Eulambia's team is constantly trying to push the boundaries of existing technology. In order to offer evolutionary products and services we never stop to research new ideas and try to bring them to life.
5G NR baseband Unit (BBU)
5G NR BBU targeting high bandwidth and performance beyond 5G specifications
- Implemented in a FPGA device for real-time processing of the radio signal
- Supports 3GPP 5G NR (v15.1) numerologies for Frequency Range 2 (FR2)
- Extends 5G NR numerologies achieving a record bitrate of 5.67Gb/s over an IF/RF bandwidth of 760MHz
- Offers large flexibility in parameterization (through software) 
- 5G NR numerology
- Bandwidth usage (up to 760MHz with sub-MHz resolution)
- Modulation scheme (QPSK, 16-QAM, 64-QAM, 256-QAM)
See more details about EULAMBIA's 5G NR BBU in the video below or download the detailed presenation from the following link: EULAMBIA_5G_BBU
 R Munoz, R Vilalta, C Manso, L Rodríguez, J M Fàbrega, R Martínez, R Casellas, J Brenes, G Landi, M Capitani, G Otero, D Larrabeiti, C Vazquez, J D Lopez-Cardona, E Grivas, T Lagkas, D Klonidis, S Rommel, I Tafur Monroy, Experimental demonstration of advanced service management in SDN/NFV fronthaul networks deploying ARoF and PoF, Demo Paper ECOC 2019
Photonic Physical Unclonable Functions
In photonic PUF (p-PUF) implementations, various objects can be used as PUF tokens: simple paper, due to the uniqueness of the paper-pulp; sandblasted glass, exploiting the randomness of defects; multimode waveguides (EULAMBIA’s patented technology).
These tokens are illuminated with a laser, under varying conditions (INPUT), and different OUTPUTS are produced (speckle patterns - very complex images of bright and dark spots). These images are cryptographically processed, and unique keys can be produced.
These outputs/keys can be used for:
- User/device/process authentication
- As high-quality cryptographic keys
- As seeds for pseudo-RNGs
- PUFs can be used as Random Number Generators (non-reproducible)
- PUFs can be used as a hardware key for physical access control
- Keys are not permanently stored in secure Non-Volatile Memory
- Keys are cryptographically compatible; they can be used for symmetric encryption without post-processing
- p-PUFs can be used as non-reproducible Random Number Generators providing high-quality seeds for pseudo-RNGs
- Immune to cloning/spoofing attacks
- Anti-tampering capabilities due to sensitivity to even minor variations (stress, tension, heating)
- Advantages compared to electronic implementations:
- Increased resiliency to machine-learning attacks
- Modelling attacks are much harder
- Reduced vulnerability to side-channel attacks
- Better entropy source, due to more complex underlying physical mechanism; exponentially large pool of numbers providing keys at Mbps rates
Learn more about Eulambia's photonic PUFs reading the white paper: p-PUF White_Paper
Secure 5G communications
Securing the physical layer of current and future communication systems
Nowadays communication networks support all crucial human activities from personal communications to financial transactions, massive data management, industrial processes, energy infrastructures, health/medical data exchange, transport, etc. Military communications are another highly demanding area in terms of security. Although a lot of attention has been given to the performance of the networks, targeting higher bandwidth and lower latency, security aspects become more and more crucial for obvious reasons.
Since the communication systems and networks, especially those based on fiber-optic media, were not designed from the ground up taking security aspects into account, the current solutions are mainly applied at the upper network layers rather than employing a holistic new approach. Particularly in the physical layer of the networks insufficient progress in security has been made. Facing the potential threats at the lower network levels, will significantly impact the security aspects at the higher layers as well.
In almost every current and future communication paradigm the transceiving elements use a number of various modulation parameters in order to achieve the desired operation. These same parameters can be used in order to enhance the communication security exploiting directly the physical layer of communications.
The OFDM-based optical communication case
Subcarrier scrambler module, to cyber-harden OFDM-based communications
Orthogonal Frequency Division Multiplexing (OFDM) has been an outstanding choice for 4G networks providing significant spectrum efficiency and performance improvement in frequency selective channels. The Cyclic-Prefix (CP) OFDM is the predominant candidate for 5G networks for the cases of downlink and uplink, in the sub-6GHz frequency band and for the mmWave range.
In the case of the OFDM transmitter, a high bit rate stream after the parallel to serial converter is driven to QAM mapper and the mapping process forms the buffered bit stream to QAM symbols. In a conventional OFDM system, the complex stream is given as input to the IFFT stage, modulating each subcarrier with QAM symbols. In this scheme, an extra stage – scrambler - is added performing re-distribution of the subcarriers across the frequency domain. The scrambler module performs the subcarrier scrambling operation using a random number sequence. This random number sequence is produced by a pseudo Random Number Generator (pseudo-RNG). The seeds for the RNG can for example be produced by exploiting the unique responses of a photonic Physical Unclonable Function (p-PUF) device. Thus, a scrabbling number sequence is produced, and the subcarriers are scrabbled accordingly. A Cyclic Prefix (CP) in order to combat the multipath is added and afterwards the produced complex OFDM signal is RF up-converted, amplified by power amplifier (PA) and radiated from the antenna.
At the receiver side, the reverse operations are followed including synchronization, frequency domain estimation/equalization and de-scrambling. Given that the same random number sequence is used (in case of a p-PUF the seed reproduced on the fly and fed into the RNG), the de-scrambling sequence would precisely follow the scrabbling one.
Typical 5G communication scenario with hardened physical layer – the case of a PUF-based scrambler
In a typical 5G communication scenario where a central office receives data from the backhaul and sends data to a Remote Radio Head (RRH), the RRH communicates with several terminals (T1, T2, etc.). Equipping the RRH with a scrambling module as described before we can cyber-harden the whole communication process, both between the Central Office and the RRH, as well as between the RRH and the terminals.
- The RRH is equipped with the scrambling module
- This scrambling module (e.g. PUF-based) will act as a “server” for the terminals and as a “client” for the Central Office.
- The terminals are equipped with a secure terminal chip that securely stores one random number sequence (PUF response in the form of a bit-string). At the terminal it’s unique ID and some helper data are also stored. These are public so no secure storage is needed.
- When the terminal wants to communicate with the RRH it sends its ID and the helper data. At the PUF-equipped RRH the key of the terminal is recreated (using the terminal ID as a challenge and the helper data). Both the terminal and the RRH use the key as a seed for the same pseudo-RNG so they acquire the same number sequence with is used for the scrabbling / de-scrabbling of the subcarriers.
- No mass leak can occur; each terminal has its own key and while the RRH is not a secure entity it is equipped with the PUF. So, the keys are generated on-the-fly, when they are needed. No-keys-at-rest property.
Central Office – RRHs communication
- As before, each RRH is not a secure entity but it is equipped with the PUF.
- The Central Office is considered a secure entity, so, a mapping between challenges and responses can be stored there for each one of the RRHs
- A unique ID (challenge) is used for each RRH and the response (key) is used for symmetrical encryption in the higher layers
UV-C diffused light communications
Free space optical communications that are based on Line-Of-Sight (LOS) in the visible and Near-Infrared (NIR), suffer from fast and deep fadings due to atmospheric turbulence, signal-to-noise degradation due to solar background radiation noise and they require accurate transceiver alignment.
By using transceivers in the UV-C solar blind wavelength range (200-280nm), we exploit the increased scattering characteristics of the atmosphere and the low solar background noise at the lower end of the visible spectrum.
- No strict transceiver alignment is needed
- Transceiver communication at high elevation angles
- Uninterrupted communication channel even if physical obstacles are present between the transceivers
- The spectrum is in principle unused
- Very sensitive photo-detectors with wide field of view are available
- Point-to-Point, Point-to-Multipoint, Multipoint-to-Multipoint NLOS connections
- Covertness for secure operation
- Resistance against jamming
- Resilient to electromagnetic interference
- Secure, free-space, short-area optical communication networks in harsh and electromagnetically polluted environments.
- Backup or main communication channel in distributed unattended sensor networks.
- Secure communication channel for UAVs and drones
- Underwater communications
- Atmospheric condition evaluation (visibility, luminance, precipitation, etc.)
- Local weather condition prediction
Ice detection sensor
- A lightweight, small form-factor, wireless/wired ice detection sensor based on optical back-scattering, that can detect the presence of ice, identify its type (rime, glaze, mixed) and estimate its thickness.
- The sensor can be mounted in a large number of moving or stationary surfaces offering real-time monitoring of the icing conditions at the monitoring surface and generates customized alarm notifications.
- Due to its wireless connectivity, the sensor can be attached in different critical points of the monitoring surface or infrastructure creating a distributed ice detection network.
- Ice detection (>=10μm)
- Ice-Water Discrimination
- Ice type classification (Glaze, Rime, Mixed)
- Ice thickness estimation (<=1cm)
- Surface temperature monitoring
- Sensor enclosure temperature monitoring
- Ambient light level monitoring
- Wired or wireless (BLE, ZigBee, etc.) communication with the base station
- Full software platform for calibration and monitoring of the sensors and real time notification centre
- Mounted on wings, fuselage or other critical points on the aircraft providing information about the aerodynamics
- Deployed across the runway of airports in order to detect the asphalt icing conditions
- Mounted across the blades for direct and accurate measurement of accreted ice (power optimisation, infrastructure damage avoidance, personnel safety)
- Mounted at the fuselage of the wind turbine
- Installed sensors under the vehicle chassis for black ice detection
- Deployed sensors at the asphalt for direct icing condition detection
Reservoir Computing techniques
- An RC is a three-layer neural network, with a typical input layer and an output layer, comparable to a feedforward linear perceptron
- The computational core resides in the hidden layer that is an assembly of non-linear nodes exhibiting random inter-node connections that remain untrained during the training phase
- The existence of connection loops in the hidden layer offers system memory and enables the processing of time dependent signal
The RC’s non-linearity and random connectivity offers an increase in the dimension-space of incoming signals, thus allowing their classification by a simple perceptron, which is subject to back-propagation training techniques.
State of the art implementations
- Dedicated photonic integrated circuits (PICs)
- Nano-photonic elements as computational cores
Typical applications include
- Prediction of medical conditions from heterogeneous data streams:
- Early diagnosis of neurodegenerative diseases (Parkinson, Alzheimer)
- Early prediction of patient’s health status in intensive care units
- Real-time video classification
- Handwritten pattern identification
- High-speed network traffic prediction for security/routing application
- Deep-packet sniffing directly in the physical layer
- Risk analysis and anomaly prediction/detection in multi-sensory scenarios
- Autonomous decision making / analysis / response for automotive applications
- Event-driven control in soft-robotics