Software and Communication Platforms for High-Performance Collaborative Grid

Pavia team

First year activity

  • WP1 - Device level
    • Development of wavelength converters
  • WP1 - Infrastructure level
    • Development of coding and transmission techniques for multimedia content
  • WP1 - Application level
    • Investigation on possible applications of perceptive interfaces to e-learning environments
    • Development of perceptive interfaces and graphic environments for e-learning applications
    • Use of eye-tracking techniques for usability studies

Summary of achieved results

At the device level, the activity has been mainly focused on the development of wavelength converters and their utilization in optical-communication systems. The converters are based on a cascade of second-order nonlinear processes taking place in a periodically-poled lithium-niobate channel waveguide. The aim of the research activity was that of eliminating (or strongly reducing) the photorefractivity of the nonlinear crystal, in order to be able to operate the converter at room temperature. We have demonstrated that doping the crystal with Hafnium Oxide at a molar concentration between 2 and 3% strongly reduces photorefractivity; we are now investigating whether doping is affecting appreciably the nonlinear response. Also, we have carried out an experiment demonstrating that the insertion of the wavelength converter in an appropriate position within the optical-communication system can produce a compensation not only of optical dispersion, but also of the distortions due to thenonlinear interactions occurring inside the monomodal optical fiber. Bit-error-rate (BER) measurements performed on a 600-km line with 100-km spans have confirmed the effectiveness of the proposed approach in a single-channel transmission system with RZ modulation format and 10 Gb/s transmission rate.

As regards the infrastructure level, our unit has studied and implemented coding techniques for multimedia content, video in particular, able to support adaptive techniques for quality of service control and robustness in case of transmission errors. In addition, we have studied methods for content distribution, based on multicast and peer-to-peer networks, able to reduce the needed overall bandwidth.

In the context of the application level, the Pavia unit has worked mainly on the activity "1.B: Real-time collaborative e-learning environments" of WP1. In particular, in-depth studies have been carried out about the possible applications of perceptive interfaces to e-learning ([3] and [12]), so as to make the most of their potentials. From a practical point of view, an original solution has been implemented for explicit communication through the combination of mouse and hand gestures. For the identification of implicit signals in the communication between the user and the computer (e.g. stress situations or difficulty in understanding something), studies have been started which exploit eye tracking as a source of data for thorough analysis of user behaviour. In addition, eye tracking techniques have also been used for usability studies. As concerns visual communication, a graphic enviroment has been implemented which helps in learning the basics of relational database interaction.

Research products developed

  1. Detailed simulators of multicast peer-to-peer networks and video source coders able to provide, in a flexible and combined way, different coding possibilities characterized by multiple descriptions and scalabilities based on the H.264 coder.
  2. GEM - Gesture-Enhanced Mouse, a vision-based perceptive interface (implemented in Java) for the recognition of hand gestures performed near the mouse; the purpose is to improve the interaction through more natural communication techniques.
  3. GraphSQL Builder, a graphic environment which allows both direct visual composition of queries and automatic visualization of textual SQL queries. The purpose of the system, implemented in C#, is mainly to help the inexperienced user learn the interaction modalities with relational databases, thanks to the expressive power of graphic representations.
With regard to GEM and GraphSQL Builder, two papers have been submitted (acceptance pending) to, respectively, ICIAP '07 (International Conference on Image Analysis and Processing, Modena, Italy, September 2007) and MIMIC'07 (International Workshop on Management and Interaction with Multimodal Information Content, Regensburg, Germany, Settebre 2007).

Publications

  1. P. Minzioni, I. Cristiani, V. Degiorgio, L. Marazzi, M. Martinelli, C. Langrock, M. M. Fejer, "Experimental demonstration of nonlinearity and dispersion compensation into an embedded link by optical phase conjugation", IEEE Photonics Technology Letters 18 (9), 995-997 (2006)
  2. P. Galinetto, F. Rossella, I. Cristiani, P. Minzioni, V. Degiorgio, E. P. Kokanyan, "Structural and optical properties of Hafnium-doped Lithium-Niobate crystals", Physica Status Solidi (to be published)
  3. M. Porta, "E-learning and machine perception: in pursuit of human-like interaction in computer-based teaching systems", International Journal of Knowledge and Learning, Inderscience Publishers (to be published)
  4. F. Bertagnolli, M. Lanati, L. Favalli, F. Gruppi, "Open Loop Transcoding Using Perceptually Adaptive Requantization of DCT Coefficients," Proc. MobiMedia Conference 2006, Alghero, Sardinia (Italy), 18-20 September 2006
  5. P. Minzioni, I. Cristiani, V. Degiorgio, L. Marazzi, A. Colciago, M. Martinelli, C. Langrock, M. M. Fejer, "Efficient Nonlinearity Cancellation through Optical Phase Conjugation into an Embedded Link with Asymmetrical Power Profiles", Proc. Conference on Optical Fiber Communications (OFC/NFOEC), 5-10 March 2006, Anaheim, California, USA, paper OthA7
  6. P. Minzioni, I. Cristiani, V. Degiorgio, L. Marazzi, M. Martinelli, C. Langrock, M. M. Fejer, "Simultaneous Nonlinearity and Dispersion Compensation into an Embedded Link: Experimental Demonstration", Proc. Conference on laser and electro-optics (CLEO), 21-26 May 2006, Long Beach, California, USA, paper CThE2
  7. P. Minzioni, L. Razzari, I. Cristiani, V. Degiorgio, P. Galinetto, F. Rossella, A. Morbiato, A. De Simone, E. P. Kokanyan, "Hafnium-Doped Lithium Niobate Crystals: Electrical and Optical Properties", Proc. Conference on laser and electro-optics (CLEO), 21-26 May 2006, Long Beach, California, USA, paper CTuJ5
  8. P. Minzioni, L. Razzari, I. Cristiani, V. Degiorgio, P. Galinetto, F. Rossella, E. P. Kokanyan, "Hafnium-Doped Lithium Niobate Crystals: Electrical and Optical Properties", Proc. 2006 China-Italy Bilateral Workshop on Photonics for Communication and Sensing, 23-24 October 2006, Xian, China
  9. P. Minzioni, I. Cristiani, V. Degiorgio, L. Marazzi, M. Martinelli, C. Langrock, M. M. Fejer, "Nonlinearity Cancellation in an Embedded Link with Asymmetrical Power Profiles: Experimental Demonstration of two new Techniques based on Optical Phase Conjugation", Proc. 2006 China-Italy Bilateral Workshop on Photonics for Communication and Sensing, 23-24 October 2006 Xian, China
  10. P. Galinetto, F. Rossella, I. Cristiani, P. Minzioni, V. Degiorgio, E. P. Kokanyan, "Structural and optical properties of Hafnium-doped Lithium-Niobate crystals", Proc. 10th Europhysical Conference on Defects in Insulating Materials, 10-14 July 2006, Milano
  11. F. Rossella, P. Galinetto, D. Grando, V. Degiorgio, E. P. Kokanyan, "Photoconductive and electro-optical properties of Hf doped lithium niobate crystals", Proc. 8th European Conference on Applications of Polar Dielectrics, 5-8 September 2006, Metz, France
  12. M. Porta, "Exploiting Vision and Speech in Advanced E-learning Systems", Proc. 3rd E-learning Conference, Coimbra, Portugal, 7th - 8th September, 2006

Dissemination

The results were presented and discussed during the first Plenary Meeting project held in Tunis on March 29th, 2007. The presentations given by Prof. Vittorio Degiorgio and Dr. Marco Porta are available for download.

Second year activity

Summary of achieved results

Several research units focused their activities on specific problems.

One group came to the following results: building of a set of tools for encoding video sequences according to the H.264 standard where it is possible to simultaneously exploit multiple descriptions encoding scalability techniques; realization of an amendment to a multicast algorithm at the application level that allows to control the construction of multicast trees under asymmetric in the contest of activities regarding peer-to-peer content distribution; startup recently, in collaboration with the unit of Pisa and the University of Manouba, of a campaign for measurement and characterization of inbound traffic to the respective universities gateway to pinpoint potential attacks.

Another research group of the Pavia unit followed two research lines, the first devoted to the development of wavelength converters to be used in optical-communication systems, the second concerning new types of optical fibers with enhanced nonlinear properties.
The converters are based on a cascade of second-order nonlinear processes taking place in a periodically-poled lithium-niobate channel waveguide. The aim of the research activity was that of eliminating (or strongly reducing) the photorefractivity of the nonlinear crystal, in order to be able to operate the converter at room temperature. Through collaborations with groups of crystal growers, we had available a set of lithium niobate crystals doped with increasing concentrations of hafnium oxide. Our measurements of linear and nonlinear optical properties have clearly shown that crystals doped at a molar concentration above 2% possess the desired properties. Experiments about periodical poling and waveguide fabrication are now in progress.
In collaboration with the CirtaCom Laboratory of the Engineering School of Communication of Tunis, has been experimentally and numerically studied the propagation of femtosecondi pulses in microstrutturate fibres with core diameter of 2.5 µm. Since fibres used allow a selective excitation of different modes of propagation, it was possible to study the generation of a supercontinuo spectrum of frequency exciting ways of higher order. The supercontinuo spectrum may be of interest to produce, from a single lase and channel spacing adjustable, the whole set of wavelengths used in a system of WDM optical communications.

As regards the application component of the project, concerning e-learning environments, in the second year of activity the Pavia unit has mainly focused on eye tracking methodologies. In particular, the work carried out can be divided into three main kinds of research: 1) development of "intelligent" e-learning environments, able to recognize users' "behaviours" and keep track of their activities; 2) implementation of active interfaces for pure eye-based control of the computer; and 3) use of eye tracking techniques for the analysis and evaluation of information presentation modes.
As for point 1, we have created e5Learning, a prototype e-learning environment characterized by three key functionalities: a) detection of basic user activities, such as reading text and observing multimedia content, in order to maintain a "history" of user actions and guide the student in the learning process; b) generation of additional contents depending on the context (e.g. textual or multimedia descriptions shown when the user is reading a specific portion of text, so as to limit the simultaneous presentation of too much information; and c) recognition of stress, high workload and tiredness states in the user, using physiological data obtainable from the eye tracker (pupil size, blink rate, fixation frequency and speed, saccade speed).
As regards point 2, we have developed Eye-S, a system for pure eye-based communication (allowing both text and generic commands input). Eye-S is thought for users who cannot move, and therefore need alternative input methods rather than those based on keyboard and mouse. Using an "eye gesture" approach, the user creates alphabet letters, as well as punctuation marks or commands definable according to the particular application to control, by means of sequences of fixations on nine predefined screen areas. Compared to other eye writing systems, such as those based on virtual keyboards, with Eye-S the screen is totally free for the display of applications, since there is no need to display specific graphical input interfaces.
With reference to point 3, we have carried out several experiments to asses the effectiveness of different RSVP (Rapid Serial Visual Presentation) techniques for image presentation, aimed also at finding guidelines and design solutions for the implementation of browsing methods based on eye selection.
Moreover, in the Web context, we have studied on-line newspapers, and in particular how to optimally present multimedia content and advertising banners. Besides the specific fields in which they have been accomplished, such researches can provide important advices on the effective design of e-learning contents, more and more characterized by a marked multimedia component.

Research products developed

One of the Pavia research units has extensively used the eye-tracking methodology both as a means of assessment and as a means of communication in ocular interfaces based systems.
During the second year was also carried out an extensive simulation in order to investigate the effect of the insertion of the wavelength converter into a WDM optical-communication system with various modulation formats. The simulation data are helpful for the preparation of new experiments extending the results obtained during the first year.

In particular, have been developed and implemented:

  • E5Learning, an e-learning environment based on eye tracking. The system, developed in C # within the Microsoft .NET framework is composed of an "Author Tool", which allows the creator of the course to specify constraints on the content and on the user activities, and a "Learner Tool ", through which the user takes the course, using a special browser.
  • Eye-S, a system for writing and generation of generic commands through the exclusive use of the eyes. The software, developed in C # within the Microsoft .NET framework., it is widely configurable and adaptable to the specific user needs.
  • detailed networks multicast-peer-to-peer simulators and video source encoders able to flexibly provide and combine various multiple descriptions encoding and scalability based on H.264 encoder.

Publications

  1. C. Calvi, M. Porta, D. Sacchi, "e5Learning, an E-Learning Environment Based on Eye Tracking". Proceedings of the 8th IEEE International Conference on Advanced Learning Technologies, July 1st-5th 2008, Santander, Cantabria, Spain.
  2. S. Corsato, M. Mosconi, M. Porta, "An Eye Tracking Approach to Image Search Activities Using RSVP Display Techniques". Proceedings of the 9th International Working Conference on Advanced Visual Interfaces (AVI 2008), Naples, Italy, 23-26 May 2008, ACM Press.
  3. M. Porta, "Implementing Eye-Based User-Aware E-Learning". Proceedings of CHI 2008 (International Conference on Human Factors in Design) ext. abstr., Florence, Italy, April 5-10, 2008, ACM Press.
  4. M. Porta, M. Turina, "Eye-S: a Full-Screen Input Modality for Pure Eye-based Communication". Proceedings of the 5th Symposium on Eye Tracking Research & Applications (ETRA 2008), Savannah, GA, USA, March 26-28, 2008, ACM Press.
  5. M. Porta, "Real-Time Analysis of Eye Behaviour for Enhanced E-Learning". Journal of Communication and Cognition, Vol. 41, Nr. 1 & 2 (2008).
  6. R. Cherif, M. Zghal, L. Tartara, and V. Degiorgio "Supercontinuum generation by higher-order mode excitation in a photonic crystal fiber" Optics Express 16, 2147-2152 (2008).
  7. L. Tartara, V. Degiorgio, R. Cherif, M. Zghal, "Single-sided Supercontinuum Generation by High Order Mode Excitation in a Photonic Crystal Fiber", IEEE Winter Topical Meeting, Sorrento, Italy, 14-16 January 2008.
  8. F. Baroncelli, B. Martini, V. Martini, and P. Castoldi, "A novel application-to-network interaction paradigm for the on-demand provisioning of VPN services in transport networks", Annual Review of Communications, vol.60, International Engineering Consortium (IEC), December 2007.
  9. P. Galinetto, F. Rossella, I. Cristiani, P. Minzioni, V. Degiorgio, and E. P. Kokanyan, "Structural and optical properties of hafnium-doped lithium-niobate crystals" physica status solidi (c) 4, 1372-1375 (2007).
  10. F. Rossella, D. Grando, P. Galinetto, V. Degiorgio, and E. P. Kokanyan, "Photoconductive and Electro-optical Properties of Hf Doped Lithium-Niobate Crystals", Ferroelectrics 352, 143-147 (2007).
  11. P. Minzioni, I. Cristiani, V. Degiorgio, and E.P. Kokanyan "Strongly sublinear growth of the photorefractive effect for increasing pump intensities in doped lithium-niobate crystals", Journal of Applied Physics 101, 116105 (2007).
  12. P. Minzioni, I. Cristiani, J. Yu, J. Parravicini, E. P. Kokanyan and V. Degiorgio, "Linear and nonlinear optical properties of Hafnium-doped lithium-niobate crystals" Optics Express 15, 14171-14176 (2007).
  13. L. Tartara, V. Degiorgio, R. Cherif, M. Zghal, "Single-sided Supercontinuum Generation by High Order Mode Excitation in a Photonic Crystal Fibre", invited talk at the International Conference on Transparent Optical Networks (ICTON-MW), Sousse, Tunisia, 6-8 December 2007.
  14. L. Tartara, V. Degiorgio, R. Cherif, M. Zghal, "Single-sided Supercontinuum Generation by High Order Mode Excitation in a Microstructured Fiber", invited talk at the National Laser Symposium, Vadodara, Gujarat, India, 17-20 December 2007.
  15. I. Cristiani, P. Minzioni, J. Parravicini, V. Degiorgio, E. P. Kokanyan, "Hafnium-Doped Lithium Niobate Crystals For Nonlinear Optics Applications", Optical Micro Systems 2007, Capri, 30 September - 3 October 2007.
  16. L. Lombardi, M. Porta, "Adding Gestures to Ordinary Mouse Use: a New Input Modality for Improved Human-Computer Interaction". Proceedings of the 14th IEEE International Conference on Image Analysis and Processing (ICIAP '07), Modena, Italy, 11-13 September 2007.
  17. C. Cerullo, M. Porta, "A System for Database Visual Querying and Query Visualization: Complementing Text and Graphics to Increase Expressiveness". Proceedings of the IEEE DEXA Workshops 2007, Regensburg, Germany, 3-7 September 2007.
  18. M. Porta, "Eye Tracking for E-Learning: When the Eyes Can Tell a Lot About the User". Proceedings of the 4th E-Learning Conference, Istanbul, Turkey, 27-28 August 2007.
  19. L. Favalli, M. Folli, "Multiple Description Coding algorithms for H.264 coder," Proceedings of Mobimedia 2007 Conference, Nafpaktos, Greece, 27-29 agosto 2007.
  20. P. Minzioni, I. Cristiani, V. Degiorgio, L. Marazzi, M. Martinelli, C. Langrock, M.M. Fejer, "Compensation of Nonlinearity in an Optical-Communication Link by using Optical Phase Conjugation", Japan-Italy Bilateral Workshop on Photonics for Communication, Osaka, Japan, 5-7 July 2007.
  21. V. Degiorgio, "Photorefractive and Nonlinear Optical Properties of Hafnium-doped Lithium-Niobate Crystals", invited talk at "Novel Optical Materials and Applications", Cetraro, Italy, 3-8 June 2007.
  22. I. Cristiani, P. Minzioni, J. Parravicini, J. Yu, L. Tartara, V. Degiorgio, E.P. Kokanyan, "Nonlinear Coefficients of Hafnium doped Lithium Niobate Crystals", CLEO Europe 2007, Munich, Germany, 17-22 June 2007.
  23. P. Minzioni, I. Cristiani, V. Degiorgio, E.P. Kokanyan, "Photorefractive-resistant Hafnium-doped Lithium Niobate Crystals at very low dopant concentration", CLEO Europe 2007, Munich, Germany, 17-22 June 2007.
  24. L. Marazzi, P. Parolari, P. Martelli, A. Gatto, M. Martinelli, P. Minzioni, I. Cristiani, V. Degiorgio, "Impact of OPC insertion in a WDM link", CLEO Europe 2007, Munich, Germany, 17-22 June 2007.

Dissemination

Ample evidence of the participation in this project by the various units has been given in various fields at national and international level, including presentations made at various meetings of the Telecommunications Group, the Networks Group and the ISI Technology platform.
The research results acquired during the project were presented in the course "Multimedia Data Transmission" held at the University of Pavia, using slides compiled by one of the PhD student granted by the project. Seminars with students from the same course, with examples of the potential of multiple and scalable descriptions coding techniques, are planned.
Several students of different degree courses in Computer Engineering and in Publishing and Multimedia Communication of Pavia benefited of the results of this reseach regarding the application and e-learning environments topic.

Third year activity

Summary of achieved results

Within the "application" level, the activity of the Pavia unit has continued the research line started in the past two years, expanding the results previously obtained and enlarging the examined areas. In particular, the activity carried out can be grouped into the following macro categories: 1) use of eye tracking techniques for the evaluation of interfaces and information presentation strategies; 2) use of eye tracking techniques for the development of eye-based active interfaces; 3) development of algorithms for the recognition of user "states" and face expressions, to be used in "intelligent" e-learning systems; 4) implementation of vision-based perceptive interfaces, for a more intuitive interaction with the PC.
As regards point 1, researches have been carried out in order to evaluate different information presentation solutions for on-line newspapers, with special reference to multimedia contents and ad banners. Studies have also been performed on different subtitling and captioning strategies for audio visual contents, in collaboration with experts in linguistics. Altogether, more than 140 testers have been involved in the experiments. The results of such researches, partly closed and partly still to be analyzed, because of the size of the eye data obtained, can also be exploited for the definition of guidelines for the implementation of e-learning interfaces and contents. The identification of effective communication strategies is in fact a transversal problem, which can be found, in slightly different forms, in different contexts. Moreover, an international collaboration with Imperial College (London) and the Mitsubishi Research Laboratory of Cambridge (MA, USA) has been established, to assess various methods for the presentation of large amounts of images, using RSVP (Rapid Serial Visual Presentation) techniques studied throughout the last two years.
The activity of point 2 has produced a web browsing system totally controllable through the eyes. Thinking about e-learning platforms accessible to all kinds of users, also those affected by motor disabilities, the implemented prototype can contribute to increase the availability of e-learning contents, which are more and more widespread but that often cannot be exploited by those who could actually get most benefits from them. Moreover, studies are now being carried out aimed at comparing the Eye-S eye writing system, developed during the second year of activity, with two alternative systems currently most widespread and studied, namely, the on-screen keyboard and the EyeWrite method.
As for point 3, an algorithm has been developed which is able to recognize user "states". The implemented prototype, which uses a normal webcam to acquire data, distinguishes among the user states "present", "absent", "frontal to the screen", "back turned on the screen", "head turned to the left", "head turned to the right", "phoning". The main feature of this perceptive system is the semiautomatic initial calibration phase, always critical but necessary for vision-based perceptive interfaces, depending on lighting conditions and on the characteristics of the specific user. Integrated within an e-learning platform, a system recognizing user states can improve the quality of the interaction, better adapting to user behavior. Currently, an algorithm for the recognition of facial expressions is also under development, to be used (possibly along with the "eye states" analysis system studied in the second year of activity) for the identification of the user's emotional states.
Regarding point 4, at last, an integration of the so-called "sketch interfaces" with vision-based perceptive interfaces has been studied. In the developed prototype, the interface is drawn freehand by the user on an ordinary sheet of paper; however, the user can not only draw the graphical elements, but also interact with interface parts (sliders and buttons in the implemented prototype) directly on the sheet. This makes it possible more natural forms of interaction, which will potentially find application in the e-learning field (for instance, in evaluation tests). The activity of point 4 is within the WP1 and WP3 (B), and in particular in WP1.3B and in WP3.1.
The research line concerning the development of wavelength converters to be used in optical-communication systems has been mainly focused on the all-optical wavelength conversion of a 100-Gb/s polarization-multiplexed (POLMUX) signal. Each polarization channel is modulated by means of differential quadrature phase-shift keying at 25 Gbaud. The experiment, performed in collaboration with Politecnico di Milano and Stanford University, has produced very interesting results, which have been presented in an oral communication at the most important world meeting on optical communications (Optical Fiber Communication Conference, San Diego, March 2009). We have also continued studies devoted to improving the performance of the converter by properly doping the lithium niobate crystal.
The research line concerning optical fibers, developed in collaboration with the CirtaCom Laboratory of the Engineering School of Communication of Tunis, was addressed in particular to the role of higher-order modes in the nonlinear propagation. The study of supercontinuum generation by femtosecond laser pulses in microstructured fibers was brought to completion, and a new activity devoted to a detailed investigation of the gain spectrum of stimulated Brillouin scattering was started.
The activities started on the previous years have been continued with emphasys on the aspects of video coding and distribution by means of P2P networks. For what the video coding aspects are concerned, reference has been on the H.264 standard with hibrid multiple description and scalable techniques for which an optimal subdivision of the total bit rate has been achieved among base and enhancement layers and the evaluation of the possibiliy to generate unbalanced descriptions. The activities on the peer to peer networks have benefitted from the research on video coding and performance of the evolved ZigZag algorithm proposed in the previous year has been integrated with a scalable video coder to enhance reliability. Extension to multiple trees and multiple descriptions is ongoing. Also ongoing is the possibility to implement by the end of the project a demonstrator of the enhanced video coder based on a handheld device. Finally, the activity started with the units in Pisa and La Manouba regarding traffic measurements and characterizationhas been slowed by the differen measurement conditions.

Research products developed

Development of innovative software:
  • WeyeB (Web eye Browser), a pure eye-based web browsing system (implemented for the Tobii 1750 eye tracker).
  • System for the recognition of potential user "states" (namely "present", "absent", "frontal to the screen", "back turned on the screen", "head turned to the left", "head turned to the right", "phoning") by means of an ordinary webcam.
  • SIWHI (Sketch Interface With Hand Interaction), a vision-based perceptive interface integrated with a "sketch" approach, where the user can draw interface elements directly on a sheet of paper, and interact with them by moving the hand on the sheet.

Publications

  1. M. Porta, "New Visualization Modes for Effective Image Presentation". International Journal of Image and Graphics, Vol. 9, No. 1 (2009), pp. 1-23.
  2. E.P. Kokanyan, P. Minzioni, I Cristiani, V. Degiorgio, "Photorefractive Effect in the Single-Domain Doped Lithium Niobate Crystal: Dependence on the Light Intensity", Ferroelectrics 373, 32-36 (2008).
  3. M. Mosconi, M. Porta, A. Ravarelli, "On-Line Newspapers and Multimedia Content: An Eye Tracking Study". Proceedings of the 26th ACM International Conference on Design of Communication (SIGDOC 2008), Lisboa, Portugal, 22-24 September 2008, ACM Press.
  4. V. Cantoni, M. Porta, A. Ravarelli, "Enhancing Assistive Technology with Eye-Controlled Interfaces". Proceedings of the 2nd International Conference on Information and Communication Technologies and Accessibility, Hammamet, Tunisia, May 7-9, 2009 (to appear).
  5. M. Porta, A. Ravelli, "WeyeB, an Eye-Controlled Web Browser for Hands-Free Navigation". Proceedings of the 2nd IEEE International Conference on Human System Interaction, Catania, Italy, May 21-23 (to appear).
  6. R. Cherif, M. Zghal, L. Tartara, V. Degiorgio, "Single-sided Supercontinuum Generation by High Order Mode Excitation in a Photonic Crystal Fiber", in "Photonic Crystal Fibers II", edited by Kyriacos Kalli, Waclaw Urbanczyk, Proc. SPIE Vol. 6990, 699000-1 (2008).
  7. V. Degiorgio, L. Tartara, R. Cherif, M. Zghal "Single-sided supercontinuum generation in a photonic crystal fiber by selective excitation of the third-order mode", 1st Mediterranean Conference on Nanophotonics MediNano-1, Istanbul, Turkey, 6-7 October 2008.
  8. L. Tartara, C. Codemard, J.N. Maran, R. Cherif, M. Zghal "Full modal analysis of the Brillouin gain spectrum of an optical fiber", invited presentation, ICTON-MW'08 - International Conference on Transparent Optical Networks "Mediterranean Winter", December 11-13, 2008, Marrakech, Morocco.
  9. V. Pusino, P. Minzioni, I. Cristiani, V. Degiorgio, L. Marazzi, P. Boffi, M. Ferrario, P. Martelli, P. Parolari, A. Righetti, R. Siano, M. Martinelli, C. Langrock, M. M. Fejer "Wavelength Conversion of Real-Time 100 Gb/s POLMUX RZ-DQPSK", paper OThS3, Optical Fiber Communication, OFC/NFOEC, March 22-26, 2009, San Diego, California, USA.
  10. L.Favalli, M. Folli, M. Lanati, "Parameters optimization for scalable multiple description coding scheme based on spatial subsampling", Mobimedia 2009, July 7-9 2008, Oulu, Finland.
  11. M. Folli, L. Favalli, "Scalable multiple description coding of video sequences", Riunione annuale Gruppo Nazionale Telecomunicazioni e Teoria dell'informazione, Florence, Italy, June 16 - 18, 2008.
  12. L. Favalli , M. Folli, D. Reforgiato, G. Schembra, "A Bandwidth-Aware P2P Platform for the Transmission of Multipoint Multiple Description Video Streams", Italian networking workshop, Cortina d'Ampezzo, Italy, January 14 - 16, 2009.

Dissemination

This project has been widely highlighted in various national and international presentations made at various meetings of the Group of Telecommunications, the Network group and the ISI technology platform.
Knowledge acquired in the project has been transferred in the cours "Trasmissione dati multimediali (Multimedia Trasmission)" at the University of Pavia in the form of course material, tutorials and examples of the capabilities of the proposed advanced techniques of video coding and delivery. The partecipation of a Ph.D. student is in itself a value added to the promotion of his skills.
The activities have been published in National and International conferences.

Final activity

Summary of achieved results

During the extension period the Pavia research unit has developed activities on the following WorkPackages:

WP1 - Hardware and Software platforms
Activity "1.B Application level: real-time collaborative e-learning environments " and, in particular:

  • regarding "1.2 B Vision-based perceptive interfaces for natural interaction with learning environments - Explicit communication", the following activities have been carried out: (i) further testing to improve the performance of WeyeB, the fully ocular interactive system for web browsing developed during the third year, in order to optimize some technical aspects. The work is part of a goal concerning the implementation of e-learning platforms available to all kinds of users, even those with motor disabilities [12]; (ii) it is in its final stage of development an eye-controlled 'smart cursor' for the interaction with ordinary operating environments such as MS Windows. Depending on the position on the screen, the cursor can directly select icons (by fixation) or can be moved precisely to a desired point, to perform virtually any operation that can be obtained using a mouse;
  • regarding "1.3B Vision-based perceptive interfaces for natural interaction with learning environments - Implicit communication": (i) an algorithm for the recognition of user states has been improved (the algorithm was developed during the third year of the project and is described in a paper presented at an international conference [13]). The prototype, which uses a simple webcam as an input tool, distinguishes among user states 'present', 'absent', 'in front of the monitor', 'back turned to the monitor', 'turned to the left', 'turned to the right', 'at the telephone'. Incorporated into an e-learning platform, a system for recognizing the user's status may improve the quality of the interaction, better adapting to the student behavior; (ii) it is in its final stage of development an algorithm for specific facial expressions recognition, to be used for the identification of the 'emotional state' of the user.

Activity "1.C Application level: adaptive QoS": during the final six months it was possible to finish the research on scalability techniques with extension to the case of 'medium grade scalability' (MGS) and the case of 3D Multiview transmission sequences. It was also possible to refine the simulation software for peer-to-peer multi-cast network and simulate an integrated environment of transmission able to exploit the features of reorganization of the network and of the video stream scalability. This integration has led to some scientific publications. Quality models of the generated sequences were also realized. In particular:

  • regarding "1.1 C Analysis of the current multimedia collaborative standards", the heterogeneity of networks and devices is driving the standardization bodies to introduce in the video coding standard scalability options that allow the coexistence of different systems and equipments. From this point of view, the latest version of the H.264 standard introduced the concept of Medium Grain Scalability that allows, together with techniques for bit allocation, to monitor the relationship between quality and bit rate in a finer way compared to the Coarse Grain Scalability introduced since the time of the MPEG2 standard, but simplified compared to the Fine Grain Scalability in MPEG4. A trend also evident in commercial products is the use of stereoscopic sequences to increase the feeling of interaction between user and sequence. The analysis of the literature has highlighted the lack of a true standard for 3D encoding and at the same time there are a number of approaches that allow to recreate the 3D experience from corecorded sequences from two cameras or from a single 2D sequence plus a depth map able to re-create any number of visions. None of these approaches is so far a winning result and the research in the field is still refining the two techniques;
  • regarding "1.2 C Scalability and multiple descriptions", the activity has concentrated on a detailed ILPS algorithm for H.264 video coding based on multiple scale descriptions developed during the project. In particular, optimizations of the allocation of bandwidth between the base and enhancement layer [17] have been carried out and an unbalanced allocation of bandwidth between the different descriptions [14] has been analyzed. An algorithm has been developed for management of multicast network structure with decentralized control that can take advantage of the video coding with multiple scalable descriptions to minimize the effects of changes in network bandwidth and instability due to abandonment (churn) of intermediate nodes that causes service interruptions. It has been shown that the algorithm is able to maintain continuity of service with a small and temporary degradation of quality [16]. This algorithm must be extended from the case of coarse grain scalability, where there is only one level of scalability, to the case of MGS recently introduced in the standard coding H.264. For MGS coding it has now been developed a model to predict the loss of quality in case of truncation of the information flow in precise points and thereby to determine a compromise between quantity of transmitted bits and quality [15] [18]. The research activity on sequence transmission for the three-dimensional view is documented by the publication [3] currently under final review.

The activity on "1.D Device level" has regarded the theoretical and experimental study of nonlinear optical devices suitable to carry out the wavelength conversion and the phase conjugation of high frequency figure optical signals. In particular:

  • regarding "1.1D Photo refractive effect", studies have been carried out on (i) the photorefractive properties of lithium niobate crystals, (ii) the conversion of the wavelength of signals with frequency between 10 Gb/s and 100Gb/s, modulated in amplitude and in phase, and with polarization multiplexing, (iii) the compensation of nonlinear effects in optical communication systems with phase modulation, (iiii) the conversion of wavelength and phase conjugation in communication systems with very low attenuation (near-lossless) for simultaneous compensation of dispersive and nonlinear effects produced during the propagation in the fiber. As regards (i), the results showed the presence of certain optical effects not previously reported in the literature that have been the subject of a presentation at an international conference and of an article which is currently under review[9]. With regard to objective (ii), the work has been done in collaboration with the Politecnico di Milano. In particular, it was shown for the first time that through a proper design and development of the wavelength conversion apparatus it is possible to change the optical carrier of a multiplexing polarization signal, and at very high bit-rate (100Gb/s) without penalty during the operation. This activity has led to the publication of an article [1] on an international journal and to contributions at major international conferences [7], [10]. Regarding activity (iii), the study, both theoretical and experimental, of techniques for the compensation of the Gordon-Mollenauer effect has been done. This effect (also known as 'nonlinear phase noise') is in fact crucial in diminishing the services received by the phase modulation communication systems. In this area it was carried out a systematic study on the comparison of different techniques of nonlinear noise reduction due to the Gordon-Mollenauer effect. The results, not yet published, were presented at an international conference [8] and showed how the technique MNTI (Mid-Nonlinearity Temporal Inversion), conceived and developed by the University of Pavia, is extremely promising in this area too. Activity (iiii) was carried out in collaboration with Aston University (Birmingham, UK). Thanks to a particular structure of Raman amplification, developed by Aston University, it has been possible to analyze the impact of wavelength conversion in systems practically free of losses. This has yielded interesting results both for dispersion compensation and nonlinear effects, and for the maximum length of transmission reachable. The results were presented at an international conference [11].
  • regarding "1.2D Micro-structured fibres", a study on the obtainable performance of bandwidth has been conducted. The activity has led to a publication in an international journal [2]. This publication shows experimentally and theoretically justifies a method for generating broadband coherent radiation through microstructured fibers. It is important to note that this method allows to obtain, unlike what has already been proposed in the past, a broadening of the optical domain only in a specific spectral region.

WP2 - Integration and assessment of solutions
Regarding the activity "2.B Application level: Assessment of software solutions", studies have been continued to evaluate different modes of information presentation in on-line newspapers, with special reference to multimedia content and ad banners (an article has been drafted for an international journal). The results of this research can be usefully exploited for the definition of guidelines for interface and content implementation in e-learning. It has also been increased the international collaboration with the Imperial College of London and the Mitsubishi Research Laboratory in Cambridge (MA, USA), in evaluating different modes of presentation of large amounts of images, according to RSVP (Rapid Serial Visual Presentation) techniques studied over the past two years.

Publications

  1. P. Martelli, P. Boffi, M. Ferrario, L. Marazzi, P. Parolari, R. Siano,V. Pusino, P. Minzioni, I. Cristiani, C. Langrock, M. M. Fejer, M. Martinelli, V. Degiorgio, "All-Optical Wavelength Conversion of a 100-Gb/s Polarization-Multiplexed Signal", Optics Express 17, September 2009.
  2. L. Tartara, V. Degiorgio, R. Cherif, M. Zghal, "Setting an upper wavelength limit to the supercontinuum generated in a photonic crystal fiber", J. Telecomm. Inform. Technol. 1, 34, 2009.
  3. L. Favalli, M. Folli, "A scalable multiple descriptions scheme for 3D video coding based on the inter layer prediction structure", Hindawi Journal on Digital Multimedia Broadcasting, Special Issue on 3DTVP, to appear.
  4. L. Bianchi, R. Gatti, L. Lombardi, L. Cinque, "Relative Distance Method for Lossless Image Compression on Parallel Architectures", in: Proceedings of Fourth International Conference on Computer Vision Theory and Applications, VISAPP 2009, Lisboa, Vol. 1, pp. 20-25, Feb 5-8, 2009.
  5. L. Bianchi, R. Gatti, L. Lombardi, L. Cinque, "Parallel Lossy Compression for HD Images - A New Fast Image Magnification Algorithm for Lossy HD Video Decompression Over Commodity GPU", in: Proceedings of Fourth International Conference on Computer Vision Theory and Applications, VISAPP 2009, Lisboa, Vol. 2, pp. 16-21, Jan 5-8, 2009.
  6. M. Piastra, "A growing self-organizing network for reconstructing curves and surfaces", IJCNN 2009, International Joint Conference on Neural Networks, Atlanta, GA, pages 2533-2540, June 14-19, 2009.
  7. L. Marazzi, P. Parolari, P. Martelli, R. Siano, P. Boffi, M. Ferrario, A. Righetti, M. Martinelli, V. Pusino, P. Minzioni, I. Cristiani, V. Degiorgio, C. Langrock, M.M. Fejer, "Real-Time 100-Gb/s POLMUX RZDQPSK Transmission over Uncompensated 500 km of SSMF by Optical Phase Conjugation", Paper JWA44, Optical Fiber Communication Conference, OFC/NFOEC, San Diego, California, USA, March 22-26, 2009.
  8. V. Degiorgio, "Nonlinear effects in optical communication systems: a compensation method based on optical phase-conjugation", invited talk at the Conference "Novel Optical Materials and Applications", Cetraro, Italy, June 7-12, 2009.
  9. J. Parravicini, P. Minzioni, V. Degiorgio, E. DelRe, "Photovoltaic self-focusing in lithium niobate", Paper EF4.4 Oral presentation at Conference on Lasers and Electro-Optics CLEO/Europe-EQEC 2009, Munich, Germany, June 14-19, 2009.
  10. P. Martelli, P. Boffi, M. Ferrario, L. Marazzi, P. Parolari, R. Siano, V. Pusino, P. Minzioni, I. Cristiani, C. Langrock, M. M. Fejer, M. Martinelli, V. Degiorgio, "All-optical wavelength conversion of a polarization multiplexed signal", Paper CI2.2 Oral presentation at Conference on Lasers and Electro-Optics, CLEO/Europe-EQEC, Munich (Germany), June 14-19, 2009.
  11. P. Minzioni, P. Harper, V. Pusino, L. Barker, C. Langrock, M. M. Fejer, J. Ania-Castanon, I. Cristiani, V. Degiorgio, "Optical Phase Conjugation for Dispersion and Nonlinearity Compensation in a 1600km, 42 Gb/s Quasi-Lossless System", OSA-Nonlinear Optics Conference, Hawaii, USA, July 12-17, 2009.
  12. V. Cantoni, M. Porta, A. Ravarelli, "Towards Fully-Accessible E-Learning Systems: Some Eye Tracking Solutions", Proceedings of the 5th E-Learning Conference, Berlin, Germany, August 31 - September 1, 2009.
  13. F. Fosso, M. Porta, "A Vision-Based Attentive User Interface with (Semi-) Automatic Parameter Calibration", Proceedings of the 10th International Conference on Computer Systems and Technologies, Rousse, Bulgaria, June 18-19, 2009.
  14. L. Favalli, M. Folli, "ILPS: a scalable multiple description coding scheme for H.264", Future Multimedia Networking: Second International Workshop, FMN 2009, Coimbra, Portugal, June 22-23, 2009, Proceedings. Also appears in Lecture Notes in Computer Science ... Networks and Telecommunications, E. Cerqueira, M. Curado eds., Springer, Berlin 2009.
  15. L. Favalli, M. Folli, M. Cesari : "Quality modeling for the Medium Grain Scalability option of H.264/SVC", 5th International Mobile Multimedia Communications Conference - Mobimedia 2009, London, United Kingdom, September 07-09, 2009.
  16. L. Favalli, M. Folli, M. Lanati, "Improved multicast algorithm for overlay multicast in P2P based video streaming", To appear, 2010 IEEE Consumer Communications and Networking Conference, Las Vegas, Jan. 9-12, 2010.
  17. L. Favalli, M. Folli (2008), "Scalable multiple description coding of video sequences", Streaming Day 2009, Genova, 21 settembre 2009.
  18. L. Favalli, M. Folli, M. Cesari, "A semianalytical model for the quality vs. rate tradeoff in Medium Grain Scalable H.264 coders", Streaming day 2009, Genova, 21 settembre 2009.

Dissemination

The participation in this project by the units involved, and in particular the cooperation with the Tunisian group, was given ample evidence in various fields, both nationally and internationally. In particular, the 8th of May 2009, was held in Hammamet, Tunisia, a final workshop of the project with the presentation of the results achieved, which was attended by the leaders of the scientific research unit of Pavia, of Pisa, by the coordinator of the Tunisian group and by researchers from the different units. The presentation day was open to the public and advertized during the "ICTA 2009 - International Conference on Information and Communication Technology & Accessibility". The project work has been highlighted in terms of new knowledge in seminars and events.

Site of the Scuola Superiore Sant'Anna di Pisa Site of the Universita' degli Studi di Pavia