NEuromorphic Reconfigurable Integrated
photonic Circuits as artificial image processor NEuromorphic Processor Based on qUantum-Dot LAsers Deep Learning in Optical Communication Systems

RESEARCH UNIT ON NEUROMORPHIC COMPUTING AND PHOTONICS (RNCP)

RNCP unit is a 2020 initiative for collaboration between the Computer and Communication Systems Laboratory (CCSL) of the University of the Aegean, department of Information and Communication Systems Engineering and the Parallel and Distributed Systems and Networks Lab (PDSN) of the University of West Attica, Department of Informatics and Computer Engineering. The two labs decided to unite forces and expertise forming a joint research unit specializing on photonic neuromorphic computing systems and their applications, dedicated machine-learning hardware and software techniques, optical communications and physical layer security. The group has thorough expertise gained through the participation of its members in diverse EU and national research projects in the broad area of photonics and optical communications.

REASEARCH ACTIVITIES

Photonic neuromorphic computing is a rapidly evolving scientific field which attracts the interest worldwide due to its intrinsic scientific value and the niche applications it can enable. Bio-inspired computing can be performed in photonics with the use of pulsed lasers and complex waveguide structures achieving neuron-like behavior in unprecedented speeds and low consumption. The intrinsic parallelism of photonic circuits paves the wave for enhanced connectivity and scalability so as to solve complex problems at the speed of photons. RNCP members have a strong activity in the area as can be found in the publication record and in the participation in EU and national R&D projects.

Optical communication systems suffer from linear and nonlinear effects of the optical channel. Although modern ASICs can handle linear effects such as polarization mode dispersion and chromatic dispersion, the major limitation regarding the maximum capacity that can be achieved comes from the nonlinearities attributed to Kerr effect. RCNP investigates state of the art recursive neural networks and reservoir computing techniques so as to mitigate the nonlinear effects in long-haul WDM transmission systems. The same activity seeks for applying low complexity machine learning algorithms in short-area networks where consumption matters.

One of the key areas that MCP has invested is the envision-design and development of cryptographic devices based on electronic-photonics hardware for cyber-physical applications. In particular MCP has developed multiple designs of optical modules as non-replicable authentication tokens and secure pseudo-random generators able to be integrated in IoT ecosystems and solidifying their resilience against cyber-physical attacks. More importantly MCP is working towards combining its solid background on neuromoprhic engineering and crypto-systems, so as to spawn a new generation of electronic-photonic neuro-cryptographic devices able to offer a twofold advantage. On one hand, provide solid security features, such as data encryption and authentication, based on physical properties and secondly employ the same modules for anomaly detection and edge machine learning

RNCP PROJECTS

NEBULA

NEuromorphic Processor Based on qUantum-Dot LAsers

Nebula Project is a research program funded by GSRT-ELIDEK (GR), hosted by the Dept. Informatics & Telecommunications of the National & Kapodistrian University of Athens and particularly the Optical Communications & Photonics Technology Laboratory, whereas it is supported by the Dept. of Photonic & Quantum Sciences of Heriot-Watt University. ​

It focus on the design, optimisation and development of fully isomorphic to biological structures photonic neurons with ultra-fast response  and marginal power consumption. These structures will be incorporated to a electro-optic scheme so as to mimic in the optical domain artificial sensorimotor processing and exploit disruptive computational paradigms like the Reservoir Computing Concept. ​

Its long-term vision is to provide an alternative path for computation harvesting the merits of photonics and neuromoprhic engineering and transforming them to a new paradigm

read more

NEoteRIC

NEuromorphic Reconfigurable Integrated
photonic Circuits as artificial image processor

An unconventional approach to demanding image applications

NEoteRIC’s primary objective is the generation of holistic photonic machine learning paradigms that will address demanding imaging applications in an unconventional approach providing paramount frame rate increase, classification performance enhancement and orders of magnitude lower power consumption compared to the state-of-the-art machine learning approaches.

NEoteRIC’s implementation stratagem incorporates multiple innovations spanning from the photonic “transistor” level and extending up to the system architectural level, thus paving new, unconventional routes to neuromorphic performance enhancement.

read more

SafeIT

“SafeIT – Wearable systems for the safety and wellbeing applied in security guards”, Research-Create-Innovate 2nd Cycle, National Strategic Reference Framework (NSRF) 2014-2020, 2020

Wearable devices and augmented reality are rapidly evolving and are particularly applicable in the service sector, such as safety, health, education, etc. In the field of security services, in particular the no-contact / lone security positions, device handler technology can offer innovative services. In particular, it can further secure the individual protection of staff in existing security services and improve their mental concentration and perceptual capacity, which assists in decision-making. Portable devices can detect the presence, biometric data of the guardian, and many additional elements that are useful both in the service they perform and the personal safety of the workers.

The goal of the project is to develop systems and applications that will receive and process data from smart wearable to record biometric data for the purposes of health and vital safety monitoring, positioning and presence / commencement / ending of the shift, direct notification and other required services, such as indications of danger, attack, etc. These devices will automatically wirelessly connect with additional equipment (ambient sensors and positional beacons) and their data will be combined with the cameras data. Systems of Augmented / Mixed reality will also be developed in addition to simply record and track signals, more effectively manage the data from incident management center operators and supporting decision-making by the end user (guardian).

Due to the variety of data to be transmitted and stored, a combination of modern signal processing and machine learning techniques is needed through the creation of a knowledge base. The latter will be shaped by the data collected per event. There will also be use of pioneering security techniques which combine multi-parametric authentication, such as the use of biometric elements (something the user is) with equipment components (somethingthe user has), which will be physically resistant to cloning, that is to say they will exhibit structural features that will make them unique. Some of the most important issues that will be addressed in the project are the data transmission security and the use of blockchain technology for secure data storage (protection against unauthorized data alteration, PUF assisted encryption), as well as the use of security technologies to protect personal data (use of nicknames and aliases, PUF reciprocal authentication). In the strand of augmented reality, the technological innovation that the project introduces is multidimensional. More specifically, this project is the first national effort to develop an integrated augmented/mixed reality platform that acts on both the field and the control center. The development of AR interfaces for managing information in the field of security services is an innovation of the project. Also, the development of ergonomic interfaces based on the geometric representation of the area of interest combined with the use of maps in an augmented reality environment is another innovation the project aspires to introduce. Finally, all applications that will be deployed will respect the privacy of users by complying to the requirements of innovative and academically recognized methodologies, while ensuring all the technical and organizational requirements set by the new General Data Protection Regulation 2016/679 (GCC-GDPR) are met.

read more

RNCP PEOPLE

RNCP Unit Directors

<b>Adonis Bogris</b>
Adonis BogrisProfessor
Adonis Bogris was born in Athens. He received the B.S. degree in informatics, the M.Sc. degree in telecommunications, and the Ph.D. degree from the National and Kapodistrian University of Athens, Athens, in 1997, 1999, and 2005, respectively. His doctoral
thesis was on all-optical
processing by means of fiber-based devices. He is currently a Professor at the Department of
Informatics and Computer Engineering at the University of
West Attica, Greece. He has authored or co-authored more than 150 articles published in international scientific journals and conference proceedings and he has participated in plethora of
EU and national research projects. His current research interests
include high-speed all-optical transmission systems and
networks, nonlinear effects in optical fibers, all-optical signal
processing, neuromorphic photonics, mid-infrared photonics
and cryptography at the physical layer. Dr. Bogris serves as a
reviewer for the journals of the IEEE and OSA.
<b>Charis Mesaritakis</b>
Charis MesaritakisAssociate Professor
Prof. Charis Mesaritakis acquired his diplom, M.Sc and Ph.D from National and Kapodistrian Univeristy of Athens (Greece). His Ph.D thesis focused on the experimental characterization and numerical modelling of novel regimes of quantum dot mode locked lasers. He has participated as a researcher in 10 FP6-FP7 and Horizon2020 EU projects. He has been awarded a postdoctoral EU Marie-Curie Fellowship, involving high precision laser telemetry in III-V Labs (France); Followed by two competitive national research grants, PROMITHEAS from the G. Latsis foundation and HFRI-GSRT NEBULA, both focusing on the investigation of photonic neuromorphic technologies and photonic machine learning. Currently he serves as technical manager for the H2020 NEoteRIC project also focusing on photonic neuromorphic paradigms. Since 2019, he is an associate professor at the department of Information and Communication Systems Engineering at the University of the Aegean, splitting his research interest among design/implement photonic neuromorphic systems for high bandwidth applications and photonic physical layer cryptographic modules (Η2020 KONFIDO, GSRT- SAFE-IT). He is author and co-author of more than 70 publications in highly cited journals and international conferences focusing on quantum-dot laser dynamics, neuromorphic schemes and physical layer security. He is a patent holder for photonic-physical unclonable functions modules for implementing physical layer security. He serves as regular reviewer for IEEE, OSA, AIP and Springer Journals whereas he serves as guest editor for MDPI’s Applied Sciences journal.


Mesaritakis CV

RNCP Researchers – PhD Students

<b>Stavros Deligiannidis</b>
Stavros Deligiannidis
Stavros Deligiannidis holds a BSc in Physics, a MSc degree in Microelectronics and VLSI from the National and Kapodistrian University of Athens. Since 2010 he is with the Department of Computer Engineering of the Technological Educational Institute of Peloponesse, Greece where he serves as a Lecturer. He is currently pursuing his PhD
degree at the University of West Attica in the field of novel signal processing techniques for optical communication systems under the supervision of Prof. Adonis Bogris. He has worked as a researcher in local and European projects. His current research interests include optical communications, deep learning, digital signal processing, and parallel computing.
<b>Dimitris Dermanis</b>
Dimitris Dermanis
Dimitris
Dermanis was born in Athens, Greece. He holds a 5 years diploma (integrated MSc) in computer science and telecommunications from the department of Information and Communication Systems Engineering (ICSD) of the University of the Aegean. He is currently pursuing
his Ph.D at the University of the Aegean, in the field of Neuromorphic Computing under the supervision of Prof. Charis Mesaritakis. His main research interests include neural networks, deep learning and cryptography.
<b>George Sarantoglou</b>
George Sarantoglou
George Sarantoglou received the Diploma degree in electrical and computer
engineering from the University of Patras, Patras, Greece, on 2016. He is
currently working towards his Ph.D. degree with the Department of Information and Communication Systems Engineering, University of the Aegean, Samos Greece. His Ph.D. thesis focuses on the experimental analysis and development of photonic processors for unconventional, bio-inspired information processing, targeting machine learning applications . His research interests include photonic systems for analog pattern recognition and multi-sensory applications.
<b>Menelaos Skontranis</b>
Menelaos Skontranis
Menelaos Skontranis received his bachelor in 2016 from Hellenic Air Force Academy as an Telecommunication-Electronic Engineer. He received his master in Microelectronics in 2019 from the Department of Informatics and Telecommunications of the National and Kapodistrian Univeristy of Athens. He currently is a PhD candidate in the Department of Information and Communication Systems Engineering of the University of Aegean under the supervision of Associate Professor Charis Mesaritakis. His research interest focuses on Quantum Dot Lasers, Vertical Cavity Surface Emitting Lasers and Optical Neural Networks.
<b>Kostas Sozos</b>
Kostas Sozos
Kostas Sozos received his B.S degree in Physics from the University of Patras in 2018 and the M.Sc in Microsystems & Nanodevices from the National Technical University of Athens in 2020. He is currently pursuing his Ph.D at the University of West Attica in the field of Neuromorphic Photonic Computing under the supervision of Prof. Adonis Bogris. His main research interests include photonic neural networks, deep learning, pattern recognition and optical communications.
<b>Giannis Tsilikas</b>
Giannis Tsilikas
Ioannis (Giannis) Tsilikas holds a 5 years Diploma (integrated MSc) in applied Physics, from School of Applied Mathematical and Physical Sciences, National Technical University of Athens. He is currently pursuing his PhD degree at the National Technical University of Athens in the field of ultra short laser pulses interaction with matter under the supervision of Prof. Charis Measaritakis and Prof. Georgios Tsigaridas. He has worked as a researcher in 5 National research projects and in 3 European research projects (up to May 2020). His current research interests include optical experimental set ups of Femto Laser Systems, photonics signal processing, short and ultra short duration laser pulses interaction with biological matter, biophotonics and photonics.
<b>Aris Tsirigotis</b>
Aris Tsirigotis
Aris Tsirigotis received his B.S. degree in Physics in 2016 and the Μ.Sc. in Electronics and Radioelectrology in 2020 from the National and Kapodistrian University of Athens.He is currently pursuing his Ph.D at the University of the Aegean in the field of Neuromorphic Photonic Computing under the supervision of Prof.Charis Mesaritakis. His main research interests include photonic neural networks, photonic design, deep learning, pattern recognition, optical communications and optical design.

RNCP PUBLICATIONS

Publications in international scientific journals

  1. S. Deligiannidis, C. Mesaritakis, A. Bogris, “Performance and Complexity Analysis of i-directional Recurrent Neural Network Models vs. Volterra Nonlinear Equalizers in Digital Coherent Systems”, ΙΕΕΕ Journal of Lightwave Technology , Early Access (2021) (Impact Factor: 4.142)
  2. M. Skontranis, G. Sarantoglou, S. Deligiannidis, A. Bogris, C. Mesaritakis, “Time-Multiplexed Spiking Convolutional Neural Network based on VCSELs for Unsupervised Image Classification”, Applied Sciences, accepted for publication (2021) (Impact Factor: 2,474)
  3. G. Sarantoglou, M. Skontranis, A. Bogris, C. Mesaritakis, “Experimental study of Neuromorphic Node based on a Multi- Waveband Emitting two – section Quantum Dot Laser” OSA Photonic Research, doi: 10.1364/PRJ.413371 (2021) (Impact Factor: 6.099)
  4. M. Skontranis, G. Sarantoglou, S. Deligiannidis, A. Bogris, C. Mesaritakis, “Unsupervised Image Classification Through Time-Multiplexed Photonic Multi-Layer Spiking Convolutional Neural Network” MDPI Applied Sciences, special issue on Optical Computing, accepted for publication (2021) (Impact Factor: 2.471)
  5. Bogris, C. Mesaritakis, Stavros Deligiannidis, Pu Li “All Optical Integrate and Fire Neuromorphic Node based on Single Section Quantum Dot Laser” IEEE Selected Topics in Quantum Electronics, Volume: 27, Issue: 2, (2020) (Impact Factor: 4.681)
  6. Mesaritakis, P. Rizomiliotis, M. Akriotou, C. Chaintoutis, A. Fragkos, D. Syvridis “Photonic Pseudo-Random Number Generator for Internet-of-Things Authentication using a Waveguide based Physical Unclonable Function” Arxiv.org (2020)
  7. Deligiannidis, A. Bogris, C. Mesaritakis, Y. Kopsinis, “Compensation of Fiber Nonlinearities in Digital Coherent Systems Leveraging Long Short-Term Memory Neural Networks” ΙΕΕΕ Journal of Lightwave Technology Volume: 38, Issue: 21, pp. 5991-5999 (2020) (Impact Factor: 4.288)
  8. Sarantoglou, M. Skontranis, C. Mesaritakis, “All Optical Integrate and Fire Neuromorphic Node based on Single Section Quantum Dot Laser” IEEE Selected Topics in Quantum Electronics, accepted for publication (2019) (Impact Factor: 4.681)

Conference Proceedings

  1. Y. Hong, S. Deligiannidis, N. Taengnoi, K. Bottrill, N. Thipparapu, Y. Wang, J. Sahu, D. Richardson, C. Mesaritakis, A. Bogris, P. Petropoulos, “Performance-enhanced Amplified O-band WDM Transmission using Machine Learning based Equalization”, CLEO – San Jose, California (2021)
  2. G. Sarantoglou, M. Skontranis,A. Bogris, C. Mesaritakis, “Resonate and Fire Neuromorphic Node based on two – section Quantum Dot Laser with multi-waveband dynamics”, ECOC-CLEO– Brussels, December (2020)
  3. M. Skontranis, G. Sarantoglou, S. Deligiannidis, A. Bogris, C. Mesaritakis,”Unsupervised Image Classification Through Time-Multiplexed Photonic Multi-Layer Spiking Convolutional Neural Network”, ECOC-CLEO– Brussels, December (2020)
  4. S. Deligiannidis, C. Mesaritakis, A. Bogris, “Performance and Complexity Evaluation of Recurrent Neural Network Models for Fibre Nonlinear Equalization in Digital Coherent Systems”, ECOC – Brussels, December (2020)
  5. Mesaritakis, M. Skontranis, G. Sarantoglou, A. Bogris, “Micro-Ring-Resonator Based Passive Photonic Spike-Time- Dependent-Plasticity Scheme for Unsupervised Learning in Optical Neural Networks” OFC USA – San Diego, March (2020)
  6. Sarantoglou, M. Skontranis, A. Bogris, C. Mesaritakis, “Temporal Resolution Enhancement in Quantum-Dot Laser Neurons due to Ground State Quenching Effects” OFC USA – San Diego, March (2020)
  7. A. Bogris “Multi-mode lasers as potential machines for reservoir computing of enhanced power” ERC International Workshop – Invited – Photonic Reservoir Computing and Information Processing in Complex Networks, Trento-Italy (2019)
  8. Mesaritakis “Passive Photonic Components as Building Blocks for Ultra-Fast Reservoir Computing and as Photonic Spike Dependent Plasticity Enabling Structures” ERC International Workshop – Invited – Photonic Reservoir Computing and Information Processing in Complex Networks, Trento-Italy (2019)
  9. Mesaritakis, “Photonic Reservoir Computing based on the Random-Interaction of Transverse Optical Modes in Large-Cross Section Waveguides” CLEO/EQEC Europe, Munich-Germany (2019)
  10. Μ. Skontranis, G. Sarantoglou, C. Mesaritakis, “Inhibitory Integrate and Fire Neuron based on Quantum-Dot Intra-Band Transitions in a Semiconductor Laser” CLEO/EQEC Europe, Munich-Germany (2019)
  11. Μ. Skontranis, G. Sarantoglou, C. Mesaritakis, “All-optical Inhibitory Integrate and Fire Neuron based on a Single-Section Quantum-Dot Semiconductor Laser” CLEO USA, San-Diego California USA (2019)

CONTACT RNCP

Adonis Bogris

abogris@uniwa.gr

Charis Mesaritakis

cmesar@aegean.gr

    Follow RNCP On Social Media

    RNCP

    Ag. Spyridonos, 122 43 Egaleo,
    Attica, Greece

    email: Adonis Bogris: abogris@uniwa.gr || Charis Mesaritakis: cmesar@aegean.gr

    web: rncp.eu