The large bandwidth, low propagation loss and immunity to eletromagnetic disturbances are major advantages of microwave photonics, whereby optical communication technologies are used in analog signal processing. The local oscillator is the cornerstone of these architectures; here, it consist of an optical carrier with a high purity modulation, a sort of optical clock which can be distributed reliably and easily all over the microwave system. Such a clock can be implemented in an opto-electronic loop involving a carrier which is modulated by a positive electronic feedback. These oscillators challenge in purity the established technology of quartz clock when they are compared in the microwave domain. The performance is ascribed to the long time delay (µs) in the feedback, owing to an optical fiber. CRONOS project is about a radically new approach allowing the miniaturization of the delay line. As the opto-electronic oscillator is based on a km long fiber, integration is impossible. Here, we will develop an opto-electro-mechanical loop with a millimeter-sized acoustic delay line, which indeed provides the required delay owing to the much lower speed of sound as compared to light. This requires a new technology where the interaction between light and sound results into an effective transduction. The heart of the system is an opto-mechanical crystal, allowing both mechanical and optical resonant modes which are spatially overlapped. Sub-micron patterning is here the key enabling the strong confinement (within a few µm) resulting into an effective light-sound interaction. Thus, a very sensitive detection of the displacement is possible, as the quantum limit for mechanical displacement results in 0.1 MHz to 1 MHz frequency shift of the resonance. The signal generated by the movement of the resonator is transduced in the optical domain, detected, amplified and finally re-injected as an acoustic wave propagating with the desired delay through an acoustic circuit before reaching the resonator. To this purpose, CRONOS will use a hybrid technology where a silicon photonic circuit will be associated to a III-V compound semiconductor alloy ensuring the best opto-mechanical performances and enabling the excitation of acoustic wave through the piezo-electric effect. The propagation of the acoustic wave is controlled through phononic band engineering owing to the periodic patterning of the material. Our first free-running optomechanical crystal have already demonstrated a fairly low phase noise (corresponding to a short-term linewidth about 100 Hz). CRONOS will improve this figure drastically, primarily by exploiting an electro-optomechanical loop which will reduce both the intrinsic phase noise and a servo loop to defeat instabilities and environmental perturbations. These challenging goals will be reached owing to the partnership between an academic institution, C2N (Centre de Nanosciences et de Nanotechnologies -UMR 9001, Palaiseau) and an industrial laboratory, TRT (Thales Research & Technology, Palaiseau) with a long track record of fruitful and ongoing collaboration. The consortium benefits from cutting-edge facilities for nanofabrication and an established expertise in semiconductor processing, nanophotonics, acoustic and microwave photonics, opto-electronic oscillators in particular.

The eProcessor ecosystem combines open source software (SW)/hardware (HW) to deliver the first completely open source European full stack ecosystem based on a new RISC-V CPU coupled to multiple diverse accelerators that target traditional HPC and extend into mixed precision workloads for High Performance Data Analytics (HPDA), AI, and Bioinformatics. eProcessor will be extendable (open source), energy-efficient (low power), extreme-scale (high performance), suitable for uses in HPC and embedded applications, and extensible (easy to add on-chip and/or off-chip components). eProcessor combines cutting edge research utilizing SW/HW co-design to achieve sustained processor and system performance for (sparse and mixed-precision) HPC and HPDA workloads by combining a high performance low power (architecture and circuit techniques) out-of-order processor core with novel, adaptive on-chip memory structures and management, as well as fault tolerance features. These software-hardware co-design solutions span the full stack from applications to runtimes, tools, OS, and the CPU and accelerators.

Highly coherent – that is narrow linewidth - laser sources are at the heart of numerous applications: ranging from terabit per second coherent communication, LiDAR for autonomous driving or driver assistance, to fibre sensing or optical atomic clocks. Yet, neither the principles of narrow linewidth lasers nor how they are manufactured, have changed in the past 30 years; fiber lasers, the workhorse for narrow linewidth, rely on bulk fibre based optical components assembled manually. In recent years, triggered by new applications in particular in FMCW LiDAR, the demand for integrated lasers that combine the coherence of a narrow linewidth fibre laser with high frequency agility and fast tuning, and can be manufactured wafer-scale in large volume at low cost has become a technological bottleneck. Here we overcome these challenges and will demonstrate for the first time a mass manufacturable, compact, wafer-scale narrow linewidth laser with unprecedentedly agile tuning and precise laser tuning. To achieve this novel technology, we will employ recent findings on a hybrid electro-opto—mechanical integrated laser obtained in the FET Proactive project “Hybrid Optomechanical Technologies.” To combine the conflicting properties of ultrahigh coherence and fast and precise tuning, we will combine sub-micron piezo-electrical actuators that rely on AlN – a proven MEMS technology - that combine an electrical and mechanical engineered degrees of freedom with silicon nitride ultra-low loss integrated photonic circuits. The combined hybrid opto-electro-mechanical device exhibits unique performance characteristics in terms of linewidth and frequency agility not attained anywhere to date, making them ideal sources for LiDAR. FMCW LiDAR is a next generation perception technology enabled by narrow linewidth highly tunable lasers that has major advantages compared to the currently used time of flight LiDAR: it is eye-safe and crucially can give both velocity and distance information in the same pixel – massively simplifying the object classification. FMCW can even operate in glaring sunlight, is immune to crosstalk from other sensors, and is ideally suited for long-range detection as required for autonomous driver assistance. It is, however, compounded by the very stringent requirements: it requires exceptionally narrow linewidth, highly linear tuning of the laser, and above all, this technology, being applied to mass markets, is required to be manufactured in large volumes at low cost. Ultra-stable, fast and linearly tunable lasers are also key enabling technology for reconfigurable, ultra-broadband wireless transmitters for future wireless communications. Using the unoccupied mmWave band above 100 GHz requires low-cost yet very high performance tunable laser systems and high-bandwidth photoreceivers to cope with the requirements of dense mesh area coverage imposed by the significant atmospheric absorption at these frequencies. In this project, we will manufacture such hybrid integrated lasers using volume-manufacturing compatible techniques only – using both monolithic and transfer printing techniques - improve their tuning and linewidth further, and test them in a relevant application scenario for FMCW LiDAR and validate the combination of NOEMS technologies and integrated photonics devices for use in optical communication by demonstration of tight phase locking using the novel actuator technology.

Compact and efficient antenna architectures are key enabler solutions for on-board electromagnetic applications (telecommunications, radars, electronic warfare, etc.). Current solutions use cumbersome mechanical systems, or expensive but flat electronically steered phased arrays. The main goal of the project is to develop a disruptive, modular and ultra-low-profile antenna architecture for the next generation of high data rate satellite communication systems for moving platforms. The proposed system will be developed in Ka-band to benefit from the high data rates offered by future satellite constellations. High data rates require the large bandwidth available at Ka-band but also extremely flat and steerable antennas to be integrated on the fuselage of moving platforms (airplanes, drones, missiles, trains, etc.). The main technical and scientific challenge of the project is the development of an extremely flat antenna architecture able to steer in a fast and efficient way its main beam over a large angular sector (> ± 65°) and wide band (20% relative bandwidth centered at 20 GHz for the downlink). The KAPLA project will develop and integrate two very innovative building blocks at the core of the proposed novel architecture. The proposed architecture will be validated experimentally to demonstrate scanning performances beyond the state-of-the-art. In details, these building blocks are: - An extremely low profile beam-forming network (BFN) using leaky-wave modes, - A thin deflector based on artificial materials and made of novel hybrid metal-dielectric cells at sub-wavelength scales. These elements will be fabricated using: - Low-cost PCB technology for the leaky-wave BFN using guided structures (substrate integrated waveguide (SIW), or parallel plate waveguide), - Novel additive manufacturing processes for metallic/polymer elements for the deflector, as this element cannot be manufactured with standard fabrication processes. This hybrid and pragmatic approach and the proposed modular architecture will guarantee a dramatic price reduction with an enhanced modularity of the system fulfilling stringent requirements for civil and military needs. The civil and military stakes of the project are clear. Satcom communications are essential for civil (e.g. high data links for flight entertainment) and military applications (e.g. drone communication). The KAPLA project will address both needs with a single antenna concept operating in the military and civil bands (transmitting or receiving band) to reduce drastically the development and industrialization time and thus the final cost of the product. The economic stakes are also impressive. The civil and military market for Satcom applications is growing at a fast and constant pace. The KAPLA project will guarantee the excellence of the civil and defense French industry in such competitive field facing a strong American competition.