Reconfigurable optoelectronic memristive architecture based on doped nanowire array for in-memory parallel perception...

Area of Science:

• Materials Science
• Computer Engineering
• Nanotechnology

Background:

• Optimizing computational redundancy in artificial intelligence requires advanced hardware functional integration for emergent computing paradigms.
• Device design for parallel perception and in-memory computing presents significant challenges.

Purpose of the Study:

• To develop a state-of-the-art integrated functional memory for in situ parallel perception and in-sensor computation.
• To demonstrate a reconfigurable optoelectronic memristive architecture (ROMA) based on a doped nanowire array.

Main Methods:

• Fabrication of a memristor based on In2S3-XAsX exhibiting tunable optoelectronic properties via vacancy engineering.
• Doping modulation to control memristive performance and photoconductive retention.
• Monolithic integration of a nanowire array for parallel processing.

Main Results:

• The In2S3-XAsX memristor showed favorable photoconductive retention and reconfigurable optoelectronic modulation.
• A two-orders-of-magnitude improvement in discriminative state was achieved with double output signal sampling.
• Recognition and encoding of 12-bit binary optical signals on a single column were demonstrated.

Conclusions:

• The reconfigurable optoelectronic memristive architecture (ROMA) offers an efficient hardware foundation for highly parallel and distributed computational paradigms.
• Vacancy engineering in In2S3-XAsX provides a pathway for tunable memristive properties.
• This work advances in-sensor computation for artificial intelligence applications.