Blog

NSense presented in IEEE Healthcom2016

NSense: A People-centric, non-intrusive Opportunistic Sensing Tool for Contextualizing Social Interaction

The scientific paper NSense (Rute C. Sofia, Saeik Firdose, Luis Amaral Lopes, Waldir Moreira and Paulo Mendes) has been accepted in IEEE Healthcom 2016 (September 14th-18th, Munich). NSense is a software tool developed by COPELABS that tracks and infers social interaction aspects in the form of computational utility functions that aim at describing two indicators of interaction: propinquity, and social interaction level.

Aspects of Global Mobility Management for Next Generation Networks

posted Nov 12, 2018, 11:51 AM by Rute Sofia   [ updated Nov 12, 2018, 12:37 PM ]

Handling mobility management in the Internet is not trivial, given the heterogeneity of devices involved; providers involved; service requirements. The Internet evolution requires re-thinking mobility management and to understand how to best distribute functionality across the network. The data transmission itself needs to take into consideration mobility, and to dynamically adjust to human movement, as next generation networks are information-centric, and user-centric.

During the last decades, several mobility management approaches have been design and validated, and are today operationally available. In the context of distributed mobile edge computing, mobility management architectures can be addressed in a novel way integrating functions such as mobility prediction and learning. For that purpose, there are a few topics that are relevant to be revisited, as explained next.


1. Mobility management functional splitting
The different approaches available today in multi-access heterogeneous networks, across different TCP/IP stack layers, which have been extensively worked upon in the context of the IETF are evolving towards decentralized mobility management. In this process, it is relevant to understand the limitations that current solutions face in next generation networks. It is also relevant to understand which functional blocks compose mobility management architectures, independently of the layer where such solutions reside.

2. From centralized to decentralized mobility management
Moving from centralized to distributed mobility management architectures can be designed in a way that is "closer" to the Internet end-user. Understanding how and where to position the different blocks that compose mobility management imply analyzing how to best decouple mobility management functionality.

3. Mobility modelling and anticipation
Moreover, in dynamic environments (which today are the basis of the Internet fringes), human interaction and computational models that can estimate aspects related with such interaction (e.g., frequency of visits to networks; roaming habits) can be provided via mobility estimation mechanisms. Mobility estimation is therefore a required mobility management function, still missing in today's architectures. Adding such mechanisms to mobility management architectures, centralized or decentralized, is beneficial.

4. Moving towards content-centric mobility management architectures
A more adequate distribution of mobility management requires a move towards a data-centric mobility management perspective. Today, mobility management solutions are focused on mobility management of devices, and this aspect is particularly challenging, due to the inherent addressing schemes, among other aspects. Simultaneous mobility of source and destination devices is an aspect that is not trivial to address. Still, analysing mobility management support from an information-centric perspective, instead of from a host reachability perspective, is relevant in the quest for mobility management that can support future Internet paradigms, where all devices are highly mobile.

Read More:

Evolving Communications in IoT

posted Sep 21, 2018, 11:58 AM by Rute Sofia

Internet of Things (IoT) communication architectures and protocols are evolving to be able to cope with new challenges such as the processing of large amounts of data; filtering; data mining and classification; high heterogeneity in devices and software.
Today, IoT communication is supported by TCP/IP which were not designed having in mind time sensitive networks or low power networks. Energy is wasted each time data is transmitted, due to protocol overhead, and non-optimized communication patterns.
The most recent trend on communication protocols follow a publish/subscriber broker approach, which creates an abstraction between things that produce information, and people or devices that consume such information. Nevertheless, issues concerning mobility management, privacy, security, and resource consumption subsist mostly tied to the networking semantics of 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 seem to bring in relevant features to IoT environments. ICN approaches such as NDN have integrated security support; reliable multi-path data-based routing;  built-in mobility support such as an interface abstraction, which is relevant for multihoming. Being information-centric, ICN does not transmit data based on host identifiers, such as addresses.

The evolutionary trend and interoperability aspects of publish/subscriber approaches as well as of network architectures and protocols in general are being exploited by the COPELABS IoT Lab team.

DABBER: Data reAchaBility BasEd Routing for Named-data Networking Wireless Environments

posted Apr 6, 2018, 4:04 AM by Rute Sofia   [ updated Apr 6, 2018, 4:07 AM ]

The European UMOBILE project is working on methods to extend NDN towards opportunistic wireless networks. In this context, COPELABS has been implementing an NFD extension, called NDN-OPP, which supports NDN multi-hop wireless communication: NDN-OPP V1.0 for Android does broadcast of Interest packets over a Wi-Fi direct network. 
 
In order to reduce resource consumption in a NDN multi-hop wireless networks, COPELABS, SENCEPTION and ATHENA have created the DABBER routing protocol, which allows selective forwarding of Interest packets based on the local announcement of name prefixes + node awareness about the properties of neighbour nodes (centrality and availability). DABBER aims to operate on a point-to-point wireless network (e.g. Wi-Fi direct, DTN) as well as on a broadcast network (e.g. ah-hoc Wi-Fi).
 
DABBER was submitted to IRTF, and will be presented in the ICNRG meeting in London (Mar 17th - Mar 23rd).
 

URL:            https://www.ietf.org/internet-drafts/draft-mendes-icnrg-dabber-00.txt
Status:         https://datatracker.ietf.org/doc/draft-mendes-icnrg-dabber/

The Role of Smart Data in Inference of Human Behavior and Interaction - book Chapter

posted Apr 6, 2018, 3:31 AM by Rute Sofia   [ updated Apr 6, 2018, 3:33 AM ]

Rute C. Sofia, Liliana I. Carvalho., Francisco de Melo Pereira, Samrat Dattagupta.The Role of Smart Data in Inference of Human Behavior and Interaction. Book chapter. "Smart Data: State-of-the-Art and Perspectives in Computing and Applications". Editors:K.-C. Li, Q. Zhang. L. T. Yang, B. Di Martino. CRC Press, Taylor & Francis Group, USA. April 2019. ISBN: 1138545589. (Accepted April 2018, To appear).

This chapter explores features, concepts, and provides guidelines concerning the role and applicability of smart data captured in a non-intrusive way, in the inference and contextualization of human behavior and interaction.
The chapter starts by introducing aspects related to human interaction, for instance, how to define and to best model physical and psychological proximity; models for social awareness and social contextualization. The next part of the chapter deals with interaction inference and interaction contextualization, namely: classification models that best suit the inference of behavior in the vere of smart data; challenges in regards to small data capture and behavior inference derived from small data, in particular when considering decentralized, mobile cyber-physical systems; guidelines to model interaction based on pervasive wireless sensing systems, including available middleware and systems (tools). The chapter then provides information concerning specific applicability use-cases, namely, Points of Interest detection via smart data, and how smart data can be used to boost social interaction in a pervasive, non-intrusive way. The chapter concludes with a set of recommendations.

Connecting the Edges: A Universal, Mobile-Centric, and Opportunistic Communications Architecture

posted Feb 19, 2018, 8:04 AM by Rute Sofia

New H2020 UMOBILE paper: C. A. Sarros et al., "Connecting the Edges: A Universal, Mobile-Centric, and Opportunistic Communications Architecture," in IEEE Communications Magazine, vol. 56, no. 2, pp. 136-143, Feb. 2018.doi: 10.1109/MCOM.2018.1700325

The Internet has crossed new frontiers with access to it getting faster and cheaper. Considering that the architectural foundations of today’s Internet were laid more than three decades ago, the Internet has done remarkably well until today to cope with the growing demand. However, the future Internet architecture is not only expected to support the ever-growing number of users and devices but also a diverse set of new applications and services.
Departing from the traditional host-centric access paradigm where access to a desired content is mapped to its location, an information-centric model enables the association of access to a desired content with the content itself, irrespectively of the location where it is being held. UMOBILE tailors the information-centric communication model to meet the requirements of opportunistic communications, integrating those connectivity approaches into a single architecture.By pushing services near the edge of the network, such an architecture can pervasively operate in any networking environment and allows for the development of innovative applications, providing access to data independently of the level of end-to-end connectivity availability.





Senception 'Ones to Watch', running for Public European Champion 2017/2018

posted Feb 1, 2018, 2:19 AM by Rute Sofia   [ updated Feb 1, 2018, 2:48 AM ]

01.2018:
Senception Lda has been awarded by the European Business Awards as 'Ones to Watch' 2017/2018. Senception is one of the 11 small portuguese SMEs running for 'Public European Champion 2017/2018'.
Please support us by voting via: European Business Awards

TUM Kolloquium: Contextual Interaction inference and characterization derived from wireless network mining

posted Sep 18, 2017, 5:10 AM by Rute Sofia

07.2017: Video available via the TUM media portal


This talk is focused on wireless network mining and on context and behavior characterization derived from the application of non-intrusive, pervasive sensing approaches in connected wireless environments. The research described in this talk goes over the development of networking solutions and mechanisms that can assist data capture and distributed inference of roaming habits in a way that may lead to social interaction stimulation (and as a consequence, to a better design of networking communication).


NSense presented in IEEE Healthcom2016

posted Sep 18, 2017, 5:05 AM by Rute Sofia   [ updated Sep 18, 2017, 5:06 AM ]

NSense: A People-centric, non-intrusive Opportunistic Sensing Tool for Contextualizing Social Interaction

09.2016

The scientific paper NSense (Rute C. Sofia, Saeik Firdose, Luis Amaral Lopes, Waldir Moreira and Paulo Mendes) has been accepted in IEEE Healthcom 2016 (September 14th-18th, Munich). NSense is a software tool developed by COPELABS that tracks and infers social interaction aspects in the form of computational utility functions that aim at describing two indicators of interaction: propinquity, and social interaction level.

Food for Thought

posted Nov 5, 2015, 3:17 PM by Rute Sofia

Set of material that I have been developing towards new aspects in networking (updated when there is time...)

Dynamic Frequency Sharing, wireless multi-station downstream transmission

posted Apr 17, 2015, 6:21 AM by Rute Sofia   [ updated Apr 17, 2015, 6:23 AM ]


DFS is a software-based mechanism that targets short-range wireless networks where transmission is based on a shared medium (e.g. broadcast) and which rely on OFDM for data transmission. DFS is applicable downstream, from the antenna to the station, and relies on techniques both from OSI Layer 1 and 2 to assist data transmission to multiple stations within a time-frame that based only in OFDM could only serve the purpose of serving a single station.

Allowing downstream transmission via one symbol to multiple stations provides the means to improve the performance of current solutions three-fold. Firstly, by allowing data to be transmitted on the same time-frame to multiple stations, the control overhead is reduced in comparison to the current standards, as the same control information is used to transmit data to multiple stations. Secondly, for real-time traffic there is an upper bound on usable data rates. For instance, for Voice over IP (VoIP) traffic it is approximately 486 kilobits per second (kbps), and for video traffic such limit is of 1 megabit per second (Mbps). Due to this limit, increasing the capacity of the wireless link does not suffice to improve performance as buffering cannot be used in real-time traffic. With the increasing popularity of VoIP, online-gaming, this inefficiency becomes an important problem to be solved. By multiplexing data downstream (from controller to stations) to several stations our solution is expected to provide a better usage of high data rate channels, which is a beneficial aspect in terms of real-time traffic. Thirdly, instead of transmitting to stations one by one, thus wasting time in particular if the first station that captures the medium is what is known as „slow“ station (e.g. away from the antenna or attaining severe interference around), our solution provides a way to transmit „simultaneously“ data to several stations within the same time frame thus decreasing the round-trip time and the latency of the transmission.

 

DFS has been conceived, validated, and implemented by COPELABS (Rute Sofia and Luis LOpes) and University of Kent (Huseyin Haci, Hassan Osman, Huiling Zhu) in the context of the European project ULOOP - User-centric Wireless Local Loop.

Related work:

 

1-10 of 13