2022 4th International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP), Journal Year: 2024, Volume and Issue: unknown, P. 123 - 130
Published: Nov. 29, 2024
Language: Английский
2022 4th International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP), Journal Year: 2024, Volume and Issue: unknown, P. 123 - 130
Published: Nov. 29, 2024
Language: Английский
ACS Photonics, Journal Year: 2024, Volume and Issue: 11(2), P. 723 - 730
Published: Jan. 10, 2024
With the rapid development of Internet Things, how to efficiently store, transmit, and process massive amounts data has become a major challenge now. Optical neural networks based on nonvolatile phase change materials (PCMs) have breakthrough point due their zero static power consumption, low thermal crosstalk, large-scale, high efficiency. However, current photonic devices cannot meet multilevel requirements in neuromorphic computing limited storage capacity. Here, new way for increasing capacity is paved from perspective modulation crystallization kinetics PCMs. A more progressive transition amorphous crystalline states achieved through grain-refinement phenomenon induced by nitrogen (N) element doping Ge2Sb2Te5 (GST), giving precise access multibit states. By integrating N-doped (N-GST) with waveguide, high-capacity device enabling over 7 bits (∼222 levels) first time. The increased nearly times compared state-of-the-art (∼32 levels). An optical convolutional network successfully established MINIST handwritten digit recognition task mapping synapse weight multiple levels, accuracy up 96.5% achieved. Our work provides strategy integrated demonstrates enormous application potential field large-scale networks.
Language: Английский
Citations
13IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 26562 - 26580
Published: Jan. 1, 2024
In recent years, the energy consumption of IoT edge nodes has significantly increased due to communication process. This necessitates need offload more computation minimize data transmission over network. To achieve this, higher-performance CPUs and memory are required on nodes. this context, we propose an energy-efficient architecture specifically designed for STT-MRAM is a promising technology that offers potential replacements SRAM Flash in devices. exhibits notable advantages traditional technologies, such as non-volatility retention without continuous power supply efficiency, resulting extended battery life portable devices applications. Its higher density scalability through standard fabrication processes further enhances its appeal next-generation solutions. However, high write main disadvantage. Previous works have explored relaxation CPU cache but there extend approach paper, multi-retention memory. Additionally, mapping scheme suggested examine impact relaxed levels consumption. best our knowledge, first study thoroughly investigate optimal thermal stability factor value applications while also considering mapping. The proposed reduces by average 70% up 83% compared currently used non-volatile architecture. Furthermore, two mappings easy use savings just 5% away from ideal
Language: Английский
Citations
8Journal of Systems Architecture, Journal Year: 2023, Volume and Issue: 144, P. 103008 - 103008
Published: Oct. 9, 2023
Language: Английский
Citations
11IEEE Access, Journal Year: 2025, Volume and Issue: 13, P. 19898 - 19908
Published: Jan. 1, 2025
Language: Английский
Citations
0Proceedings of the 28th Asia and South Pacific Design Automation Conference, Journal Year: 2025, Volume and Issue: unknown, P. 128 - 134
Published: Jan. 20, 2025
Language: Английский
Citations
0Physica B Condensed Matter, Journal Year: 2025, Volume and Issue: unknown, P. 417274 - 417274
Published: April 1, 2025
Language: Английский
Citations
0Journal of Solid State Chemistry, Journal Year: 2025, Volume and Issue: unknown, P. 125407 - 125407
Published: May 1, 2025
Language: Английский
Citations
0physica status solidi (b), Journal Year: 2025, Volume and Issue: unknown
Published: May 11, 2025
Phase‐change memory (PCM) represents a next‐generation advancement in nonvolatile data storage technology. Its potential stems from high scalability, rapid switching response, and long‐term stability, making it suitable for nanoscale applications. PCM operates by inducing structural transitions within phase‐change materials, toggling between an ordered crystalline state disordered amorphous phase. This transformation significantly alters optical electrical characteristics, which is crucial electronic applications recording. Among notable GST225 In 2 Se 3 exhibit remarkable properties, enabling efficient information storage. this article, we have discussed phase change materials. technology presents significant advantages, but there are several challenges limitations must be addressed to become universal solution. Some of these such as speed, retention, power consumption, cycling endurance article.
Language: Английский
Citations
0Acta Optica Sinica, Journal Year: 2025, Volume and Issue: 45(9), P. 0913001 - 0913001
Published: Jan. 1, 2025
Citations
0Published: March 25, 2024
Language: Английский
Citations
3