Compressed and Deep Sensing: Models, Structures, and Tasks

Informacje na temat projektu

CoDeS

Identyfikator umowy o grant: 101000967

DOI

10.3030/101000967

Projekt został zamknięty

Data podpisania przez KE 30 Kwietnia 2021

Data rozpoczęcia 1 Lipca 2021

Data zakończenia 30 Czerwca 2026

Finansowanie w ramach

• EXCELLENT SCIENCE - European Research Council (ERC)

Koszt całkowity

€ 3 000 000,00

Wkład UE

€ 3 000 000,00

Koordynowany przez

WEIZMANN INSTITUTE OF SCIENCE
Israel

Periodic Reporting for period 2 - CoDeS (Compressed and Deep Sensing: Models, Structures, and Tasks)

Okres sprawozdawczy: 2023-01-01 do 2024-06-30

Podsumowanie kontekstu i ogólnych celów projektu

In this project we are exploring how we can build systems that are far more efficient than those existing today by designing them in an end-to-end holistic fashion. Throughout their design we take into account ay structure in the input or system, and we also consider the final system task. We have been exploring ways in which this approach can make various types of systems more efficient including communication systems, radar, medical devices and more. In all of these areas we consider the basic question: Can we measure only what is necessary to perform our desired system task by jointly optimizing the entire signal acquisition and processing chain?

Prace wykonane od początku projektu do końca okresu sprawozdawczego oraz najważniejsze dotychczasowe rezultaty

In our work so far we have been developing fundamental trade-offs between sampling rate, quantization resolution, and task fidelity. We have also developed new results on the bounds of model-based deep learning that takes models into account within the learning process.
On the engineering side we have developed several new ADC devise based on this principle like modulo-based sampling that can accommodate high dynamic range, and time-based sampling that allows removing the system clock all together. On the application side we have been considering 6G networks, joint sensing and communication systems, efficient vital sign monitoring, ECG monitoring and more.

Innowacyjność oraz oczekiwany potencjalny wpływ (w tym dotychczasowe znaczenie społeczno-gospodarcze i szersze implikacje społeczne projektu)

On the engineering side we have developed several new ADC devise based on this principle like modulo-based sampling that can accommodate high dynamic range, and time-based sampling that allows removing the system clock all together. On the application side we have been considering 6G networks, joint sensing and communication systems, efficient vital sign monitoring, ECG monitoring and more.