Talks

  • Invited Talks

  • flag_germany   Indoor Data Management in Anyplace May 9, 2017

    Location: Heidelberg University, Germany


    The pervasiveness of smartphones is leading to the uptake of a new class of Internet-based Indoor Navigation (IIN) services, which might soon diminish the need of Satellite-based localization technologies in urban environments. These services rely on geo-location databases that store spatial models along with wireless, light and magnetic signals used to localize users and provide better power efficiency and wider coverage than predominant approaches. In this talk I will overview the research behind the building blocks of the Anyplace IIN, an open, modular, extensible and scalable navigation architecture that exploits crowdsourced Wi-Fi data to develop a novel navigation service that won several international research awards for its utility and accuracy (i.e., less than 2 meters). Our MIT-licenced open-source software stack has to this date been used by hundreds of researchers and practitioners around the globe, with the public Anyplace service reaching over 80,000 real user interactions. In the second part of this talk, I will focus on an algorithm we developed for protecting users from location tracking by the IIN service, without hindering the provisioning of fine-grained location updates on a continuous basis. Our algorithm exploits a k-Anonymity Bloom filter and a generator of camouflaged localization requests, both of which are shown to be resilient to a variety of privacy attacks. My talk will conclude with a summary of other recent research work.
  • flag_france   Building All k Nearest Neighbor Social Communities April 12, 2017

    Location: Paris Descartes University, France


    A wide spectrum of Internet-scale mobile applications, ranging from social networking, gaming and entertainment to emergency response and crisis management, all require efficient and scalable All k Nearest Neighbor (AkNN) computations over millions of moving objects every few seconds to be operational. Most traditional techniques for computing AkNN queries are centralized, lacking both scalability and efficiency. Only recently, distributed techniques for shared-nothing cloud infrastructures have been proposed to achieve scalability for large datasets. These batch-oriented algorithms are sub-optimal due to inefficient data space partitioning and data replication among processing units. In this talk I will present Spitfire, a distributed algorithm that provides a scalable and high performance AkNN processing framework. The proposed algorithm deploys a fast load-balanced partitioning scheme along with an efficient replication-set selection algorithm, to provide fast main-memory computations of the exact AkNN results in a batch-oriented manner. I will also overview Rayzit, an experimental and open-source mobile AkNN service we established and operate that reached over 45,000 real user interactions to this date. My talk will conclude with a summary of other recent research work related to Indoor Localization and Privacy as well as Telco Big Data.
  • flag_france   IoT-enabled Localization and Navigation June 16, 2016

    Location: 2016 PERCCOM Summer School, Erasmus Mundus Master on Pervasive Computing & COMmunications for sustainable development, Nancy, France


    The advances of Internet-of-Things (IoT) technology in recent years is leading to the uptake of a new class of Internet-based navigation services, which might soon diminish the need of Satellite-based technologies in urban environments. These services rely on geolocation databases that store spatial models along with wireless, light and magnetic signals used to localize users. Developing IoT-based navigation services creates a new spectrum of information management challenges ranging from crowdsourcing indoor models, acquiring and fusing big-data velocity signals, localization algorithms, location privacy of custodians and others. In this talk, I will overview the current landscape of academic and industrial such services using a multi-dimensional taxonomy of emerging topics in this domain, including location, crowdsourcing, privacy and modeling. I present the dimensions of our taxonomy through the lens of an open, modular, extensible and scalable navigation architecture, coined Anyplace, concluding with open challenges.
  • flag_finland   Spatial Big Data Research and Applications at the University of Cyprus May 16, 2016

    Location: The First Europe-China Workshop on Big Data Management, Helsinki, Finland


    Spatial big data architectures have transformed the way enterprises collect, store and analyze massive amounts of velocity data that features a spatial extend. In this talk, I will summarize an array of research problems and applications that our laboratory has investigated or aims to investigate in this scope. I will start out by overviewing Anyplace (http://anyplace.cs.ucy.ac.cy/), our in-house Internet-based Indoor Navigation service that won several international research awards for its accuracy and utility. Particularly, I will discuss the challenges in processing and visualizing indoor big data (e.g., Wi-Fi and magnetic signals). My talk will be succeeded by a summary of Rayzit (http://rayzit.com/), which is our award-winning location-based crowd messaging service that allows a mobile crowd to instantly connect to their k Nearest Neighbors (kNN) as they move in space. I will particularly summarize Spitfire, which is a distributed algorithm that provides a scalable and high-performance All k Nearest Neighbor processing framework to Rayzit. I will then overview Spate, which is a novel spatio-temporal time machine for telecommunication data (e.g., CDR, NMS, PCHR). Spate supports a range of visual analytic cues (heatmaps, POI clusters, etc.) that enables a Telco operator to quickly compare abstract models to real velocity data but also to monitor and replay network traces using a declarative SQL language that breaks down to SPARK jobs. I will finally also introduce GreenCharge, which is a big data GIS architecture to guide electric vehicles to green energy excess. GreenCharge is expected to maximize self-consumption of electricity by prosumers, contributing in that way to the stability of power grids and a sustainable future. GreenCharge is anticipated to be pilot at the University of Cyprus, a self-sufficient 17GWh/annum producer of solar electricity.
  • flag_spain   Anyplace Indoor Information Service May 3, 2016

    Location: Symposium on Challenges of Fingerprinting in Indoor Positioning and Navigation, Open University of Catalonia, Barcelona, Spain


    People spend 80-90% of their time in indoor environments such as offices, undergrounds, shopping malls and airports. On the other hand, the uptake of interesting applications in indoor spaces (e.g., navigation, inventory management and elderly support) has so far been hampered by the lack of technologies that can provide indoor location (position) accurately, in real-time, in an energy-efficient manner and without expensive additional hardware. Modern smartphones currently rely on Internet-based Indoor Navigation (IIN) services, which can provide the location of a user upon request. Unfortunately, those technologies are both inaccurate and additionally raise important location privacy concerns, as the IIN can know where the user is at all times. In this talk, I will start out by overviewing the building blocks of Anyplace (http://anyplace.cs.ucy.ac.cy/), our in-house IIN services that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. In the second part of this talk, I will focus on an algorithm we developed for protecting users from location tracking by the IIN service, without hindering the provisioning of fine-grained location updates on a continuous basis. Our algorithm exploits a k-Anonymity Bloom filter and a generator of camouflaged localization requests both of which are shown to be resilient to a variety of privacy attacks. In the third part of this talk, I will focus on an innovative framework for accurate fast indoor localization over an intermittently connected WiFi coined Prefetching Localization (PreLoc).
  • flag_cyprus   Prefetching Indoor Navigation Structures in Anyplace March 23, 2016

    Location: University of Nicosia, Nicosia, Cyprus


    People spend 80-90% of their time in indoor environments such as offices, undergrounds, shopping malls and airports. On the other hand, the uptake of interesting applications in indoor spaces (e.g., navigation, inventory management and elderly support) has so far been hampered by the lack of technologies that can provide indoor location (position) accurately, in real-time, in an energy-efficient manner and without expensive additional hardware. Modern smartphones currently rely on Internet-based Indoor Navigation (IIN) services, which provide the location of a user upon request using structures called RadioMaps (RMs). In this talk, I will start out by overviewing the building blocks of Anyplace (http://anyplace.cs.ucy.ac.cy/), our in-house IIN services that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. In the second part of this talk, I will focus on an innovative framework for accurate fast indoor localization over an intermittently connected WiFi coined Prefetching Localization (PreLoc).
  • flag_netherlands   Internet-based Indoor Navigation Services February 25, 2016

    Location: ACROSS Meeting, Centrum Wiskunde & Informatica (CWI), Amsterdam, Netherlands


    People spend 80-90% of their time in indoor environments such as offices, undergrounds, shopping malls and airports. On the other hand, the uptake of interesting applications in indoor spaces (e.g., navigation, inventory management and elderly support) has so far been hampered by the lack of technologies that can provide indoor location (position) accurately, in real-time, in an energy-efficient manner and without expensive additional hardware. Modern smartphones currently rely on Internet-based Indoor Navigation (IIN) services, which can provide the location of a user upon request. Unfortunately, those technologies are both inaccurate and additionally raise important location privacy concerns, as the IIN can know where the user is at all times. In this talk, I will start out by overviewing the building blocks of Anyplace (http://anyplace.cs.ucy.ac.cy/), our in-house IIN services that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. In the second part of this talk, I will focus on an algorithm we developed for protecting users from location tracking by the IIN service, without hindering the provisioning of fine-grained location updates on a continuous basis. Our algorithm exploits a k-Anonymity Bloom filter and a generator of camouflaged localization requests both of which are shown to be resilient to a variety of privacy attacks.
  • flag_cyprus   Multifaceted Engagement of BSc Students in Computer Science Research December 15, 2015

    Location: Research-led Teaching Workshop, Centre for Teaching & Learning, University Cyprus, Nicosia, Cyprus.


    Research entails a deep systematic investigation of sources in order to reach new, previously unknown conclusions. In Computer Science, these conclusions are often in the form of algorithmic developments or in the form of software artifacts tackling unconventional, real-life challenges. In both cases, the conclusions have to be publishable at high caliber journals, conferences or competitions, such that the credibility of the developments is externalized and recognized by the wider research community. While the above type of research is a fundamental attribute of postgraduate studies, undergraduate students are often not part of this process for a variety of reasons, including: i) lack of advanced domain knowledge (usually taught at the post-graduate level); ii) lack of mathematical or laboratory skills (e.g., proving a theorem or managing a complex software stack); and iii) lack of writing and presentation skills (i.e., the dissertation is written at the end of their studies). During the last decade, I have applied a wide range of empirical methods to engage undergraduate students in research as soon as they are admitted to the University. Particularly, students have been encouraged through a variety of methods to familiarize themselves with research through our courses and become active contributors to our quest for academic excellence. The applied methods have had some very promising results, particularly: i) several summer research internships since 2012 (for 1st year BSc students); ii) several research interviews available on Youtube since 2014 (for 2nd year BSc students); iii) 5 distinctions in the Intl. ACM SIGMOD programming competition since 2010 (for 3rd year BSc students); iv) 21 journal and conference proceedings since 2007 (for 4th year BSc students); and v) 3 international awards for innovative research systems since 2007 (again for 3rd and 4th year students). In the scope of my case study, I will detail the applied methodology for each different type of activitity and also summarize the lessons learnt. I claim that the provided achievements have helped our undergraduate students to reach their personal goals and our Data Management Systems Laboratory (DMSL) to meet its research agenda. Particularly, the above engagement allowed 6 students to obtain fellowships for graduate studies (e.g., Fulbright, Oxford), 5 students to advance to PhD programs, while all of them have moved on with MSc studies in the US, UK, Canada and Switzerland.
  • flag_cyprus   Prefetching Indoor Navigation Structures in Anyplace October 27, 2015

    Location: 2nd Cyprus Workshop on Computing: Scientific Applications of Computing, European University Cyprus, Nicosia, Cyprus.


    Wi-Fi (or WLAN) based indoor navigation applications mobiles rely on cloud-based services (s) that take of a user’s (u) localization task using structures called (RMs). It is imperative for u to have a stable WiFi in order to either continuously receive location from s or to download RMs a priori for offline navigation. networks however, suffer from intermittent connectivity due poor network planning that results in sparse deployment access points and effectively areas where Wi-Fi coverage be guaranteed. This inherently affects the localization and therefore the navigation experience of users. In paper, we propose an innovative framework for accurate fast indoor localization over an intermittently connected WiFi coined Prefetching Localization (PreLoc). In Preloc, prioritize the download of RM records based on knowledge from historic traces of other users inside the same Instead of downloading the complete RM from s to we propose a Probabilistic Group Selection (PGS) strategy, identifies RM records that have a higher probability of necessary to a user moving inside a target area. We have our framework using a real prototype developed in as well as realistic Wi-Fi traces we collected at the of Cyprus. Our experimental study reveals that PreLoc PGS and conventional fingerprint-based indoor positioning can yield accuracy that is as good as using the same with a complete RM even under scenarios of weak Wi-Fi coverage.
  • flag_greece   Panel: The Beckman Database Research Self-Assessment Meeting July 31, 2015

    Location: 13th Hellenic Data Management Symposium (HDMS 2015), Athens University of Economics and Business, Athens, Greece


    Panel Co-Chairs: Yannis Ioannidis and Panagiotis Karras. Other Panelists: Alexandros Labrinidis, Peter Triantafillou and Herodotos Herodotou.
  • flag_usa   Indoor Data Management in Anyplace June 24, 2015

    Location: Department of Computer Science, University of Pittsburgh, PA USA


    People spend 80-90% of their time in indoor environments such as offices, undergrounds, shopping malls and airports. On the other hand, the uptake of interesting applications in indoor spaces (e.g., navigation, inventory management and elderly support) has so far been hampered by the lack of technologies that can provide indoor location (position) accurately, in real-time, in an energy-efficient manner and without expensive additional hardware. Modern smartphones currently rely on Internet-based Indoor Navigation (IIN) services, which can provide the location of a user upon request. Unfortunately, those technologies are both inaccurate and additionally raise important location privacy concerns, as the IIN can know where the user is at all times. this talk, I will start out by overviewing the building blocks of Anyplace (http://anyplace.cs.ucy.ac.cy/), our in-house IIN services that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. In the second part of this talk, I will focus on an algorithm we developed for protecting users from location tracking by the IIN service, without hindering the provisioning of fine-grained location updates on a continuous basis. Our algorithm exploits a k-Anonymity Bloom filter and a generator of camouflaged localization requests, both of which are shown to be resilient to a variety of privacy attacks. My talk will be succeeded by a summary of Rayzit which is an award-winning location-based crowd messaging service that addresses big-data velocity with parallel algorithms and distributed NoSQL databases.
  • flag_usa   Tutorial: Mobile Data Management in Indoor Spaces June 17, 2015

    Location: The 16th IEEE International Conference on Mobile Data Management (MDM '15), Pittsburgh, PA, USA.


    This advanced seminar presents the fundamental mobile data management concepts behind the realization of innovative indoor information services that deal with all aspects of handling indoor data as a valuable resource, including data modeling, data acquisition, query processing, privacy and energy consumption. The goal is to provide an overview of the emerging field of indoor data management with a particular emphasis on mobile systems. We tackle the topic from a wide range of perspectives: fundamentals, definitions, current state, academic & industrial perspective, reality & visionary scenarios as well as future challenges. The seminar captures the big picture, such that interested researchers and practitioners can expand their study by following the references. Our presentation will be carried out through the lens of an experimental Indoor Information System we developed at the University of Cyprus, coined Anyplace, which has obtained three international awards and was ranked the second most accurate indoor localization technology by Microsoft Research at IEEE/ACM IPSN '14.
  • flag_cyprus   Indoor Data Management: Status and Challenges February 17, 2015

    Location: Department of Computer Science, University of Cyprus, Nicosia Cyprus


    People spend 80-90% of their time in indoor environments such as offices, undergrounds, shopping malls and airports. On the other hand, the uptake of interesting applications in indoor spaces (e.g., navigation, inventory management and elderly support) has so far been hampered by the lack of technologies that can provide indoor location (position) accurately, in real-time, in an energy-efficient manner and without expensive additional hardware. Modern smartphones currently rely on cloud-based Indoor Positioning Services (IPS), which can provide the location of a user upon request but those are both inaccurate and additionally raise important location privacy concerns, as the IPS can know where the user is at all times. In this talk, I will start out by overviewing the building blocks of Anyplace, our in-house IPS that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. In the second part of this talk, I will focus on an algorithm we developed for protecting users from location tracking by the IPS, without hindering the provisioning of fine-grained location updates on a continuous basis. Our algorithm exploits a k-Anonymity Bloom filter and a generator of camouflaged localization requests both of which are shown to be resilient to a variety of privacy attacks.
  • flag_hungary   Smartphone Cloud Testbeds and Applications December 5, 2014

    Location: European Institute of Innovation & Technology (EIT), ICT Labs, Budapest Hungary


    The explosive number of smartphones with ever growing computing and sensing capabilities have brought a paradigm shift to many traditional domains of the computing field. In this talk, I will present three (3) ongoing testbeds and applications we are developing in-house in the space of smartphone and human-centric systems: In will start out with SmartLab (smartlab.cs.ucy.ac.cy), a Mobile Infrastructure-as-a-Service cloud we have developed and deployed at the University of Cyprus. In SmartLab, an intuitive web-based interface supplies a variety of complex mobile management utilities that provide fine-grained and low-level control over real smartphones, e.g., usage of networking, storage and sensors as well as automated mockup executions. We present our research experiences from using SmartLab in different research settings as well as our envisioned future scenarios for urban-scale deployment, federation issues and security studies. I will continue with Anyplace (anyplace.cs.ucy.ac.cy), our in-house Indoor Positioning Service that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. I will conclude with Rayzit (rayzit.com) which is an award-winning location-based crowd messaging service that addresses big-data velocity with parallel algorithms and distributed NoSQL databases.
  • flag_russia   Indoor Data Management: Status and Challenges October 2, 2014

    Location: Skolkovo Institute of Science and Technology (Skoltech), Moscow Russia


    People spend 80-90% of their time in indoor environments such as offices, undergrounds, shopping malls and airports. On the other hand, the uptake of interesting applications in indoor spaces (e.g., navigation, inventory management and elderly support) has so far been hampered by the lack of technologies that can provide indoor location (position) accurately, in real-time, in an energy-efficient manner and without expensive additional hardware. Modern smartphones currently rely on cloud-based Indoor Positioning Services (IPS), which can provide the location of a user upon request but those are both inaccurate and additionally raise important location privacy concerns, as the IPS can know where the user is at all times. In this talk, I will start out by overviewing the building blocks of Anyplace, our in-house IPS that recently won several international research awards for its accuracy (i.e., less than 2 meters) and utility. Anyplace deploys a number of innovative concepts, including crowdsourcing, big-data management, energy-aware processing, multi-device optimization and mobile data management, in order to realize a power-efficient and accurate indoor localization and navigation technology. In the second part of this talk, I will focus on an algorithm we developed for protecting users from location tracking by the IPS, without hindering the provisioning of fine-grained location updates on a continuous basis. Our algorithm exploits a k-Anonymity Bloom filter and a generator of camouflaged localization requests, both of which are shown to be resilient to a variety of privacy attacks. My talk will be succeeded by a summary of related research efforts, namely SmartLab, which is a novel in-house programming cluster of smartphones that we use in our experimental studies; and Rayzit which is an award-winning location-based crowd messaging service that addresses big-data velocity with parallel algorithms and distributed NoSQL databases.
  • flag_austria   Managing Smartphone Cloud Testbeds September 30, 2014

    Location: Distributed Systems Group, Technical University of Vienna, Vienna Austria


    The explosive number of smartphones with ever growing computing and sensing capabilities have brought a paradigm shift to many traditional domains of the computing field. Smartphone users nowadays gain access to unprecedented possibilities, knowledge and power due to a diverse landscape of applications. Developers on the other hand are challenged with a fragmented mobile landscape that is extremely dynamic to changes in both hardware and software. Re-programming smartphones and instrumenting them for application testing and data gathering at scale is currently a tedious and time-consuming process that poses significant logistical challenges. In this talk, we present the abstractions comprising SmartLab, a Mobile Infrastructure as a Service cloud we have developed and deployed at the University of Cyprus. In SmartLab, an intuitive web-based interface supplies a variety of complex mobile management utilities that provide fine-grained and low-level control over real smartphones, e.g., usage of networking, storage and sensors as well as automated mockup executions. We present our research experiences from using SmartLab in different research settings as well as our envisioned future scenarios for urban-scale deployment, federation issues and security studies. My talk will be succeeded by a summary of related mobile data management research efforts, namely Anyplace, which is our in-house indoor localization and navigation service; and Rayzit which is our award-winning location-based crowd messaging service.
  • flag_cyprus   Mobile Crowdsourcing: Challenges and Applications March 31, 2014

    Location: 7th Webdatanet (COST Action IS1004) Conference on Mobile Research, Larnaca Cyprus.


    Crowdsourcing refers to a distributed problem-solving model in which a crowd of undefined size is engaged to solve a complex problem through an open call. This novel problem-solving model found its way into numerous applications on the web for voting, fund-raising, micro-works and wisdom-of-the-crowd scenarios. Multi-sensing capabilities and multi-modal connectivity means of smartphones offer a great platform for extending and diversifying web-based crowdsourcing applications to a larger contributing crowd, making contribution easier and omnipresent. Unfortunately, numerous new challenges ranging from big data stream volume and velocity, to location privacy and energy consumption as well as device diversity issues among other, arise in this new context. In this talk, I will summarize mobile crowdsourcing challenges and applications related to Microblogging Urban Sensing and Indoor Information Systems.
  • flag_greece   Crowdsourcing Urban Data with Smartphones March 28, 2014

    Location: The 17th International Conference on Extending Database Technology (EDBT/ICDT), Mining Urban Data (MUD) Workshop, Athens, Greece


    Crowdsourcing refers to a distributed problem-solving model in which a crowd of undefined size is engaged to solve a complex problem through an open call. This novel problem-solving model found its way into numerous applications on the web for voting, fund-raising, micro-works and wisdom-of-the-crowd scenarios. Multi-sensing capabilities and multi-modal connectivity means of smartphones offer a great platform for extending and diversifying web-based crowdsourcing applications to a larger contributing crowd, making contribution easier and omnipresent. Unfortunately, numerous new challenges ranging from big data stream volume and velocity, to location privacy and energy consumption as well as device diversity issues among other, arise in this new context. this talk, I will start out by a discussion on how primitive crowdsourcing challenges emerge and evolve into urban data collection scenarios. I will present these new challenges through the lens of some in-house systems we’ve developed over the last few years, particularly: i) a smartphone programming cluster coined SmartLab, which provides means for general-purpose urban-scale sensing scenarios and smart cities; ii) an indoor information system coined Anyplace, which addresses big-data, privacy and energy consumption for building WiFi Radiomaps of indoor places; and iii) a location-based crowd messaging service coined Rayzit which addresses big data velocity with parallel algorithms and distributed nosql databases.
  • flag_italy   Panel: Large-Scale Participatory Urban Sensing: A Fad or Reality? July 5, 2013

    Location: 14th IEEE International Conference on Mobile Data Management (MDM '13), Milan, Italy


    One of the popular research trends at present focuses on the use of sensor data generated/collected by consumer mobile devices to infer the ‘urban state’. There are a fairly large number of research initiatives that view such a citizen-centric distributed and mobile sensing platform as one of the most promising ways to gather data about various aspects of cities, such as environmental parameters & pollution levels, traffic congestion, popularity of events at various public spaces, etc. there are many skeptics who doubt that this “decentralized, bottom-up” approach can be an effective & commercially viable approach in the long run, due to open challenges in many aspects, such as resource limitations, privacy data quality and incentives. This panel will explore how academia and industry are tackling these challenges and debate on what types of applications are likely to be sustainable under this crowd-sourced paradigm.
  • flag_italy   Tutorial: Crowdsourcing for Mobile Data Management June 4, 2013

    Location: The 14th IEEE International Conference on Mobile Data Management (MDM '13), Milan Italy.


    Crowdsourcing refers to a distributed problem-solving model in which a crowd of undefined size is engaged to solve a complex problem through an open call. This novel problem-solving model found its way into numerous applications on the web for voting, fund-raising, micro-works and wisdom-of-the-crowd scenarios. On the other hand, the shift of desktop users to mobile platforms in the post-PC era, along with the unique multi-sensing capabilities of modern mobile devices are expected to eventually unfold the full potential of Crowdsourcing. This is true, as smartphones offer a great platform for extending and diversifying web-based crowdsourcing applications to a larger contributing crowd, making contribution easier and omnipresent. This advanced seminar presents the fundamental concepts behind crowdsourcing and its applications to mobile data management. In the first part of the seminar, we will overview the crowdsourcing landscape from a variety of perspectives, with a particular emphasis on the latest data management trends. In the second and more extended part of the seminar, we will focus on an in-depth coverage of emerging mobile crowdsourcing architectures and systems, through a multi-dimensional taxonomy that will address location, sensing, power, performance, big-data and privacy among others. Furthermore, we will overview a number of in-house crowdsourcing prototypes we have developed and deployed over the last few years. The seminar concludes with challenges opportunities and new directions in the field.
  • flag_cyprus   Big Data - What is it? March 12, 2013

    Location: 4th Architect Club Meeting, Nicosia Cyprus


    Big data refers to data sets whose size and structure strains the ability of commonly used relational DBMSs to capture, manage, and process the data within a tolerable elapsed time. Big data sizes commonly range from a few dozen terabytes to many petabytes in a single database and their underlying data model might be anything from structured (relational or tabular) to semi-structured (XML or JSON) or even unstructured (Web text and log files). Big data architectures are highly parallel and distributed in order to cope with the inherent I/O and CPU limitations. Such systems typically perform on mid-scale private clouds, offering higher privacy, to large-scale public clouds, both exposing operational and analytic functionality stand-alone or as-a-Service. This talk aims to overview the current big-data management landscape, the underlying technologies and their provenance, the latest NoSQL and NewSQL trends, possible applications of big-data management systems for online and offline processing of sensor data, text data, social data and medical data in enterprise environments. The talk will also overview ongoing big-data research and teaching activities at the University of Cyprus.
  • flag_cyprus   Smartphone Sensing: Testbeds and Applications February 11, 2013

    Location: Workshop on Social Platforms for Urban Sensing, Dept. of Comp. Science, University of Cyprus, Nicosia Cyprus


    Smartphone devices have emerged as powerful computational platforms equipped with multitude of sensors that are capable of generating vast amounts of data (geo-location, audio, video, etc.) Collections of such devices connected to the Internet yield Smartphone Networks, which can be utilized for opportunistic and participatory sensing applications in intelligent transportation systems, social networking applications, city planning and others. In this talk, I will present a collection of ongoing testbeds and applications we are developing in-house for this new era of ubiquituous smartphone computing. In particular, I will be presenting SmartLab, which an innovative open programming cloud of 40+ Android devices deployed at our Department over the last three years. SmartLab is the first open Smartphone IaaS (Infrastructure-as-a-Service) cloud that enables fine-grained, low-level interactions over static or moving smartphones via an intuitive web-based interface. Such a testbed provides means for general-purpose urban-scale sensing scenarios and personal gadget management. This talk will also briefly cover other smartphone sensing testbeds and applications we 've developed: a localization engine for fine-grained positioning without GPS, a trajectory comparinson framework with privacy gurranttees, a P2P smartphone searching framework and a neighborhood sensing framework.
  • flag_cyprus   Querying Sensor Data in Smartphone Networks October 11, 2012

    Location: CS Colloquium Series @ UCY, Nicosia Cyprus


    Smartphones have emerged as powerful computational platforms equipped with multitude of sensors that are capable of generating vast amounts of data (geo-location, audio, video, etc.) Collections of smartphones connected to the Internet are nowadays proposed for opportunistic and participatory sensing applications in intelligent transportation systems, social networking applications, city planning and many other domains, prompting undeniably the post-PC era. In this talk, I will present distributed architectures for querying and managing such sensor data by taking into account energy, data disclosure and networking aspects. I will particularly focus on SmartTrace, a powerful query processing framework for finding similar smartphone trajectories without disclosing the traces of participating users. I will also present SmartLab, a first-of-a-kind programmable cloud of 40+ smartphones deployed at our department enabling a new line of systems-oriented research on smartphones. Finally, I will also overview other related smartphone data management frameworks we 've developed for peer-to-peer search, crowdsourcing and indoor positioning, concluding with an outlook to our future research agenda.
  • flag_greece   Data Management Techniques for Smartphone Networks June 12, 2011

    Location: 10th Intl. ACM Workshop on Data Engineering for Wireless and Mobile Access (MobiDE '11), Athens Greece


    Smartphone devices have emerged as powerful computational platforms equipped with multitude of sensors that are capable of generating vast amounts of data (geo-location, audio, video, etc.) Collections of such devices connected to the Internet yield Smartphone Networks, which can be utilized for opportunistic and participatory sensing applications in intelligent transportation systems, social networking applications, city planning and others. The uptake of applications in this domain, is currently severely hampered by the fact that these devices have: i) a limited energy budget (i.e., smartphone devices still operate on batteries), ii) limited connectivity (i.e., not all regions offer unlimited Internet connectivity at the same cost); and iii) high privacy constraints (i.e., these devices might reveal the identity and habits of their custodians.) In this talk, I will present a collection of data management techniques that deal with Smartphone Networks. In particular, I will start out with SmartTrace, a powerful framework for finding similar trajectories in a smartphone network without disclosing the traces of the participating users. SmartTrace relies on an in-situ data storage model, where geo-location data is recorded locally on smartphones for both performance and data-disclosure reasons. SmartTrace then deploys an efficient top-K query-processing algorithm that exploits distributed trajectory similarity measures, resilient to spatial and temporal noise, in order to derive the most relevant answers quickly and efficiently. I will then introduce SmartOpt, a multi-objective query optimizer that enables efficient content searches in smartphone networks. I will also introduce Proximity, a spatial neighborhood computation framework for smartphone networks. My talk will be succeeded by the presentation of SmartNet, our in-house programming cloud for smartphone networks
  • flag_usa   Energy Efficient Data Management in Smartphone Networks April 2, 2011

    Location: US National Science Foundation Workshop on Sustainable Energy Efficient Data Management, Arlington, Virginia USA


    Smartphone computational platforms equipped with multitude of sensors and capable of generating vast amounts of data (geo-location, audio, video, etc.) On the other hand, these devices operate on a strict energy budget, thus have a limited lifetime on a single charge. Consequently, we need to identify new energy-aware algorithms and techniques to provide innovative, feature-rich applications and services. In this white paper, we start out by providing recent trends in Smartphone technology and Smartphone networks. Our description is succeeded by an anatomy of the energy costs associated with data processing in a Smartphone Network. We conclude with prominent research directions in energy-aware data management for Smartphone networks.
  • flag_usa   Querying Smartphone Networks with SmartTrace March 29, 2011

    Location: Dept. of Computer Science, University of Pittsburgh, Pittsburgh, PA, USA, March 29 2011


    Smartphone devices have emerged as powerful computational platforms equipped with multitude of sensors that are capable of generating vast amounts of data (geo-location, audio, video, etc.) Collections of such devices connected to the Internet yield Smartphone Networks, which can be utilized for opportunistic and participatory sensing applications in intelligent transportation systems, social networking applications, city planning and others. The uptake of applications in this domain, is currently severely hampered by the fact that these devices have: i) a limited energy budget (i.e., smartphone devices still operate on batteries), ii) limited connectivity (i.e., not all regions offer unlimited Internet connectivity at the same cost); and iii) high privacy constraints (i.e., these devices might reveal the identity and habits of their custodians). In this talk I will present SmartTrace, a powerful framework for finding similar trajectories in a smartphone network, without disclosing the traces of participating users. Our framework, coined SmartTrace, quickly answers queries of the form: “Report the users that move more similar to Q, where Q is some query trace.” SmartTrace relies on an in-situ data storage model, where geo-location data is recorded locally on smartphones for both performance and data-disclosure reasons. SmartTrace then deploys an efficient top-K query-processing algorithm that exploits distributed trajectory similarity measures, resilient to spatial and temporal noise, in order to derive the most relevant answers to Q quickly and efficiently. We assess our propositions with realistic and real workloads from Microsoft Research Asia and other sources. Our study reveals that SmartTrace computes the desired results with 74% less energy consumption and 13% faster than its centralized and decentralized counterparts. My talk will be succeeded by a summary of related research efforts, namely SmartNet, an innovative programming cloud for smartphone networks; and SmartOpt, a multi-objective query optimizer that enables efficient content searches in smartphone networks.
  • flag_cyprus   Query Routing Trees for Wireless Sensor Networks March 9, 2011

    Location: Information Systems, Open University of Cyprus, Nicosia Cyprus


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor the physical world at an extremely high fidelity. In order to collect the data generated by these tiny-scale devices, sensors are typically organized in structures coined Query Routing Trees (QRTs). Our study reveals that predominant data acquisition systems construct QRTs in ad-hoc manners leading to a significant waste of energy. In this talk I will present MicroPulse+, a framework for minimizing the consumption of energy during data collection in Sensor Networks. MicroPulse+ eliminates a variety of data transmission and data reception inefficiencies using a collection of in-network algorithms. In particular, MicroPulse+ introduces: i) the Workload-Aware Routing Tree (WART) algorithm, which is established on profiling recent data collection activity and on identifying the bottlenecks using an in-network execution of the critical path method; and ii) the Energy-driven Tree Construction (ETC) algorithm, which balances the workload among nodes and minimizes data collisions. The talk will conclude with an outlook into current and future research work.
  • flag_greece   Ranking Query Results in a Networked World March 9, 2011

    Location: Department of Informatics, University of Athens Greece


    In this talk I will present a family of algorithms for Top-k ranking of query results in a distributed environment. A Top-K query focuses on the subset of most relevant answers for two reasons: i) to minimize the cost metric that is associated with the retrieval of all answers; and ii) to improve the quality of the answer set such that the user is not overwhelmed with irrelevant results. I will start out by providing an overview of Top-K query processing algorithms for centralized and middleware systems. I will then highlight the limitations of these algorithms and focus on three novel algorithms we developed designated for networked environments (i.e., Peer-to-Peer Networks, Wireless Sensor Networks and Smartphone Networks). I will also present evaluation studies of these algorithms on: i) a Wireless Sensor Network testbed of 54 sensor devices; ii) a Peer-to-Peer testbed of 1000 peers deployed on 75 linux workstations; and iii) A smartphone network deployment on Android-based smartphone devices. The talk will conclude with an overview of related research problems that I am currently working on and an outlook to future work.
  • flag_cyprus   Panel: Data Management in Clouds: Research Challenges and Opportunities July 3, 2010

    Location: 9th Hellenic Data Management Symposium (HDMS '10), Ayia Napa, Cyprus


    Panel Chair: Anastasia Ailamaki. Other Panelists: Marios Dikaiakos, Evangelia Pitoura, Peter Triantafillou and Akrivi Vlachou
  • flag_usa   Ranking Query Results in a Networked World May 27, 2010

    Location: IBM T.J. Watson Research Center, Hawthorne, NY, USA


    In this talk I present the fundamental concepts behind distributed Top-K query processing algorithms. A Top-K query focuses on the subset of most relevant answers for two reasons: i) to minimize the cost metric that is associated with the retrieval of all answers; and ii) to improve the quality of the answer set such that the user is not overwhelmed with irrelevant results. I will start out by providing an overview of state-of-the-art Top-K query processing algorithms for centralized and middleware systems. I will then highlight the limitations of these algorithms and focus on two novel algorithms we developed designated for networked environments (i.e., Wireless Sensor Networks, Peer-to-Peer Networks, Vehicular Networks, etc.) I will also present evaluation studies conducted on: i) a Peer-to-Peer testbed of 1000 peers deployed on 75 workstations; ii) a Wireless Sensor Network testbed of 54 sensor devices and iii) A Smartphone Network, deployed on a number of Android-based smartphone devices. The talk will conclude with an overview of related research problems that I am currently working on and an outlook to future applications of the presented ideas.
  • flag_usa   Spatio-Temporal Query Processing in Smartphone Networks May 23, 2010

    Location: SSPC-WAN Workshop, 11th Intl. Conference on Mobile Data Management (MDM '10), Kansas City, MO, USA


    In this presentation I will present a powerful and distributed spatio-temporal query processing framework, coined HUB-K. Our framework can be utilized to promptly answer queries of the form: ' 'Report the objects (i.e., trajectories) that follow a similar spatio-temporal motion to Q, where Q is some query trajectory. ' ' HUB-k, relies on an in-situ data storage model, where spatio-temporal data remains on the smartphone that generated the given data, as well a state-of-the-art top-k query processing algorithms, which exploit distributed trajectory similarity measures in order to identify the correct answers promptly. We present preliminary design choices, an outline of our preliminary implementation and an outlook to future challenges.
  • flag_germany   Semantic Challenges in (Mobile) Sensor Networks January 26, 2010

    Location: Seminar 10042: Semantic Challenges in Sensor Networks, Dagstuhl, Germany


    The widespread deployment of mobile phones along with the massive production of sensors for every aspect of modern life provides evidence that Computer Science research and education will evolve dramatically over the next few years. The boundaries of Mobile Devices and Sensor Devices are nowadays blurring as the former devices are already equipped with a multitude of sensing capabilities, including GPS (which enables the derivation of geospatial coordinates), accelerometers (which enable the derivation of orientation, vibration and shock) and an exciting set of other sensors (e.g., proximity sensors, ambient light sensors, while more traditional sensors such as temperature, acoustic, magnetometers and others will be integrated in these devices very soon). That creates the notion of Mobile Sensor Devices that will become even more ubiquitous than their predecessor 'smart-phone ' devices. In this talk, I will provide an overview and definitions of Mobile-Sensor-Network (MSN) related platforms and applications. In particular, I will show how applications in environmental monitoring, body sensor networks, vehicular sensor networks and intelligent transportation systems have brought a dramatic shift on how spatio-temporal data is nowadays generated. I will then outline some semantic challenges that arise in this context including: vastness, uncertainty, data integration, query processing and privacy. I will also address some more general challenges that currently hinder the evolution and uptake of semantic MSNs.
  • flag_spain   Distributed Top-K Ranking Algorithms December 15, 2008

    Location: DAMA Group, Polytechnic University of Catalonia (UPC), Barcelona, Spain


    In this talk I will present the fundamental concepts of distributed Top-K query processing algorithms. A Top-K query focuses on a subset of most relevant answers for two reasons: i) to minimize the cost metric that is associated with the retrieval of all answers; and ii) to improve the quality of the answer set such that the user is not overwhelmed with irrelevant results. I will start out by providing an overview of state-of-the-art Top-K query processing algorithms for centralized DBMS systems. I will then highlight the limitations of these algorithms and focus on the Threshold Join Algorithm (TJA), our distributed top-k query processing algorithm designated for distributed computing networks (i.e., Wireless Sensor Networks, Peer-to-Peer Networks, Vehicular Networks, etc.) I will finally present an evaluation study conducted with our middleware system deployed over a network of 1000 peers on 75 workstations.
  • flag_switzerland   MicroHash - An Efficient Index Structure for Flash-Based Sensor Devices December 12, 2008

    Location: IBM Research, Zurich, Switzerland


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor the physical world at an extremely high fidelity. Research in this area has to this day primarily focused on the trade-off between local computation and communication in order to minimize the transfer of data over the fundamentally expensive wireless link. On the contrary, we focus on the challenges of storing sensor readings locally at each node. This In-Situ storage paradigm offers a novel perspective for conserving energy in Wireless Sensor Networks as the communication channel is only accessed for answering on-demand queries rather than for percolating each and every event to a centralized database. Storing large quantities of data locally at each sensor has to be complemented by efficient access methods that will speed up the execution of queries when required. In this talk I will present MicroHash, an external memory index structure that is tailored to the distinct characteristics of the most prevalent type of non-volatile memory used in sensor systems, namely flash memory. MicroHash exploits the asymmetric read/write characteristics of flash memory in order to offer high performance indexing and searching capabilities in the presence of energy and storage media lifetime constraints.
  • flag_switzerland   An Overview of Distributed Top-K Ranking Algorithms December 12, 2008

    Location: Communication Systems Group (CSG), ETH Zurich, Switzerland


    In this talk I will present the fundamental concepts of distributed Top-K query processing algorithms. A Top-K query focuses on a subset of most relevant answers for two reasons: i) to minimize the cost metric that is associated with the retrieval of all answers; and ii) to improve the quality of the answer set such that the user is not overwhelmed with irrelevant results. I will start out by providing an overview of state-of-the-art Top-K query processing algorithms for centralized DBMS systems. I will then highlight the limitations of these algorithms and focus on the Threshold Join Algorithm (TJA), our distributed top-k query processing algorithm designated for distributed computing networks (i.e., Wireless Sensor Networks, Peer-to-Peer Networks, Vehicular Networks, etc.) I will finally present an evaluation study conducted with our middleware system deployed over a network of 1000 peers on 75 workstations.
  • flag_china   Tutorial: Distributed Top-K Query Processing in Wireless Sensor Networks April 27-30, 2008

    Location: The 9th International Conference on Mobile Data Management (MDM '08), Beijing, China.


    Wireless Sensor Networks create an innovative technology that enables users to monitor and study the physical world at an extremely high resolution. Query processing in such ad-hoc environments is a challenging task due to the complexities imposed by the inherent energy and communication constraints. To this end, the research community has proposed to take into account user-defined parameters in order to derive the K most relevant (or Top-K) answers quickly and efficiently. A Top-K query returns the subset of most relevant answers, in place of all answers, for two reasons: i) to minimize the cost metric that is associated with the retrieval of all answers; and ii) to improve the recall and the precision of the answer set, such that the user is not overwhelmed with irrelevant results. This tutorial presents the fundamental concepts behind distributed Top-K query processing and the adaptations of these algorithms to distributed and wireless sensor networks. It additionally provides a gentle overview of rudimentary and advanced techniques covering a significant body of research in this domain. The tutorial will start out with an overview of the most influential centralized and middleware Top-K query processing algorithms and then proceed with an elaborate description of distributed Top-K ranking algorithms for One-time Top-K Queries, Continuous Top-K Queries and Approximate Top-K Queries. Finally, it will provide an outlook to compelling future applications that can be constructed on the foundation of these algorithms. Although the tutorial is specifically geared towards Wireless Sensor Networks, many of the presented ideas find extensions in other mobile environments such as Adhoc Networks, Vehicular Networks and the Mobile Web.
  • flag_united_kingdom   MicroHash - An Efficient Index Structure for Flash-Based Sensor Devices January 11, 2008

    Location: Systems and Networking Group, Microsoft Research Cambridge, Cambridge, UK


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor the physical world at an extremely high fidelity. Research in this area has to this day primarily focused on the trade-off between local computation and communication in order to minimize the transfer of data over the fundamentally expensive wireless link. On the contrary, we focus on the challenges of storing sensor readings locally at each node. This In-Situ storage paradigm offers a novel perspective for conserving energy in Wireless Sensor Networks as the communication channel is only accessed for answering on-demand queries rather than for percolating each and every event to a centralized database. Storing large quantities of data locally at each sensor has to be complemented by efficient access methods that will speed up the execution of queries when required. In this talk I will present MicroHash, an external memory index structure that is tailored to the distinct characteristics of the most prevalent type of non-volatile memory used in sensor systems, namely flash memory. MicroHash exploits the asymmetric read/write characteristics of flash memory in order to offer high performance indexing and searching capabilities in the presence of energy and storage media lifetime constraints.
  • flag_sweden   Content-Based Search in Internet-Scale Peer-to-Peer Systems December 28, 2006

    Location: Department of Electronic, Computer and Software Systems (ECS), KTH - Royal Institute of Technology, Stockholm, Sweden


    The emerging Peer-to-Peer (P2P) model has become a very powerful and attractive paradigm for developing Internet-scale services for sharing resources, including files and documents. The distributed nature of these systems, where nodes are typically located across different networks and domains, inherently hinders the efficient retrieval of information. In this talk I will present techniques to perform content-based search over data repositories that are geographically scattered over peers of different networks. Data repositories in this context contain documents of text, audio, video or other semi-structured data and the task is to locate a certain set of keywords or multimedia features. We present the components of the pFusion architecture, an open source system that builds on work in unstructured P2P systems and topologically-aware overlay construction techniques. Our empirical results using datasets from AKAMAI, NLANR and TREC, show that the architecture we propose is both efficient and practical. In this talk I will also overview other related research activities in Grid, P2P and Sensor systems that we are currently involved in.
  • flag_greece   Top-K Query Processing Techniques for Distributed Environments June 8, 2006

    Location: Institute of Computer Science (ICS) of the Foundation for Research and Technology Hellas (FORTH), Crete, Greece


    Emerging applications in Sensor and Peer-to-Peer networks make the concept of data integration without centralization nowadays more meaningful than ever. In these environments, data is generated continuously and potentially automatically across geographically diverse locations. Organizing data in centralized repositories is becoming prohibitively expensive and in many occasions impractical. Storing data in-situ however, complicates query processing because data relations are fragmented over a number of remote sites. Furthermore, accessing these fragmented relations is only feasible by traversing a network of other nodes. This makes the execution of a query an even more complex task. We claim that in many occasions it might more beneficial to find the K highest ranked (or Top-K) answers, for some user defined parameter K, if this can minimize the query execution cost. In this talk, I will present techniques to efficiently answer Top-K queries in a distributed environment. A Top-K query returns the K highest ranked answers to a user defined similarity function. At the same time it also minimizes some cost metric, such as the utilization of the communication medium, which is associated with the retrieval of the desired answer set. I will provide an overview of state-of-the-art algorithms that solve the Top-K problem in a centralized setting and show why these are not applicable to the distributed case. I will then focus on the Threshold Join Algorithm (TJA), which is a novel solution for executing Top-K queries in a distributed environment. I will also present results from our performance study with a real middleware testbed deployed over a network of 75 workstations.
  • Other Talks

  • flag_greece   Distributed In-Memory Processing of All k Nearest Neighbor Queries July 28, 2016

    Location: The 14th Hellenic Data Management Symposium (HDMS 2016), Athens University of Economics and Business, Athens, Greece


    A wide spectrum of Internet-scale mobile applications, ranging from social networking, gaming and entertainment to emergency response and crisis management, all require efficient and scalable All k Nearest Neighbor (AkNN) computations over millions of moving objects every few seconds to be operational. Most traditional techniques for computing AkNN queries are centralized, lacking both scalability and efficiency. Only recently, distributed techniques for shared-nothing cloud infrastructures have been proposed to achieve scalability for large datasets. These batch-oriented algorithms are sub-optimal due to inefficient data space partitioning and data replication among processing units. this paper we present Spitfire, a distributed algorithm that provides a scalable and high-performance AkNN processing framework. Our proposed algorithm deploys a fast load-balanced partitioning scheme along with an efficient replication-set selection algorithm, to provide fast main-memory computations of the exact AkNN results in a batch-oriented manner. We evaluate, both analytically and experimentally, how the pruning efficiency of the Spitfire algorithm plays a pivotal role in reducing communication and response time up to an order of magnitude, compared to three other state-of-the-art distributed AkNN algorithms executed in distributed main-memory. Επεξεργασία Επερωτήσεων k Πλησιέστερων Γειτόνων Όλων των Αντικειμένων σε Κατανεμημένη Κύρια Μνήμη ευρύ φάσμα κινητών εφαρμογών Διαδικτυακής κλίμακας, από εφαρμογές κοινωνικής δικτύωσης και παιχνίδια μέχρι εφαρμογές αντιμετώπισης και διαχείρισης καταστάσεων έκτακτης ανάγκης, απαιτούν έναν αποτελεσματικό και κλιμακώσιμο τρόπο υπολογισμού επερωτήσεων k Πλησιέστερων Γειτόνων Όλων των Αντικειμένων (All k Nearest Neighbors - AkNN) πάνω από εκατομμύρια κινούμενα αντικείμενα κάθε λίγα δευτερόλεπτα. Οι πλείστες γνωστές τεχνικές για τον υπολογισμό AkNN ερωτημάτων δεν είναι κατανεμημένες και συνεπώς δεν μπορούν να εφαρμοστούν αποδοτικά σε μεγάλη κλίμακα και γρήγορες ροές δεδομένων. Μόνο πρόσφατα προτάθηκαν κατανεμημένες τεχνικές βασιζόμενες σε υποδομές νέφους για την επίτευξη κλιμάκωσης σε μεγάλα σύνολα δεδομένων, οι οποίες ωστόσο υστερούν σε απόδοση διότι δεν επιτυγχάνουν το βέλτιστο διαχωρισμό του χώρου δεδομένων και τη βέλτιστη αντιγραφή δεδομένων μεταξύ των μονάδων επεξεργασίας. αυτό το άρθρο παρουσιάζουμε τον Spitfire, έναν κατανεμημένο αλγόριθμο που παρέχει ένα επεκτάσιμο και υψηλής απόδοσης πλαίσιο επεξεργασίας AkNN ερωτημάτων σε κατανεμημένη κύρια μνήμη. Ο προτεινόμενος αλγόριθμος χρησιμοποιεί ένα αποδοτικό πλαίσιο διαχωρισμού των δεδομένων και ένα πλαίσιο κλωνοποίησης δεδομένων, τα οποία επιτυγχάνουν ισορροπημένα φορτία επεξεργασίας μεταξύ των υπολογιστικών μονάδων. Στην αναλυτική και πειραματική μας αξιολόγηση δείχνουμε πως η απόδοση κλαδέματος του αλγορίθμου Spitfire διαδραματίζει κεντρικό ρόλο στη μείωση του φόρτου επικοινωνίας και του χρόνου ανταπόκρισης ως και μια τάξη μεγέθους σε σύγκριση με τρεις άλλους κατανεμημένους αλγόριθμους αιχμής που επιλύουν AkNN επερωτήσεις σε κατανεμημένη κύρια μνήμη.
  • flag_greece   Privacy-Preserving Indoor Localization on Smartphones July 28, 2016

    Location: The 14th Hellenic Data Management Symposium (HDMS 2016), Athens University of Economics and Business, Athens, Greece


    Indoor Positioning Systems (IPS) have recently received considerable attention, mainly because GPS is unavailable in indoor spaces and consumes considerable energy. On the other hand, predominant Smartphone OS localization subsystems currently rely on server-side localization processes, allowing the service provider to know the location of a user at all times. In this paper, we propose an innovative algorithm for protecting users from location tracking by the localization service, without hindering the provisioning of fine-grained location updates on a continuous basis. proposed Temporal Vector Map (TVM) algorithm, allows a user to accurately localize by exploiting a k-Anonymity Bloom (kAB) filter and a bestNeighbors generator of camouflaged localization requests, both of which are shown to be resilient to a variety of privacy attacks. We have evaluated our framework using a real prototype developed in Android and Hadoop HBase as well as realistic Wi-Fi traces scaling-up to several GBs. Our analytical evaluation and experimental study reveal that TVM is not vulnerable to attacks that traditionally compromise k-anonymity protection and indicate that TVM can offer fine-grained localization in approximately four orders of magnitude less energy and number of messages than competitive approaches. Γεωτοποθέτηση Κινητών Συσκευών σε Εσωτερικούς Χώρους με Διαφύλαξη της Ιδιωτικότητας των Χρηστών Τα συστήματα εντοπισμού θέσης σε εσωτερικούς χώρους έχουν λάβει πρόσφατα ιδιαίτερη προσοχή, κυρίως επειδή το GPS είναι ενεργοβόρο και εφόσον αυτό δεν είναι διαθέσιμο σε εσωτερικούς χώρους. Τα υποσυστήματα γεωτοποθέτησης των έξυπνων κινητών συσκευών βασίζονται πλέον σε Διαδικτυακές υπηρεσίες παροχής θέσης. Αυτό ωστόσο αναδεικνύει το πρόβλημα της ιδιωτικότητας θέσης, εφόσον οι πάροχοι θέσης μπορούν να γνωρίζουν πλέον τη θέση των χρηστών ανά πάσα στιγμή. αυτό το άρθρο παρουσιάζουμε ένα καινοτόμο αλγόριθμο για την προστασία της ιδιωτικότητας θέσης των χρηστών από υπηρεσίες εντοπισμού χωρίς να εμποδίζουμε την παροχή ενημερώσεων θέσης σε συνεχή βάση. Ο προτεινόμενος αλγόριθμος, με το όνομα Temporal Vector Map (TVM), επιτρέπει στους χρήστες να προσδιορίσουν τη θέση τους με ακρίβεια χρησιμοποιώντας ένα εξειδικευμένο φίλτρο Bloom, το οποίο παρέχει k-ανωνυμία (kAB), και μια τεχνική δημιουργίας καμουφλαρισμένων αιτήσεων εντοπισμού θέσης, με το όνομα bestNeighbors. Το προτεινόμενο πλαίσιο έχει αξιολογηθεί χρησιμοποιώντας ένα πραγματικό πρωτότυπο, το οποίο αναπτύχθηκε στο Android και Hadoop HBase μαζί με αρκετά GBs από ρεαλιστικά ίχνη Wi-Fi. Η αναλυτική αξιολόγηση και η πειραματική μας μελέτη αποκαλύπτουν ότι το TVM δεν είναι ευάλωτο σε επιθέσεις που παραδοσιακά παραβιάζουν την προστασία k-ανωνυμίας και δείχνουν ότι το TVM μπορεί να προσφέρει ακρίβεια γεωτοποθέτησης με περίπου τέσσερις τάξεις μεγέθους λιγότερη ενέργεια και αριθμό μηνυμάτων από ανταγωνιστικές προσεγγίσεις.
  • flag_greece   Radiomap Prefetching for Indoor Navigation in Intermittently Connected Wi-Fi Networks July 30, 2015

    Location: The 13th Hellenic Data Management Symposium (HDMS 2015), Athens University of Economics and Business, Athens, Greece


    Wi-Fi (or WLAN) based indoor navigation applications mobiles rely on cloud-based services (s) that take of a user’s (u) localization task using structures called (RMs). It is imperative for u to have a stable WiFi in order to either continuously receive location from s or to download RMs a priori for offline navigation. networks however, suffer from intermittent connectivity due poor network planning that results in sparse deployment access points and effectively areas where Wi-Fi coverage be guaranteed. This inherently affects the localization and therefore the navigation experience of users. In paper, we propose an innovative framework for accurate fast indoor localization over an intermittently connected WiFi coined Prefetching Localization (PreLoc). In Preloc, prioritize the download of RM records based on knowledge from historic traces of other users inside the same Instead of downloading the complete RM from s to we propose a Probabilistic Group Selection (PGS) strategy, identifies RM records that have a higher probability of necessary to a user moving inside a target area. We have our framework using a real prototype developed in as well as realistic Wi-Fi traces we collected at the of Cyprus. Our experimental study reveals that PreLoc PGS and conventional fingerprint-based indoor positioning can yield accuracy that is as good as using the same with a complete RM, even under scenarios of weak coverage. Προ-ανάκτηση Ραδιοχαρτών για Πλοήγηση σε Εσωτερικούς Χώρους επί Δίκτυου Wi-Fi Διακοπτόμενης Συνδεσιμότητας σύγχρονες υπηρεσίες πλοήγησης σε εσωτερικούς χώρους εξαρτώνται από υπηρεσίες νέφους, οι οποίες εντοπίζουν τη θέση του χρήστη χρησιμοποιώντας δομές Ραδιοχαρτών (RM), οι οποίες καταγράφουν την ένταση ισχύος σημείων πρόσβασης Wi-Fi σε διάφορα σημεία στο χώρο. Σε αυτές τις υπηρεσίες, είναι σημαντικό να υπάρχει σταθερή σύνδεση του χρήστη στο δίκτυο Wi-Fi, έτσι ώστε να μπορεί ο χρήστης να βρίσκει συνεχώς την θέση του από την υπηρεσία ή για να κατεβάσει o χρήστης ολόκληρο το RM εκ των προτέρων βρίσκοντας τη θέση του αυτόνομα. Τα δίκτυα Wi-Fi, ωστόσο, υποφέρουν από διακοπτόμενη συνδεσιμότητα λόγω ελλιπούς σχεδιασμού των δικτύων ή για οικονομικούς παράγοντες. Αυτό επηρεάζει άμεσα την ακρίβεια προσδιορισμού της θέσης του χρήστη και ως εκ τούτου την εμπειρία πλοήγησης του στο χώρο. αυτό το άρθρο παρουσιάζουμε ένα καινοτόμο πλαίσιο, με την επωνυμία PreLoc, για γρήγορο και ακριβή προσδιορισμό της θέσης των χρηστών σε εσωτερικούς χώρους με υποδομή δικτύων Wi-Fi διακοπτόμενης συνδεσιμότητας. Στο PreLoc, γίνεται μια ευφυής διάταξη των εγγραφών του RM βασιζόμενοι σε γνώση που ανακτήθηκε από ιστορικές πλοηγήσεις χρηστών στο ίδιο κτίριο. Αντί να ανακτηθεί ολόκληρο το RM από την υπηρεσία στον χρήστη, προτείνουμε μια στρατηγική με το όνομα PGS, η οποία προσδιορίζει τις εγγραφές του RM που έχουν την μεγαλύτερη πιθανότητα να είναι αναγκαίες στον χρήστη καθώς πλοηγείται. Αξιολογήσαμε το πλαίσιο μας χρησιμοποιώντας ένα πραγματικό πρότυπο που αναπτύχθηκε σε Android, αλλά και ρεαλιστικά ίχνη Wi-Fi που συλλέχθηκαν στο Πανεπιστήμιο Κύπρου. Η πειραματική μας μελέτη αποκαλύπτει ότι το πλαίσιο PreLoc, με χρήση της στρατηγικής PGS και συμβατικών αλγόριθμων εντοπισμού θέσης σε εσωτερικούς χώρους, μπορεί να αποφέρει παρόμοια ακρίβεια με αυτή που επιτυγχάνεται εάν χρησιμοποιούσαμε ολόκληρο το RM ακόμη και σε σενάρια ασθενούς κάλυψης του χώρου από τα σημεία πρόσβασης Wi-Fi.
  • flag_usa   Radiomap Prefetching for Indoor Navigation in Intermittently Connected Wi-Fi Networks April 16, 2015

    Location: 16th IEEE International Conference on Mobile Data Management (MDM 2015), Pittsburgh, PA, USA


    Wi-Fi (or WLAN) based indoor navigation applications mobiles rely on cloud-based services (s) that take of a user’s (u) localization task using structures called (RMs). It is imperative for u to have a stable WiFi in order to either continuously receive location from s or to download RMs a priori for offline navigation. networks however, suffer from intermittent connectivity due poor network planning that results in sparse deployment access points and effectively areas where Wi-Fi coverage be guaranteed. This inherently affects the localization and therefore the navigation experience of users. In paper, we propose an innovative framework for accurate fast indoor localization over an intermittently connected WiFi coined Prefetching Localization (PreLoc). In Preloc, prioritize the download of RM records based on knowledge from historic traces of other users inside the same Instead of downloading the complete RM from s to we propose a Probabilistic Group Selection (PGS) strategy, identifies RM records that have a higher probability of necessary to a user moving inside a target area. We have our framework using a real prototype developed in as well as realistic Wi-Fi traces we collected at the of Cyprus. Our experimental study reveals that PreLoc PGS and conventional fingerprint-based indoor positioning can yield accuracy that is as good as using the same with a complete RM even under scenarios of weak coverage.
  • flag_usa   Scalable Mockup Experiments on Smartphones using SmartLab April 16, 2015

    Location: 16th IEEE International Conference on Mobile Data Management (MDM 2015), Pittsburgh, PA, USA


    In this paper we present a comprehensive architecture carry out experimental repeatability studies on clusters smartphones. Our architecture is founded on SmartLab (http://smartlab.cs.ucy.ac.cy/), in-house architecture for managing real and virtual smartphones an intuitive Web user interface. Our presented architecture of several exciting components for re-programming and smartphones to perform application testing and gathering in a facile manner as well as executing mockup by “feeding” the devices with GPS/sensor readings. will particularly demonstrate the various components of our that encompasses smartphone sensor data collected mobile users and organized in our distributed NoSQL document The given datasets can then be replayed on our comprising of real and virtual smartphones accessible developers through our Web 2.0 user interface. We present applicability of our architecture through various mockup over different application scenarios.
  • flag_usa   Anyplace: A Crowdsourced Indoor Information Service April 16, 2015

    Location: 16th IEEE International Conference on Mobile Data Management (MDM 2015), Pittsburgh, PA, USA


    People do most of their activities, business, commerce, and socializing indoors. As all of these increasingly aided by online services and indoor spaces becoming bigger and more complex, there is a growing for cost-effective indoor localization, mapping, navigation information services. In this paper, we present a complete Information Service, coined Anyplace (http://anyplace.cs.ucy.ac.cy/), has an open, modular, extensible and scalable architecture, making it ideal a wide range of applications. Our service features three desirable properties, namely crowdsourcing, scalability and Anyplace implements a set of crowdsourcing-supportive to handle the enormous amount of crowd-sensed filter incorrect user contributions and exploit Wi-Fi data heterogeneous mobile devices. Moreover, it uses a big-data for efficient storage and retrieval of localization and data. Finally, our service relies on the abundance of data on smartphones (e.g. Wi-Fi signal strength and measurements) to deliver reliable indoor geolocation that received several international awards.
  • flag_usa   Rayzit: An Anonymous and Dynamic Crowd Messaging Architecture April 15, 2015

    Location: 3rd IEEE International Workshop on Mobile Data Management, Mining, and Computing on Social Networks, collocated with IEEE MDM '15 (Mobisocial '15), Pittsburgh, PA, USA


    The smartphone revolution has introduced a new of social networks where users communicate over anonymous platforms to exchange opinions, ideas and even carry commerce. These platforms enable individuals to establish interactions between strangers based on a common interest attribute. In this paper we present Rayzit, a novel anonymous crowd messaging architecture which utilizes the location of each user to connect them instantly to their k Nearest Neighbors as they move in space. Contrary to the very large body location-based social networks that suffer from bootstrapping our architecture enables a user to always interact with geographically closest possible users around. We establish this using a fast computation of an All kNN query that a dynamic global social graph every few seconds. We motivating application scenarios and the detailed backend that allows Rayzit to scale. We have collected analyzed data from the interactions of thousands of active and confirm our claims..
  • flag_germany   Crowdsourced Indoor Localization and Navigation with Anyplace April 16, 2014

    Location: 13th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN 2014), Berlin, Germany


    In this demonstration paper, we present the Anyplace system that relies on the abundance of sensory data on smartphones (e.g., WiFi signal strength and inertial mea- surements) to deliver reliable indoor geolocation information. Our system features two highly desirable properties, namely crowdsourcing and scalability. Anyplace implements a set of crowdsourcing-supportive mechanisms to handle the enormous amount of crowdsensed data, filter incorrect user contributions and exploit WiFi data from heterogeneous mobile devices. More- over Anyplace follows a big-data architecture for efficient and scalable storage and retrieval of localization and mapping data.
  • flag_germany   Sensor Mockup Experiments with SmartLab April 16, 2014

    Location: 13th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN 2014), Berlin, Germany


    In this demonstration paper we present SmartLab, an architecture for managing a cluster of both Android Real De- vices (ARDs) and Android Virtual Devices (AVDs) via an intuitive web-based interface. Our architecture consists of several exciting components for re-programming and instrumenting smartphones to perform application testing and data gathering in a facile manner as well as executing mockup experiments by “feeding” the devices with GPS/sensor readings. We will particularly demonstrate the various components of our architecture that encompasses smartphone sensor data collected by mobile users and organized in our distributed NoSQL document store. The given datasets can then be replayed on our testbed comprising of real and virtual smartphones accessible to developers through our Web 2.0 user interface. We present the applicability of our architecture through various mockup experiments over different application scenarios.
  • flag_usa   Managing Smartphone Testbeds with SmartLab November 7, 2013

    Location: 27th USENIX Large Installation System Administration Conference (LISA '13), Washington D.C., USA


    The explosive number of smartphones with ever growing and computing capabilities have brought a shift to many traditional domains of the computing Re-programming smartphones and instrumenting for application testing and data gathering scale is currently a tedious and time-consuming process poses significant logistical challenges. In this we make three major contributions: First, we propose comprehensive architecture, coined SmartLab1, managing a cluster of both real and virtual smartphones are either wired to a private cloud or connected a wireless link. Second, we propose and a number of Android management optimizations command pipelining, screen-capturing, file which can be useful to the community for similar functionality into their systems. Third conduct extensive experiments and microbenchmarks support our design choices providing qualitative evidence the expected performance of each module comprising architecture. This paper also overviews experiences using SmartLab in a research-oriented setting also ongoing and future development efforts.
  • flag_austria   Spatio-Temporal WiFi Localization October 1, 2013

    Location: Final Conference of MOVE COST Action IC903, Vienna, Austria, Sept 30 - Oct 1, 2013


  • flag_cyprus   Smartphone Data Management: Frameworks and Applications June 6, 2013

    Location: ERCIM Spring Meeting 2013, June 6, 2013, Nicosia Cyprus


  • flag_italy   CLODA: A Crowdsourced Linked Open Data Architecture June 3, 2013

    Location: 1st IEEE Intl. Workshop on Mobile Data Management, Mining, and Computing on Social Networks (MobiSocial) with IEEE MDM '13, June 3, 2013, Milan Italy


    In this paper we present our Crowdsourced Linked Open Data Architecture (CLODA), a first attempt to combine crowdsourcing, localization and location-based services to gener- ate, collect, validate and relate real-world, geo-spatial and multi- dimensional information using smartphones and other mobile devices. CLODA focuses on the construction of URI addressable, interlinked and semi-structured data following the Linked-Open Data (LOD) paradigm. The validity of the constructed data is then contributed by a participating crowd. We present our prototype implementation on top of Google Maps and a blend of in-house technologies, particularly our indoor positioning framework, coined Airplace, our trajectory similarity framework, coined SmartTrace, our neighborhood detection framework, coined Proximity and our smartphone testing platform coined SmartLab.
  • flag_cyprus   Big Data - What is it? March 19, 2013

    Location: EPL671 Course, Department of Computer Science, University of Cyprus, Nicosia Cyprus


    Big data refers to data sets whose size and structure strains the ability of commonly used relational DBMSs to capture, manage, and process the data within a tolerable elapsed time. Big data sizes commonly range from a few dozen terabytes to many petabytes in a single database and their underlying data model might be anything from structured (relational or tabular) to semi-structured (XML or JSON) or even unstructured (Web text and log files). Big data architectures are highly parallel and distributed in order to cope with the inherent I/O and CPU limitations. Such systems typically perform on mid-scale private clouds, offering higher privacy, to large-scale public clouds, both exposing operational and analytic functionality stand-alone or as-a-Service. This talk aims to overview the current big-data management landscape, the underlying technologies and their provenance, the latest NoSQL and NewSQL trends, possible applications of big-data management systems for online and offline processing of sensor data, text data, social data and medical data in enterprise environments. The talk will also overview ongoing big-data research and teaching activities at the University of Cyprus.
  • flag_india   Continuous all k-nearest neighbor querying in smartphone networks July 24, 2012

    Location: The 13th IEEE International Conference on Mobile Data (IEEE MDM '12), Bangalore India


    Consider a centralized query operator that identifies to every smartphone user its k geographically nearest neighbors at all times, a query we coin Continuous All k-Nearest Neighbor Such an operator could be utilized to enhance public services, allowing users to send SOS beacons out to the rescuers, allowing gamers and social networking users to ad-hoc overlay communication infrastructures, in order carry out complex interactions. In this paper, we study the of efficiently processing a CAkNN query in a cellular or network, both of which are ubiquitous. We introduce an coined Proximity, which answers CAkNN queries in time, where n denotes the number of users and λ a parameter (λ << n). Proximity does not require additional infrastructure or specialized hardware and its is mainly attributed to a smart search space sharing we introduce. Its implementation is based on a novel structure, coined k+heap, which achieves constant O(1) time and logarithmic O(log(k*λ)) insertion/update time. being parameter-free, performs efficiently in the face high mobility and skewed distribution of users (e.g., the service equally well in downtown, suburban or rural areas). We evaluated Proximity using mobility traces from two sources concluded that our approach performs at least one order of magnitude faster than adapted existing work.
  • flag_united_kingdom   Towards planet-scale localization on smartphones with a partial radiomap June 25, 2012

    Location: 4th ACM international workshop on Hot topics in planet-scale measurement ' (HotPlanet '12), in conjunction with MobiSys '12, Lake District UK


    The majority of smartphone localization systems useAssistedGPS for fine-grained localization in outdoor spaces or WiFibased RSS (Received Signal Strength) technologies for coarsegrain positioning in indoor and outdoor spaces. The former precious energy from mobile devices, is strictly affected by the environment (e.g., cloudy day, forests, etc.) does not work in indoor spaces. The latter collects RSS WiFi beams within a user’s vicinity and transfers an vector to the server for localization, in which the position of the user is disclosed possibly violating users’ privacy. this paper, we present BloomMap, an innovative and effi- algorithm that conducts a localization process without the user’s location to the localization service, minimizing the energy consumption of the mobile unit and also the network traffic by not transferring large positioning structures to the client (i.e., known as radiomap). framework is designed for planet-scale RSS localization which are expected to emerge in the near-future. particular, a user may localize itself using a subset of a data repository of RSS signals that is updated in real by smartphone wardrivers. Our preliminary evaluation that our propositions can localize a device without unveiling its location in approximately 80% less time energy network resources than competitive approaches. We describe our WiFi-based prototype system developed the Android OS. Categories and Sub.
  • flag_sweden   Disclosure-free GPS Trace Search in Smartphone Networks June 7, 2011

    Location: The 12th IEEE International Conference on Mobile Data (IEEE MDM '11), Lulea Sweden


    In this paper we present a powerful distributed framework for finding similar trajectories in a smartphone network, without disclosing the traces of participating users. Our framework, coined SmartTrace, exploits opportunistic and participatory sensing in order to quickly answer queries of the form: 'Report the users that move more similar to Q, where Q is some query trace '. SmartTrace, relies on an in-situ data storage model, where geo-location data is recorded locally on smartphones for both performance and data-disclosure reasons. SmartTrace then deploys an efficient top-K query processing algorithm that exploits distributed trajectory similarity measures, resilient to spatial and temporal noise, in order to derive the most relevant answers to Q quickly and efficiently. We assess our ideas with realistic and real workloads from Microsoft Research Asia and other sources. Our study reveals that SmartTrace computes the desired results with 74% less energy consumption and 13% faster than its centralized and decentralized counterparts. Our experimental results also confirm our analytical study.
  • flag_sweden   Multi-Objective Query Optimization in Smartphone Social Networks June 7, 2011

    Location: The 12th IEEE International Conference on Mobile Data (IEEE MDM '11), Lulea Sweden


    The bulk of social network applications for smartphones (e.g., Twitter, Facebook, Foursquare, etc.) currently rely on centralized or cloud-like architectures in order to carry out their data sharing and searching tasks. Unfortunately, the given model introduces both data-disclosure concerns (e.g., disclosing all captured media to a central entity) and performance concerns (e.g., consuming precious smartphone battery and bandwidth during content uploads). In this paper, we present a novel framework, coined SmartOpt, for searching objects (e.g., images, videos, etc.) captured by the users in a mobile social community. Our framework, is founded on an in-situ data storage model, where captured objects remain local on their owner 's smartphones and searches then take place over a novel lookup structure we compute dynamically, coined the Multi-Objective Query Routing Tree (MO-QRT). Our structure concurrently optimizes several conflicting objectives (i.e., it minimizes energy consumption, minimizes search delay and maximizes query recall), using a Multi-objective Evolutionary Algorithm based on Decomposition (MOEA/D) that calculates a diverse set of high quality non-dominated solutions in a single run. We assess our ideas with mobility patterns derived by Microsoft 's Geolife project and social patterns derived by DBLP. Our study reveals that SmartOpt can yield query recall rates of 95%, with one order of magnitude less time and two orders of magnitude less energy than its competitors.
  • flag_cyprus   Query Routing Trees for Wireless Sensor Networks February 15, 2011

    Location: EPL671 Course, Department of Computer Science, University of Cyprus, Nicosia Cyprus


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor the physical world at an extremely high fidelity. In order to collect the data generated by these tiny-scale devices, sensors are typically organized in structures coined Query Routing Trees (QRTs). Our study reveals that predominant data acquisition systems construct QRTs in ad-hoc manners leading to a significant waste of energy. In this talk I will present MicroPulse+, a framework for minimizing the consumption of energy during data collection in Sensor Networks. MicroPulse+ eliminates a variety of data transmission and data reception inefficiencies using a collection of in-network algorithms. In particular, MicroPulse+ introduces: i) the Workload-Aware Routing Tree (WART) algorithm, which is established on profiling recent data collection activity and on identifying the bottlenecks using an in-network execution of the critical path method; and ii) the Energy-driven Tree Construction (ETC) algorithm, which balances the workload among nodes and minimizes data collisions. The talk will conclude with an outlook into current and future research work.
  • flag_usa   MHS: Minimum-Hot-Spot Query Trees for Wireless Sensor Networks June 6, 2010

    Location: The 9th International ACM Workshop on Data Engineering for Wireless and Mobile Access (MobiDE '10), withACM SIGMOD/PODS10, Indianapolis, Indiana USA


    We present a novel distributed algorithm (MHS) that constructs a query routing tree that minimizes collisions during query execution. It was shown in previous work that minimizing collisions during query execution saves significant amount of energy[1]. In the same paper it is shown that balancing the node degrees of a query routing tree significantly reduces collisions during query execution. We address the inefficiencies of the previously proposed algorithm and propose a simpler, purely distributed, parameter-free, cheaper and more efficient algorithm. Our resulting query trees are optimally balanced, guarantee minimum collisions and minimum latency for query execution and allow for opportunistic in-network processing. MHS poses the minimum possible communication overhead to the network and is parameter-free as opposed to previously proposed algorithms. Our proposed algorithm can be used for acquiring data from the nodes of any distributed systems where the main objective is to minimize the communication cost.
  • flag_usa   MDM '10 Program Report May 25, 2010

    Location: The 11th International Conference on Mobile Data Management (MDM 2010), Kansas City USA

  • flag_france   FSort: External Sorting on Flash-based Sensor Devices August 24, 2009

    Location: The 6th Intl. Workshop on Data Management for Sensor Networks (DMSN09), with VLDB09, Lyon France


    In long-term deployments of Wireless Sensor Networks, it is often more efficient to store sensor readings locally at each device and transmit those readings to the user only when requested (i.e., in response to a user query). Many of the techniques that collect information from a sensor network require that the data is sorted on some attribute (e.g., range queries, top-k queries, join queries, etc.) Yet, the underlying storage medium of these devices (i.e., Flash media) presents some unique characteristics which renders traditional disk-based sorting algorithms inefficient in this context. In this paper we devise the FSort algorithm, an efficient external sorting algorithm for flash-based sensor devices with a small memory footprint. FSort minimizes the expensive write/delete operations of flash memory minimizing in that way the consumption of energy. In particular, FSort uses a top-down replacement selection algorithm in order to produce sorted runs on flash media in a log-based manner. Sorted runs are then recursively merged in order to yield the sorted result. Our experimentation with real traces from Intel Research Berkeley show that FSort greatly outperforms the traditional External Mergesort Algorithm both in regards to time and energy consumption. We found similar advantages in regards to the wearability constraints of flash media.
  • flag_taiwan   Perimeter-based Data Acquisition and Replication in Mobile Sensor Networks May 20, 2009

    Location: The 10th International Conference on Mobile Data Management (MDM '09), Taipei Taiwan


    This paper assumes a set of n mobile sensors that move in the Euclidean plane as a swarm. Our objectives are to explore a given geographic region by detecting spatio-temporal events of interest and to store these events in the network until the user requests them. Such a setting finds applications in mobile environments where the user (i.e., the sink) is infrequently within communication range from the field deployment. Our framework, coined SenseSwarm, dynamically partitions the sensing devices into perimeter and core nodes. Data acquisition is scheduled at the perimeter, in order to minimize energy consumption, while storage and replication takes place at the core nodes which are physically and logically shielded to threats and obstacles. To efficiently identify the nodes laying on the perimeter of the swarm we devise the Perimeter Algorithm (PA), an efficient distributed algorithm with a low communication complexity. For storage and fault-tolerance we devise the Data Replication Algorithm (DRA), a voting-based replication scheme that enables the exact retrieval of events from the network in cases of failures. Our trace-driven experimentation shows that our framework can offer significant energy reductions while maintaining high data availability rates. In particular, we found that when failures are less than 60\% failure then we can recover over 80\% of generated events exactly.
  • flag_taiwan   ETC: Energy-driven Tree Construction in Wireless Sensor Networks May 20, 2009

    Location: SenTIE '09 workshop, with IEEE MDM '09, Taipei Taiwan


    Continuous queries in Wireless Sensor Networks (WSNs) are founded on the premise of Query Routing Tree structures (denoted as T ), which provide sensors with a path to the querying node. Predominant data acquisition systems for WSNs construct such structures in an ad-hoc manner and therefore there is no guarantee that a given query workload will be distributed equally among all sensors. That leads to data collisions which represent a major source of energy waste. In this paper we present the Energy-driven Tree Construction (ETC) algorithm, which balances the workload among nodes and minimizes data collisions, thus reducing energy consumption, during data acquisition in WSNs. We show through real micro-benchmarks on the CC2420 radio chip and trace-driven experimentation with real datasets from Intel Research and UC-Berkeley that ETC can provide significant energy reductions under a variety of conditions prolonging the longevity of a wireless sensor network.
  • flag_cyprus   Indexing and Searching in Wireless Sensor Networks February 14, 2008

    Location: Department of Computer Science, University of Cyprus, Nicosia Cyprus


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor the physical world at an extremely high fidelity. Research in this area has to this day primarily focused on the trade-off between local computation and communication in order to minimize the transfer of data over the fundamentally expensive wireless link. On the contrary, we focus on the challenges of storing sensor readings locally at each node. This In-Situ storage paradigm offers a novel perspective for conserving energy in Wireless Sensor Networks as the communication channel is only accessed for answering on-demand queries rather than for percolating each and every event to a centralized database. Storing large quantities of data locally at each sensor has to be complemented by efficient access methods that will speed up the execution of queries when required. In this talk I will provide an overview of recent developments in Wireless Sensor Network Technology and highlight some important data indexing and searching challenges that arise in this context. In particular, I will present MicroHash which is an external memory index structure that is tailored to the distinct characteristics of flash memory, the most prevalent type of non-volatile memory used in sensor systems.
  • flag_cyprus   ICGrid: Towards a Grid Infrastructure for Intensive Care Units January 21, 2008

    Location: Intensive Care Forum, Hilton, Nicosia Cyprus


    ICGrid (Intensive Care Grid) is a distributed platform that enables the seamless integration, correlation and retrieval of clinically interesting episodes across Intensive Care Units, which is currently under development by our group. Such a task requires huge processing and data storage capabilities, which are common attributes of Grid infrastructures. ICGrid is based on a hybrid architecture that combines i) a heterogeneous set of monitors that sense the inpatients and ii) Grid technology that enables the storage, processing and information sharing task between Intensive Care Units.
  • flag_france   Grid Failure Monitoring and Ranking using FailRank January 15, 2008

    Location: Coregrid Network of Excellence, Paris France


    The objective of Grid computing is to make processing power as accessible and easy to use as electricity and water. The last decade has seen an unprecedented growth in Grid infrastructures which nowadays enables large-scale deployment of applications in the scientific computation domain. One of the main challenges in realizing the full potential of Grids is to make these systems \em dependable. In this presentation we present \em FailRank a nove
  • flag_austria   SenseSwarm: A Perimeter-based Data Acquisition Framework for Mobile Sensor Networks September 24, 2007

    Location: The 4th Intl. Workshop on Data Management for Sensor Networks (DMSN '07), with VLDB '07, Vienna Austria


    This paper assumes a set of $n$ mobile sensors that move in the Euclidean plane as a swarm. Our objectives are to explore a given geographic region by detecting and aggregating spatio-temporal events of interest and to store these events in the network until the user requests them. Such a setting finds applications in environments where the user (i.e., the sink) is infrequently within communication range from the field deployment. Our framework, coined SenseSwarm, dynamically partitions the sensing devices into perimeter and core nodes. Data acquisition is scheduled at the perimeter in order to minimize energy consumption while storage and replication takes place at the core nodes which are physically and logically shielded to threats and obstacles. To efficiently identify the perimeter of the swarm we devise the Perimeter Algorithm (PA), an efficient distributed algorithm with a message complexity of O(p + n), where p denotes the number of nodes on the perimeter and $n$ the overall number of nodes. For storage and replication we devise a spatio-temporal in-network aggregation scheme based on minimum bounding rectangles and minimum bounding cuboids. Our trace-driven experimentation shows that our framework can offer significant energy reductions while maintaining high data availability rates.
  • flag_cyprus   Distributed Spatio-Temporal Similarity Search July 4, 2007

    Location: Cyprus Summer School on Intelligent Systems, Department of Computer Science, Nicosia Cyprus


    In this talk I will introduce the distributed spatio-temporal similarity search problem: given a query trajectory Q, we want to find the trajectories that follow a motion similar to Q, when each of the target trajectories is segmented across a number of distributed nodes. We propose two novel algorithms, UB-K and UBLB-K, which combine local computations of lower and upper bounds on the matching between the distributed subsequences and Q. Such an operation generates the desired result without pulling together all the distributed subsequences over the fundamentally expensive communication medium. Our solutions find applications in a wide array of domains, such as cellular networks, wildlife monitoring and video surveillance. Our experimental evaluation using realistic data demonstrates that our framework is both efficient and robust to a variety of conditions. In this talk, I will also present techniques to efficiently answer Top-K queries in a distributed environment. A Top-K query returns the K highest ranked answers to a user defined similarity function. At the same time it also minimizes some cost metric, such as the utilization of the communication medium, which is associated with the retrieval of the desired answer set. I will provide an overview of state-of-the-art algorithms that solve the Top-K problem in a centralized setting and show why these are not applicable to the distributed case. I will then focus on the Threshold Join Algorithm (TJA), which is a novel solution for executing Top-K queries in a distributed environment. I will also present results from our performance study with a real middleware testbed deployed over a network of 75 workstations.
  • flag_greece   FailRank: Towards a Unified Grid Failure Monitoring and Ranking System June 12, 2007

    Location: CoreGRID Workshop on Grid Programming Model Grid and P2P Systems Architecture Grid Systems, Tools and Environments, Crete Greece


    The objective of Grid computing is to make processing power as accessible and easy to use as electricity and water. The last decade has seen an unprecedented growth in Grid infrastructures which nowadays enables large-scale deployment of applications in the scientific computation domain. One of the main challenges in realizing the full potential of Grids is to make these systems \em dependable. In this paper we present \em FailRank a nove
  • flag_germany   The MicroPulse Framework for Adaptive Waking Windows in Sensor Networks May 11, 2007

    Location: The 1st IEEE International Workshop on Data Intensive Sensor Networks 2007, with MDM 2007, Mannheim Germany.


    In this paper we present MicroPulse, a novel framework for adapting the waking window of a sensing device S based on the data workload incurred by a query Q. Assuming a typical tree-based aggregation scenario, the waking window is defined as the time interval t during which S enables its transceiver in order to collect the results from its children. Minimizing the length of t enables S to conserve energy that can be used to prolong the longevity of the network and hence the quality of results. Our method is established on profiling recent data acquisition activity and on identifying the bottlenecks using an in-network execution of the Critical Path Method. We show through trace-driven experimentation with a real dataset that MicroPulse can reduce the energy cost of the waking window by three orders of magnitude.
  • flag_germany   MINT Views: Materialized In-Network Top-k Views in Sensor Networks May 11, 2007

    Location: he 8th International Conference on Mobile Data Management (MDM '07), Mannheim Germany


    In this paper we introduce MINT (Materialized In-Network Top-k) Views, a novel framework for optimizing the execution of continuous monitoring queries in sensor networks. A typical materialized view V maintains the complete results of a query Q in order to minimize the cost of future query executions. In a sensor network context, maintaining consistency between V and the underlying and distributed base relation R is very expensive in terms of communication. Thus, our approach focuses on a subset V ' (\subseteq V) that unveils only the k highest-ranked answers at the sink for some user defined parameter k. We additionally provide an elaborate description of energy-conscious algorithms for constructing, pruning and maintaining such recursively-defined in-network views. Our trace-driven experimentation with real datasets show that MINT offers significant energy reductions compared to other predominant data acquisition models.
  • flag_cyprus   Top-K Algorithms: Concepts and Applications March 20, 2007

    Location: Nicosia Cyprus (EPL 671 - Computer Science: Research and Technology)


    In this talk, I will present techniques to efficiently answer Top-K queries in a distributed environment. A Top-K query returns the K highest ranked answers to a user defined similarity function. At the same time it also minimizes some cost metric, such as the utilization of the communication medium, which is associated with the retrieval of the desired answer set. I will provide an overview of state-of-the-art algorithms that solve the Top-K problem in a centralized setting and show why these are not applicable to the distributed case. I will then focus on the Threshold Join Algorithm (TJA), which is a novel solution for executing Top-K queries in a distributed environment. I will also present results from our performance study with a real middleware testbed deployed over a network of 75 workstations.
  • flag_cyprus   MicroHash: An External Memory Indexing Structure for Wireless Sensor Devices April 26, 2007

    Location: Nicosia, Cyprus (EPL651 - Data Management for Mobile Computing Department of Computer Science (UCY))


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor and understand the physical world at an extremely high fidelity. Research to this day has primarily focused on the trade-off between local computation and communication, in order to offset the expensive transfer of data over the fundamentally unreliable wireless link. On the contrary, we focus on the challenges of storing sensor readings locally at each node. This In-Situ storage paradigm offers a novel perspective for conserving energy, as we access the communication channel to answer on-demand queries rather than for percolating each and every event to a centralized database. Storing large quantities of data locally at each node has to be complemented by efficient index structures that will enable access to data when required. In this talk we present MicroHash, an external memory index structure which is tailored to the distinct characteristics of the most prevalent type of non-volatile memory used in sensor systems, namely flash memory. Our index structure exploits the asymmetric read/write and wear characteristics of flash memory in order to offer high performance indexing and searching capabilities in the presence of a low energy budget.
  • flag_france   ICGrid: Intensive Care Grid (Best Demo) December 1, 2006

    Location: Sophia-Antipolis France (CoreGRID Industrial Conference)


    Intensive Care Units (ICUs) at hospitals utilize cutting edge technology in order to acquire the physiological state of inpatients, which are in a critical (life-threatening) physiological state, at an extremely high fidelity. In particular, ICUs utilize a very large number of monitoring and sensing devices that are continuously attached on inpatients in order to uncover the physiological state of the inpatients. Such measurements can then be utilized for i) education, ii) early diagnosis and iii) for defining early warning systems that identify when a human life is jeopardy. A problem with the current setting is that individual ICUs are limited to the locally acquired measurements. As a result, the number of clinically 'interesting ' episodes available to doctors is also very limited. ICGrid (Intensive Care Grid) is a distributed platform that enables the seamless integration, correlation and retrieval of clinically interesting episodes across Intensive Care Units, which is currently under development by our group. Such a task requires huge processing and data storage capabilities, which are common attributes of Grid infrastructures. ICGrid is based on a hybrid architecture that combines i) a heterogeneous set of monitors that sense the inpatients and ii) Grid technology that enables the storage, processing and information sharing task between Intensive Care Units. Our demonstration aims at presenting the first part of the hybrid architecture of ICGrid (i.e. the acquisition of real signals from inpatients and their storage on the Grid). Our demonstration platform operates on a standalone laptop. In a real setting, this software is able to extract the physiological parameters from monitoring devices installed at ICUs.
  • flag_cyprus   Business Processes: Behavior Prediction and Capturing Reasons for EvolutionMay 24, 2006

    Location: The 8th International Conference on Enterprise Information Systems, Paphos Cyprus

  • flag_cyprus   MicroHash: An External Memory Indexing Structure for Wireless Sensor Devices March 31, 2006

    Location: Department of Computer Science, University of Cyprus, Nicosia Cyprus


    Wireless Sensor Networks offer a non-intrusive and non-disruptive technology that enables users to monitor and understand the physical world at an extremely high fidelity. Research to this day has primarily focused on the trade-off between local computation and communication, in order to offset the expensive transfer of data over the fundamentally unreliable wireless link. On the contrary, we focus on the challenges of storing sensor readings locally at each node. This In-Situ storage paradigm offers a novel perspective for conserving energy, as we access the communication channel to answer on-demand queries rather than for percolating each and every event to a centralized database. Storing large quantities of data locally at each node has to be complemented by efficient index structures that will enable access to data when required. In this talk we present MicroHash, an external memory index structure which is tailored to the distinct characteristics of the most prevalent type of non-volatile memory used in sensor systems, namely flash memory. Our index structure exploits the asymmetric read/write and wear characteristics of flash memory in order to offer high performance indexing and searching capabilities in the presence of a low energy budget.
  • flag_cyprus   Data Storage in Sensor Databases March 14, 2006

    Location: eNEXT Workshop on Sensor and Ad-hoc Networks, Nicosia Cyprus

  • flag_cyprus   Global Internet Content DeliveryNovember 22, 2005

    Location: EPL602 Course, Department of Computer Science, University of Cyprus, Nicosia Cyprus

  • flag_cyprus   Distributed Top-K Query Processing November 16, 2005

    Location: Department of Computer Science, University of Cyprus, Nicosia Cyprus


    Modern Sensor and Peer-to-Peer data management systems have to cope with data that is generated automatically and continuously across distributed and potentially geographically diverse locations. Organizing data in centralized repositories is becoming increasingly expensive and in many occasions impractical. Additionally, users are usually only interested in finding the highest ranked answers to their queries rather that the complete range of answers. In this talk, I will present efficient techniques to answer Top-K queries in a distributed environment. A Top-K query returns the K highest ranked answers to a user defined similarity function. At the same time it also minimizes some cost metric which is associated with the retrieval of the desired answer set. My talk focuses on the Threshold Join Algorithm (TJA), which is a novel distributed Top-K query processing algorithm that combines local similarity scores available at each computing site. I will also present the LB-K and UBLB-K algorithms which utilize lower and upper bounds, when exact scores are not available. An extensive experimental evaluation with our distributed middleware testbed reveals that the proposed methods are orders of magnitudes more efficient than their competitors.
  • flag_canada   On Constructing Internet-Scale P2P Information Retrieval Systems September 2004

    Location: Second International Workshop on Databases, Information Systems, and Peer-to-Peer Computing (DBISP2P 2004), Toronto Canada


    We initiate a study on the effect of the network topology on the performance of Peer-to-Peer (P2P) information retrieval systems. The emerging P2P model has become a very powerful and attractive paradigm for developing Internet-scale systems for sharing resources, including files, or documents. We show that the performance of Information Retrieval algorithms can be significantly improved through the use of fully distributed topologically aware overlay network construction techniques. Our empirical results, using the Peerware middleware infrastructure, show that the approach we propose is both efficient and practical.
  • flag_usa   A Local Search Mechanism for Peer-to-Peer Networks November 2002

    Location: The 11th ACM CIKM International Conference on Information and Knowledge Management, McLean, VA USA.


    One important problem in peer-to-peer (P2P) networks is searching and retrieving the correct information. However, existing searching mechanisms in pure peer-to-peer networks are inefficient due to the decentralized nature of such networks. We propose two mechanisms for information retrieval in pure peer-to-peer networks. The fir, the modified Breadth-First-Search (BFS) mechanism, is an extension of the current Gnuttela protocol, allows searching with keywords, and is designed to minimize the number of messages that are needed to search the network. The second, the Intelligent Search mechanism, uses the past behavior of the P2P network to further improve the scalability of the search procedure. In this algorithm, each peer autonomously decides which of its peers are most likely to answer a given query. The algorithm is entirely distributed, and therefore scales well with the size of the network. We implemented our mechanisms as middleware platforms. To show the advantages of our mechanisms we present experimental results using the middleware implementation.
  • flag_cyprus   A Quantitative Analysis of the Gnutella Network TrafficJuly 2002

    Location: Department of Computer Science, University of Cyprus, Nicosia Cyprus