YAHYA-6G aims to propose new signal processing solutions doped with machine-learning. We will focus on the detection and compensation of RF imperfections in mMIMO (massive Multiple input Multiple output) based NOMA (Non-orthogonal multiple access) pair . In other hand, YAHYA-6G target is to minimize the long-term power consumption based on the stochastic optimization theory for mMIMO-NOMA IoT networks with EH (Energy Harvesting) in presence of RF imperfections. Thus the objectives of the YAHYA-6G project are: 1- Identify major RF imperfections that may occur in a multi-access / multi-antenna broadband system. 2- Propose new solutions to optimize the energy efficiency at the RF transmitters. This solution will focus on the power amplifier that represents 60 at 70% of the energy consumed in an RF transmitter. 3- Analyze the impact of these RF imperfections on mobile radio systems exploiting NOMA technologies. 4- Propose a Deep Learning online learning process to detect the NOMA channel characteristics and compensate the effect of HPA nonlinearity. A joint detection of the NOMA interference and HPA (High Power Amplifier) nonlinearity will be studied in mMIMO-NOMA system. 5- Resolve a non convex based problem coping with the expected 6G requirements, with a particular focus on optimal resource scheduling and computation capacity allocation and reducing energy consumption of wireless devices, through a set of new algorithms . 6- Realize a demonstrator based on the SDR (Software Defined Radio) USRP cards on which some algorithms developed in the project will be implemented.

YAHYA-6G aims to propose new signal processing solutions doped with machine-learning. We will focus on the detection and compensation of RF imperfections in mMIMO (massive Multiple input Multiple output) based NOMA (Non-orthogonal multiple access) pair . In other hand, YAHYA-6G target is to minimize the long-term power consumption based on the stochastic optimization theory for mMIMO-NOMA IoT networks with EH (Energy Harvesting) in presence of RF imperfections. Thus the objectives of the YAHYA-6G project are: 1- Identify major RF imperfections that may occur in a multi-access / multi-antenna broadband system. 2- Propose new solutions to optimize the energy efficiency at the RF transmitters. This solution will focus on the power amplifier that represents 60 at 70% of the energy consumed in an RF transmitter. 3- Analyze the impact of these RF imperfections on mobile radio systems exploiting NOMA technologies. 4- Propose a Deep Learning online learning process to detect the NOMA channel characteristics and compensate the effect of HPA nonlinearity. A joint detection of the NOMA interference and HPA (High Power Amplifier) nonlinearity will be studied in mMIMO-NOMA system. 5- Resolve a non convex based problem coping with the expected 6G requirements, with a particular focus on optimal resource scheduling and computation capacity allocation and reducing energy consumption of wireless devices, through a set of new algorithms . 6- Realize a demonstrator based on the SDR (Software Defined Radio) USRP cards on which some algorithms developed in the project will be implemented.

In recent years, transformer-based deep learning models such as BERT or GPT-3 have led to impressive results in many natural language processing (NLP) tasks, exhibiting transfer and few-shot learning capabilities. However, despite faring well in benchmarks, current deep learning models for NLP often fail badly in the wild: they are bad at out-of-domain generalization, they do not exploit contextual information, they are poorly calibrated, and their memory is not traceable. These limitations stem from their monolithic architectures, which are good for perception, but unsuitable for tasks requiring higher-level cognition. In this project, I attack these fundamental problems by bringing together tools and ideas from machine learning, sparse modeling, information theory, and cognitive science, in an interdisciplinary approach. First, I will use uncertainty and quality estimates for utility-guided controlled generation, combining this control mechanism with the efficient encoding of contextual information and integration of multiple modalities. Second, I will develop sparse and structured memory models, together with attention descriptive representations towards conscious processing. Third, I will build mathematical models for sparse communication (reconciling discrete and continuous domains), supporting end-to-end differentiability and enabling a shared workspace where multiple modules and agents can communicate. I will apply the innovations above to highly challenging language generation tasks, including machine translation, open dialogue, and story generation. To reinforce interdisciplinarity and maximize technological impact, collaborations are planned with cognitive scientists and with a scale-up company in the crowd-sourcing translation industry.

With the fast development of electronic devices and computing technologies, various emerging applications (e.g., big data analysis, artificial intelligence and 3-dimensional (3D) media, Internet of things, etc.) have been entering our society with significant amounts of data traffic. While mobile networks are already indispensable to our society for “anywhere anytime connection,” one main characteristic of future mobile networks (i.e., B5G: Beyond 5G) is the very huge amount of data, which requires very high throughput per devices (multiple Gbps, up to Tera bps: Tbps) and multiple Tbps per area efficiency (Tbps/km2). Though some disrupting 5G technologies may provide a few Gbps service, it is still not able to achieve hundreds of Gbps or Tbps rates. In the near future, the peak rate of mobile communication networks is expected to reach hundreds of Gbps or even Tbps rates, which requires either very high spectrum efficiency (e.g., much higher than 10 bits/s/Hz) in millimetre wave bands or very large bandwidth (e.g., more than 20GHz) . While the former is very challenging, the latter can be achieved in THz bands (roughly, 100GHz to 10THz). The design of ubiquitous access with very high rates in mobile and heterogeneous network (HetNet) environments is the key to the development of future mobile networks, and so the objective of this collaborative 6G-TERAFIT is to create a knowledge transfer between the researchers and the engineers who will contribute to the design and implementation of future B5G ultra-fast networks and create the pedestal for them to become potential leaders in the resulting scientific, technological, and industrial fields. This project is committed to creating an “excellent” educational training platform that is multi-disciplinary and inter-sectoral in nature.

We envision a radically new technology for in-vivo bioresorbable chemical sensing, where optical devices, power and light sources, synthetic receptors - made out of materials that completely dissolve with biologically benign byproducts in biofluids - will be developed and integrated together. The sensing system, the size of 1 EuroCent, will be coated by a long-lived biocompatible polymer designed with on-demand degradation, then implanted in the body to monitor in-vivo, in-situ, and in real-time a chemotherapeutic drug, doxorubicin, commonly used to treat cancer; the system is then fully and safely RESORBed once no more needed using an external temperature-trigger that initiates the dissolution of the protecting coating and, in turn, of the system, avoiding device-retrieval surgery that may cause tissue lesion/infection. The general objective is to demonstrate fabrication, operation (2 months) in-vivo and in real-time - then dissolution - of such a bioresorbable chemical sensing system for the detection of doxorubicin in an animal model. This will break a new ground in in-situ monitoring of chemotherapeutic drug enabling – for the first time – a fine tuning of the drug dose at the tumor site, increasing patient survival rate. Being aware of the project risks, we have broken down the general into different specific objectives, identified a set of Key Performance Indicators, alternative material synthesis/device fabrication techniques, mitigation measures to tackle major risks. The RESORB technology truly represents the foundation of a future technology for personalized medicine, enabling to address a number of medical issues for which continuous and localized monitoring of specific analytes (i.e., biomarkers and drugs) in-vivo for a prescribed time is of chief importance, e.g., acute trauma treatment, post-surgery sepsis, drug therapeutic profiling, and other, all examples for which ex-situ analysis of biofluids has proved to be not fully adequate for clinical needs.