Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 4
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Sub-milliwatt threshold power and tunable-bias all-optical nonlinear activation function using vanadium dioxide for wavelength-division multiplexing photonic neural networks

  • Parra Gomez, Jorge|||0000-0003-4610-3411
  • Sanchis Kilders, Pablo|||0000-0003-2984-4218
  • Navarro-Arenas, Juan
[EN] The increasing demand for efficient hardware in neural computation highlights the limitations of electronic-based systems in terms of speed, energy efficiency, and scalability. Wavelength-division multiplexing (WDM) photonic neural networks offer a high-bandwidth, low-latency alternative but require effective photonic activation functions. Here, we propose a power-efficient and tunable-bias all-optical nonlinear activation function using vanadium dioxide (VO2) for WDM photonic neural networks. We engineered a SiN/BTO waveguide with a VO2 patch to exploit the phase-change material¿s reversible insulator-to-metal transition (IMT) for nonlinear activation. We conducted numerical simulations to optimize the waveguide geometry and VO2 parameters, minimizing propagation and coupling losses while achieving a strong nonlinear response and low-threshold activation power. Our proposed device features a sub-milliwatt threshold power, a footprint of 5 ¿m, and an ELU-like activation function. Moreover, the bias of our device could be thermally tuned, improving the speed and power efficiency. On the other hand, performance evaluations using the CIFAR-10 dataset confirmed the device¿s potential for convolutional neural networks (CNN). Our results show that a hybrid VO2/SiN/BTO platform could play a prominent role in the path toward the development of high-performance photonic neural networks., This work is supported by the following funding: European Health and Digital Executive Agency (101070690 (PHOENIX)); Agencia Estatal de Investigación (PID2022-137787OB-I00); Generalitat Valenciana (PROMETEO Program(CIPROM/2022/14)); Ministerio de Universidades ( Margarita Salas (MS21-037)); Universitat Politècnica de València (Grants PAID-06-23, PAID-10-23).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Ultra-high endurance silicon photonic memory using vanadium dioxide

  • Seoane-Martínez, Juan José
  • Navarro-Arenas, Juan
  • Recaman, Maria
  • Koen Schouteden
  • Locquet, Jean-Pierre
  • Parra Gomez, Jorge|||0000-0003-4610-3411
  • Sanchis Kilders, Pablo|||0000-0003-2984-4218
[EN] Silicon photonics arises as a viable solution to address the stringent resource demands of emergent technologies, such as neural networks. Within this framework, photonic memories are fundamental building blocks of photonic integrated circuits that have not yet found a standardized solution due to several trade-offs among different metrics such as energy consumption, speed, footprint, or fabrication complexity, to name a few. In particular, a photonic memory exhibiting ultra-high endurance performance (>106¿cycles) has been elusive to date. Here, we report an ultra-high endurance silicon photonic volatile memory using vanadium dioxide (VO2) exhibiting a record cyclability of up to 107¿cycles without degradation. Moreover, our memory features an ultra-compact footprint below 5¿µm with the potential for nanosecond and picojoule programming performance. Our silicon photonic memory could find application in emerging photonic applications demanding a high number of memory updates, such as photonic neural networks with in situ training., This work was supported by funding from the European Union s Horizon Europe by the granting Authority HADEA (European Health and Digital Executive Agency) under Grant Agreement No 101070690 (PHOENIX). Grant PID2022-137787OB-I00 funded by MCIN/AEI/10.13039/501100011033 and by "ERDF A way of making Europe", and PROMETEO Program (CIPROM/2022/14) funded by Generalitat Valenciana are also acknowledged. J.N.A acknowledges grant from University of Valencia/Ministry of Universities (Government of Spain), modality Margarita Salas (MS21-037), funded by the European Union, Next-Generation EU. J. Parra would like to acknowledge Universitat Politècnica de València for his grant 2-PAID-10-22.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Enhanced BaTiO3/Si3N4 integrated photonic platform with VO2 technology for large-scale neuromorphic computing [Invited]

  • Seoane-Martínez, Juan José
  • Navarro-Arenas, Juan
  • Parra Gomez, Jorge|||0000-0003-4610-3411
  • Sanchis Kilders, Pablo|||0000-0003-2984-4218
[EN] The hybrid barium titanate (BaTiO3 or BTO) - silicon nitride (Si3N4 or SiN) platform integrated on silicon has been established as a promising candidate for implementing photonic integrated circuits with unique features in terms of high modulation speeds and low transmission loss. However, despite the high Pockels effect in BTO, switching device footprints are relatively large, which could compromise their use in applications with large scalability requirements, such as neuromorphic computing hardware. To address this limitation, we propose the integration of vanadium dioxide (VO2) in the SiN/BTO platform to enable ultra-compact amplitude switching devices offering a scalar multiplication functionality with multilevel operation. More concretely, an electronically reprogrammable switching device with a 5-bit amplitude encoding capability and an insertion loss of only 0.5 dB is developed. The proposed device is built with a 9-µm-long VO2/SiN/BTO waveguide structure integrated with an efficient microheater using a transparent conducting oxide. Such a vital building block would offer significant potential for developing more complex photonic integrated circuits, including dot-product or matrix-vector multiplication engines. The combination of high speed, low loss, and reduced footprint makes the proposed enhanced platform an attractive solution for application in scalable and energy-efficient neuromorphic computing hardware., European Health and Digital Executive Agency (101070690 (PHOENIX)); Agencia Estatal de Investigación (PID2022-137787OB-I00); Generalitat Valenciana (PROMETEO Program (CIPROM/2022/14)); Ministerio de Universidades ( Margarita Salas (MS21-037)); Universitat Politècnica de València (Grant 2-PAID-10-22).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

On-chip electro-optical spiking VO2/Si device with an inhibitory leaky integrate-and-fire response

  • Morcillo-Roldán, Juan Francisco
  • Sanchis Kilders, Pablo|||0000-0003-2984-4218
  • Parra Gomez, Jorge|||0000-0003-4610-3411
[EN] Integrated photonic neuromorphic computing has emerged as an appealing technology to replicate human brain functionalities by leveraging the benefits of integrated photonics. In this context, spiking neural networks (SNNs) offer a highly energy-efficient artificial neural network due to the event-based information processing. In this work, we propose an electro-optical spiking device for the silicon photonics platform based on a VO2/Si waveguide and a meander-shaped microheater. A leaky integrate-and-fire (LIF) response with inhibitory optical spiking is achieved by harnessing the temperature-driven insulator-metal transition (IMT) of VO2 combined with thermal dynamics. Hence, our proposal would not require complex electronic circuitry bottlenecking the neural network. Our numerical simulations demonstrate firing upon milliwatt and nanosecond range postsynaptic pulses and adjustable spike characteristics in the time domain with sub-microsecond values. Therefore, our spiking device could enable scalable and energy-efficient photonic-based SNNs., Universitat Politècnica de València (PAID-06-23, PAID-10-23); Agencia Estatal de Investigación (PID2022- 137787OB-I00); European Health and Digital Executive Agency (101070690 (PHOENIX)); Conselleria d Educació, Investigació, Cultura i Esport (PROMETEO Program (CIPROM/2022/14)).