect which
limits the potential applicability of wireless in dense environ-
ments. This paper proposes and evaluates a novel MAC Layer
mechanism that extends current wireless networks with the
possibility to perform downstream transmission to multiple
devices within a single transmission time-frame, resulting in
improved fairness for all devices. The mechanism, which is
software defined, is backward compatible with current wireless
standards and does not require any hardware changes. The
solution has been validated in a realistic testbed and the paper
provides details concerning the computational aspects of our
solution; a description of the implementation; results extracted
under different realistic scenarios in terms of throughput,
packet loss, as well as jitter.

Pervasive computing is today backed up by millions of personal devices, which track the most varied aspects of a citizen's online routine, usually backed up by cloud-based systems, that have been engineered to deal with large amounts of raw data. It is, however, feasible to perform behavior characterization in a non-intrusive and pervasive way, derived from the natural wireless footprint that mobile devices leave on the network. In order to assist this line of thought this paper provides a characterization study of such wireless footprint , derived from traces obtained from a non-intrusive sensing tool that resides solely on the end-user device. We give insight into the dynamics of both individual and correlated wireless footprints in time and space.

ABSTRACT Existing Quality of Service models are well defined in the data path, but lack an end-to-end control path mechanism that guarantees the required resources to bandwidth intensive services, such as video streaming. Current reservation protocols provide scalable resource reservation inside routing domains. However, it is primarily between such domains that scalability becomes a major issue, since inter-domain links experience large volumes of reservation requests. As a possible solution, we present and evaluate the Shared-segment based Inter-domain Control Aggregation Protocol, (SICAP) which affords the benefits of shared-segment aggregation, while avoiding its major drawback, namely, its sensitivity to the intensity of requests [1]. We present results of simulations that compare the performance of SICAP against that of the Border Gateway Reservation Protocol, (BGRP) which relies on sink-tree aggregation to achieve scalability.

ABSTRACT Current mobility management approaches deploy heavy centralized models that are not prepared to deal with the requirements of growing mobile environments, leading to suboptimal routing and scalability problems. As a first step towards a decoupled and distributed mobility management architecture, we propose and evaluate different approaches for the distribution of the mobility management functionalities. We initiate our mobility decoupling from the most common split into data and control planes. We go further in splitting the control plane of mobility management into location and handover management. We evaluate the distributed approaches, based on the proposed decoupling, and we compare them with the most adopted fully-centralized approaches. The results of the evaluation demonstrate that the distribution of mobility management functionalities through elements closer to the end-user improve both user and network performance, even in a hierarchical network topology.

Abstract—This paper discusses and validates energy-aware routing metrics which can be applied to any available routing protocol. We validate the metrics performance based on the two main branches of multihop routing, namely, link-state and distance-vector approaches. The validation is performed based upon discrete event simulations.

This paper addresses the problematic of mobility
management decomposition, as a first step towards the devel-
opment of decentralized mobility management architectures.
We propose and evaluate an IP mobility management solution
based on the decoupling of data and control plane. We have
carried out simulations, showing that the proposed approach
can significantly reduce packet loss and reachability time for
mobile nodes

Ethernet’s plug-&-play feature is built on its use of flat (location independent) addresses and use of broadcasts to resolve unknown MAC addresses. While plug-&-play is one of Ethernet’s most attractive features, it also affects its scalability. As the number of active MAC addresses in the network grows beyond the capacity of forwarding caches in bridges, the odds of “cache-
misses,” each triggering a broadcast, grow as well. The resulting increase in broadcast bandwidth consumption affects scalability.
To address this problem, we propose a simple address resolution scheme based on an adaptation of distributed hash tables where a single query suffices in the steady state. The new scheme is implemented on advanced bridges maintaining backward compatibility with legacy bridges and eliminating reliance on broadcasts for address discovery. Comparisons with a legacy, broadcast-based scheme are carried out along several metrics that demonstrate the new scheme’s robustness and ability to improve scalability.

Mobility management is a key aspect to consider in future Internet architectures, as these architectures include a highly nomadic end-user which often relies on services provided by multi-access networks.
In contrast, today’s mobility management solutions were designed having in mind simpler scenarios and requirements from the network and where roaming could often be taken care of with previously established agreements. With a more dynamic behavior in the network, and also with a more prominent role from the end-user, mobility management has to deal with additional requirements derived from new Internet paradigms. To assist in understanding such requirements and also how to deal with them, this paper proposes a starting point to dismantle current mobility management notions. Our contribution is an initial proposal on defining mobility management in concrete functional blocks, their interaction, as well as a potential grouping which later can assist in deriving novel and more flexible mobility management architectures.

Current Quality of Service models such as those embodied in the Differentiated Services proposal,rely on data path aggregation to achieve scalability. Data path aggregation bundles into a single aggregate multiple flows with the same quality requirements, hence decreasing the amount of state to be kept. A similar scalability concern exists on the control path, where the state required to account for individual reservations needs to be minimized. There have been several proposals aimed at control path aggregation, and the goal of this paper is to expand on these works in an attempt to gain a better understanding of the various parameters that influence the efficiency of different approaches. In particular, we focus on inter-domain control aggregation, and compare an Autonomous System (AS) sink-tree based approach with two examples of a shared AS segment based approach, in terms of the amount of state kept, both per AS and per edge router. Our main contributions are in providing a greater understanding into the design of efficient control path aggregation methods.

—IoT scenarios rely on different IP-based communication protocols to assist data transmission among heterogeneous resource constrained devices. While such solutions are being applied by different vendors to different environments, there is yet not a generalized consensus on which solution best meets networking and service requirements. At the same time, new paradigms such as information-centric networking exhibit network architectural semantics that seem to be relevant in regards of challenges such as security, mobility, scalability. This paper goes over the most recent evolution of IoT communication approaches and compares them to the most emblematic ICN architecture, NDN, from a network architectural perspective. The paper then explains challenges that NDN faces to assist IoT communication interoperability, and provides guidelines to overcome the identified challenges.

—Data classification in pervasive wireless networks is often tied to smart data captured from individual sources. Eager learner models are the ones mostly used in this context. However, activity recognition requires data fusion from multiple sources. In this case, data classification via eager learner methods gives rise to an increase in misclassification. An alternative approach is to consider lazy learning methods such as Memory Based Reasoning , as these models are known to achieve better results even with smaller data sets. This work proposes a generic Memory Based Reasoning model for pervasive wireless networks, comparing its performance against other common machine learning techniques, such as Neural Networks, Decision Tree and Logistic Regression. The performance is evaluated in terms of Misclassification Rate and Average Squared Error, based on a collected data set.

—Short-range wireless technologies bring in the possibility to explore device-to-device communication, and thus to compute proximity indicators, such as relative distance. Models based on signal strength that can be applied to distance computation are often limited by today's mobile sensing technology features, thus leading to significant margins of error, which can be reduced via adequate calibration. Nevertheless, such calibration is today performed by rule of thumb. This paper explains how relative distance can be computed, showing, via an experimental example, the margins of error attained, and proposes a classification model to assist in a better calibration of relative distance computation.

—Mobile social networks rely on regular credential mechanisms to provide users with access to networking resources. This approach has become the de-facto access method, but becomes complex as users have to reenter their credentials each time they hop across visited networks. To assist in making the access more user-friendly, this paper describes the application of social trust mechanisms as a way to lower the complexity inherent to the use of credentials and to assist in non-repudiation. Notions behind social trust schemes and security concerns that can be tackled by social trust mechanisms, as well as a few operational examples are addressed in this paper.

This paper goes over the concept of User-centric  networking,as a paradigm for networking architectures usually located in the Customer Premises, paradigm which is steadily hanging the way the Internet has been devised. Such change is due to the fact that the Internet end-user
is empowered due to novel approaches such as software defined networking, thus being in
control of functionality that so far was restricted to be placed in the core and access regions of  networks. Such change introduces the need to revisit home networking, and from an end-to-end perspective, to introduce new concepts and tehnology into the networking functionality. The paper presents a user-centric  model and its functional blocks, and describes the architectural example  that has been coneived, implemented, and validated in the context of the European project ULOOP -
User-centric  Wireless Local Loop.

This paper addresses mobility management in the
ontext of user-
entri
networking, alerting to the need to

onsider new paradigms
to adapt mobility management solutions to future Internet ar
hite
tures.

The paper provides notions
on
erning distributed mobility management

aspe
ts, as well as what is mobility estimation and how it

an be applied

to

urrent or to future mobility management solutions, based on an ex-
isting proof-of-

on
ept.

IoT communication architectures and protocols have been evolving in order to cope with new challenges derived from environments involving a large number of heterogeneous, resource-constrained devices. Examples of such challenges are the support for intensive processing of large amounts of data; filtering; data mining and classification; high heterogeneity in devices and software. Today end-to-end IoT services and data exchange are mostly supported by TCP/IP, set of protocols which originally have not been conceived having in mind time aspects such as time sensitive data, power constrains, or even the support of sub-second responses, a common requirement in Industrial IoT. Therefore, the most recent trend on communication protocols follow a publish/subscriber broker approach, which creates an abstraction between \emph{things} that produce information (producers), and people or devices that consume such information (consumers). Albeit these protocols assist, for instance, frequent data polling, issues concerning mobility management, privacy, security, and resource consumption subsist. The reason is the underlying networking semantics of TCP/IP, which follow a host-based reachability approach. The new architectural paradigm of ICN focuses on providing support to directly reach information objects, while in contrast today’s Internet is focused on reaching devices that store information objects. The design of ICN paradigms seems to bring in relevant features to IoT environments. ICN approaches such as NDN claim to have integrated security support; reliable multi-path data-based routing; built-in mobility support such as an interface abstraction. Being information-centric, ICN does not transmit data based on host identifiers, such as addresses. In order to further evolve communication protocols in a way that best sustains highly heterogeneous IoT environments, there is the need to better understand current implementation limitations and performance aspects. Such analysis needs to take into consideration the support of both industrial and consumer IIoT environments, given that stakeholders of each domain consider different protocols to support data transmission. Hence, a main issue for IoT data exchange today is \emph{interoperability}. In this context, and among several interoperability gaps, this report is focused on the evaluation of the performance for multiple IoT protocols. We aim to contribute to understand the performance of some of the most relevant protocols available today, in terms of: time to completion, reachability, or even overhead derived from the information exchanged. The work here reported and that has been explored in the COPELABS IoT Lab between September 2018 and July 2019 aims at contributing to the experimental analysis of IoT communication protocols.IoT communication architectures and protocols have been evolving in order to cope with new challenges derived from environments involving a large number of heterogeneous, resource-constrained devices. Examples of such challenges are the support for intensive processing of large amounts of data; filtering; data mining and classification; high heterogeneity in devices and software. Today end-to-end IoT services and data exchange are mostly supported by TCP/IP, set of protocols which originally have not been conceived having in mind time aspects such as time sensitive data, power constrains, or even the support of sub-second responses, a common requirement in Industrial IoT. Therefore, the most recent trend on communication protocols follow a publish/subscriber broker approach, which creates an abstraction between \emph{things} that produce information (producers), and people or devices that consume such information (consumers). Albeit these protocols assist, for instance, frequent data polling, issues concerning mobility management, privacy, security, and resource consumption subsist. The reason is the underlying networking semantics of TCP/IP, which follow a host-based reachability approach. The new architectural paradigm of ICN focuses on providing support to directly reach information objects, while in contrast today’s Internet is focused on reaching devices that store information objects. The design of ICN paradigms seems to bring in relevant features to IoT environments. ICN approaches such as NDN claim to have integrated security support; reliable multi-path data-based routing; built-in mobility support such as an interface abstraction. Being information-centric, ICN does not transmit data based on host identifiers, such as addresses. In order to further evolve communication protocols in a way that best sustains highly heterogeneous IoT environments, there is the need to better understand current implementation limitations and performance aspects. Such analysis needs to take into consideration the support of both industrial and consumer IIoT environments, given that stakeholders of each domain consider different protocols to support data transmission. Hence, a main issue for IoT data exchange today is \emph{interoperability}. In this context, and among several interoperability gaps, this report is focused on the evaluation of the performance for multiple IoT protocols. We aim to contribute to understand the performance of some of the most relevant protocols available today, in terms of: time to completion, reachability, or even overhead derived from the information exchanged. The work here reported and that has been explored in the COPELABS IoT Lab between September 2018 and July 2019 aims at contributing to the experimental analysis of IoT communication protocols.

Device-to-device communication has been introduced years ago in cellular networks as a new paradigm to enhance network performance. Device-to-Device communication, by exploiting direct communication between nearby mobile devices significantly improves system performance and constitutes a promising technology for next-generation wireless networks. And also serves several goals from a network operator perspective. In this paper we present a survey of available tools, simulators, and test-beds dedicated to Device-to-Device communications in a LTE based cellular network.

—Experimentation concerning mobile networks operation is often supported by mobility models, or by some form of mobility emulation, as having a testbed with live roaming subjects is not easy to achieve nor to control in terms of scalability. As such, mobility models are used to mimic different mobility scenarios. In our multi-hop routing experiences, we have detected different results by considering different mobility models. This paper shows how three of the most popular models used today impact different aspects of networking operation, namely, throughput, packet loss. We have analyzed the behavior of the different mobility models by carrying out simulations in the context of two different multi-hop protocols, and our research proposes relevant guidelines for peers, when performing multi-hop routing validation based on mobility models.

User-centric wireless networks abound around us
supported by a variety of networking architectures such as
infrastructured hotspots, ad-hoc or mesh architectures. Common
to all of these is an increasing node movement frequency which
strongly impacts network efficiency, but in order to better
understand such impact, there is the need to capture with some
level of accuracy mobility patterns of the wireless nodes. Hence, a
series of mobility models have been developed. This survey gives
insight into the most recent trades and evolution of mobility
models, highlighting operation and analysis for each of them.
The survey also gives insight into the applicability of mobility
models, beyond the common simulation application environment

... Control Aggregation Mechanisms Rute Sofia ( 1) Roch Guérin ( 2) Pedro Veiga ( 3) Abstract ... Sofia et al. [3] propose another inter-domain control aggregation solution, SICAP, which performs aggregation based on shared-segments along the paths of different reservations. ...

UMM (User-centric Mobility Management) is a project focused on global mobility management in spontaneous wireless environments, ie, environments that self-adjust to fast network topology changes. Global mobility management is addressed from a network ...

ABSTRACT This White Paper provides an insight on Trust Management within the context of the User-centric Wireless Local Loop (ULOOP) project, depicting the main principles and the overall trust management framework, and also describing its main individual components. It has as motivation to disseminate ULOOP concepts and to raise awareness towards trust management in user-centric wireless networks.

In the latest years the highly nomadic lifestyles that Internet users experience, and the strong entanglement between society and technology, lead to the appearance of community networks where the end-user has, most of the times, an active role in terms of sharing Internet access. Such networks range from basic functionality, such as the ability to create a wireless (ad-hoc) network on-the-fly with a simple PC (e.g., Internet Connection Sharing functionality from Microsoft), or more elaborate cases of commercial success, e.g. FON. Wireless networks provided by end-users are expected to grow, despite the limitations imposed by traditional operator-centric Internet communication models. In this new scenario the end-user (or a community of end-users) is a micro-operator in the sense that he/she shares his/her subscribed broadband Internet access based on some incentive scheme. Besides Internet access sharing, being a micro-operator also means providing other network functionality, such...

The Information System Laboratory (Institut für Informationstechnische Systeme, IIS) of
the Department of Computer Science, University of the Federal Armed Forces Munich,
Germany started two years ago research and teaching in the area of communications.
One of the closely related topics is addressing the use and application of technology and
mechanisms under economic and technical optimization measures. Therefore, during the
spring term 2004 (FT 2004) a fourth instance of the Internet Economic seminar has been
prepared and students as well as supervisors worked on this topic.
Still today, Internet Economics are run rarely as a teaching unit. This is a little in contrast
to the fact that research on Internet Economics has been established as an important area
in the center of technology and economics on networked environments. During the last
ten years, the underlying communication technology applied for the Internet and the way
electronic business transactions are performed on top of the network have changed. Alt-
hough, a variety of support functionality has been developed for the Internet case, the core
functionality of delivering data, bits, and bytes remained unchanged. Nevertheless, chan-
ges and updates occur with respect to the use, the application area, and the technology
itself. Therefore, another review of a selected number of topics has been undertaken.