Femtojoule per MAC Neuromorphic Photonics: An Energy and Technology Roadmap

Totović, Angelina; Dabos, George; Passalis, Nikolaos; Tefas, Anastasios; Pleros, Nikos

doi:10.1109/jstqe.2020.2975579

Cited by 73 publications

(42 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, we discuss the computational efficiency of the proposed circuit. Although there are many indices for expressing the performance of a computing device, the multiply-accumulate per second speed (MAC•s −1 ) is now widely considered to be a milestone in the photonic neuromorphic computation region 9,53,54 . Thus, we discuss this index for the reservoir computer.…”

Section: Discussionmentioning

confidence: 99%

Scalable reservoir computing on coherent linear photonic processor

2021

View full text Add to dashboard Cite

Photonic neuromorphic computing is of particular interest due to its significant potential for ultrahigh computing speed and energy efficiency. The advantage of photonic computing hardware lies in its ultrawide bandwidth and parallel processing utilizing inherent parallelism. Here, we demonstrate a scalable on-chip photonic implementation of a simplified recurrent neural network, called a reservoir computer, using an integrated coherent linear photonic processor. In contrast to previous approaches, both the input and recurrent weights are encoded in the spatiotemporal domain by photonic linear processing, which enables scalable and ultrafast computing beyond the input electrical bandwidth. As the device can process multiple wavelength inputs over the telecom C-band simultaneously, we can use ultrawide optical bandwidth (~5 terahertz) as a computational resource. Experiments for the standard benchmarks showed good performance for chaotic time-series forecasting and image classification. The device is considered to be able to perform 21.12 tera multiplication–accumulation operations per second (MAC ∙ s−1) for each wavelength and can reach petascale computation speed on a single photonic chip by using wavelength division multiplexing. Our results are challenging for conventional Turing–von Neumann machines, and they confirm the great potential of photonic neuromorphic processing towards peta-scale neuromorphic super-computing on a photonic chip.

show abstract

Section: Discussionmentioning

confidence: 99%

Scalable reservoir computing on coherent linear photonic processor

2021

View full text Add to dashboard Cite

show abstract

“…In practice, the devices that need to be powered in the wireless pipeline monitoring system mainly include sensors and wireless signal transmission devices, for which a microwatt level power consumption can be estimated [ 33 ]. According to the measured output of the TENG fabricated in this paper, the output may meet the power consumption requirements of these devices.…”

Section: Experimental Testmentioning

confidence: 99%

“…Substituting Equations (3) and (4) into Equation (2), we have:

in which,

is the natural circular frequency of pipe,

is the hysteretic damping coefficient of the pipe defined by

[ 33 ],

and

respectively the generalized loading and generalized mass of the j -th normal mode, L the length of the pipe.…”

Section: Theoretical Modellingmentioning

confidence: 99%

“…In Figure 5b, both the peak current and the average output power has no significant degradation after 10,000 reciprocating cycles, indicating an excellent stability and durability of the TENG device. In practice, the devices that need to be powered in the wireless pipeline monitoring system mainly include sensors and wireless signal transmission devices, for which a microwatt level power consumption can be estimated [33]. According to the measured output of the TENG fabricated in this paper, the output may meet the power consumption requirements of these devices.…”

Section: Experimental Testmentioning

confidence: 99%

“…According to the Euler-Bernoulli beam theory [34], the governing equation of the simply-supported cantilever beam can be written as Equation (1). The vibration equations for TENG device is written as Equation 2based on D'Alembert's principle: In practice, the devices that need to be powered in the wireless pipeline monitoring system mainly include sensors and wireless signal transmission devices, for which a microwatt level power consumption can be estimated [33]. According to the measured output of the TENG fabricated in this paper, the output may meet the power consumption requirements of these devices.…”

Section: Dynamic Responses Of the Pipe And The Mass-spring Basementioning

confidence: 99%

See 2 more Smart Citations

A Contact-Mode Triboelectric Nanogenerator for Energy Harvesting from Marine Pipe Vibrations

Zhang

Wang

et al. 2021

Sensors

View full text Add to dashboard Cite

Structural health monitoring is of great significance to ensure the safety of marine pipes, while powering the required monitoring sensors remains a problem because the ocean environment is not amenable to the traditional ways of providing an external power supply. However, mechanical energy due to the vortex-induced vibration of pipelines may be harvested to power those sensors, which is a convenient, economic and environmentally friendly way. We here exploit a contact-separation mode triboelectric nanogenerator (TENG) to create an efficient energy harvester to transform the mechanical energy of vibrating pipes into electrical energy. The TENG device is composed of a tribo-pair of dielectric material films that is connected to a mass-spring base to guarantee the contact-separation motions of the tribo-pair. Experimental tests are conducted to demonstrate the output performance and long-term durability of the TENG device by attaching it to a sample pipe. A theoretical model for the energy harvesting system is developed for predicting the electrical output performance of the device. It is established that the normalized output power depends only on two compound variables with all typical factors taken into consideration simultaneously. The simple scale law is useful to reveal the underlying mechanism of the device and can guideline the optimization of the device based on multi-parameters analyses. The results here may provide references for designing contact-mode TENG energy harvesting devices based on the vibration of marine pipes and similar structures.

show abstract

Photonic Neural Networks with Kramers–Kronig Activation

Tian¹,

Han

Shu

et al. 2023

Advanced Photonics Research

View full text Add to dashboard Cite

Photonic neural networks (PNNs) are promising to replace conventional deep learning hardware due to their potentially higher energy efficiency and computational speed. Other than the fast progress on optical linear transformer, the nonlinear activators are much less mature. Usually, the optical activators employ the nonlinear mapping from input amplitude to output amplitude, which is limited by high threshold and loss, or rely on additional bias voltage and heterogeneous integration of external circuits. Herein, the activation induced by the Kramers–Kronig relationship is proposed, i.e., the connection between amplitude and phase of the light field, but not purely the light amplitude itself, namely, Kramers–Kronig activation (KKA). PNN with KKA exhibits learning capability apparently better than the activation‐free linear network and comparable to the network with popular activation functions like ReLU, Softplus, and so on. Moreover, PNN with KKA is highly programmable and cascadable supporting ultra‐deep networks. The essence of KKA is attributed to the learning of nonlinear features in relatively low‐dimensional RN × N space by linear features in high‐dimensional CN × N space. Considering besides the amplitude–phase coupling, several other parameters like wavelength, frequency, polarization, etc., can be also mutually linked. This approach may expand to new avenues for optical activations.

show abstract

Femtojoule per MAC Neuromorphic Photonics: An Energy and Technology Roadmap

Cited by 73 publications

References 73 publications

Scalable reservoir computing on coherent linear photonic processor

Scalable reservoir computing on coherent linear photonic processor

A Contact-Mode Triboelectric Nanogenerator for Energy Harvesting from Marine Pipe Vibrations

Photonic Neural Networks with Kramers–Kronig Activation

Contact Info

Product

Resources

About