The University of Trento, which is already working on a research project on photonics and the brain, will soon launch an even more ambitious project that may have an impact on the ICT industry (from metropolitan networks in smart cities to data centers required in an increasingly digital society).
Funding for new project (named ALPI, whose goal is to develop an advanced optical neural platform) will be provided to the Department of Physics of UniTrento by the European Research Council. The Department secured an ERC Proof of Concept Grant within the research and innovation programme Horizon 2020, to explore the commercial or societal potential of research results.
This grant, awarded by the most important organization in Europe that finances frontier research, also involves HIT - Trentino Innovation Hub, the technology transfer organization that applied for funding with the Department of Physics and is the direct beneficiary of the grant to implement the activities that are necessary to actually provide the Proof of Concept, i.e. to make a functioning prototype of the new technology.
This type of grant aims in particular to verify the technical feasibility of the project, explore potential markets and file patent applications. In the first phase, HIT examined the innovation potential of the new technology, comparing its features with those of competing technologies from the USA and China. HIT will provide assistance to professor Pavesi in drafting of the business plan and the marketing strategy.
SM Optics (a company of the Siae Microelettronica group, a corporation providing telecommunication network equipment) and ST Microelectronics (an Italian-French electronics and semiconductor manufacturer), are also involved in the project as advisors.
News of grant were released in July, but things are now starting to move. The project will be officially launched online on Wednesday, 25 November, at 16.00.
Alpi (ALPl optical signal recovery by Photonic neural network Integrated in a transceiver module) is coordinated by Lorenzo Pavesi, professor of the Department of Physics of the University of Trento. It received funding for 150,000 euro and aims to develop, by the spring of 2022, a photonic neural network integrated within an optical transceiver providing real time compensation of fiber nonlinearities which degrade optical signals.
The project represents a step forward from another European project, Backup, that is focused on a similar problem and is coordinated by the Department of Physics of UniTrento involving, in a complex multidisciplinary research activity, the Department of Information Engineering and Computer Science, the Department of Cellular, Computational and Integrative Biology - Cibio, and the Center for Mind/Brain Sciences.
The project
Alpi (ALPl optical signal recovery by Photonic neural network Integrated in a transceiver module) aims at the integration of a photonic neural network within an optical transceiver which is able to recover distortion on the optical transmitted data. Based on a deep learning approach, the new compact device aims to provide real time compensation of fiber nonlinearities which degrade optical signals. To achieve that goal, the research team propose to integrate in the optical link the neuromorphic photonic circuits that they are currently developing in the Backup project.
Backup, whose goal is to unveil the relationship between brain connectivity and function by integrated photonics, aims to develop hybrid interfaces between a neuronal network and a photonic integrated circuit, to emulate the way in which the brain processes and memorizes information.
The second goal of Alpi is the technology transfer of the prototype. For these purposes, patents will be filed and a business plan will be developed in partnership with other companies. Based on market analysis, the device could be commercialized in the field of optical interconnection (from data centers to metropolitan networks).