The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
We demonstrate experimentally for the first time the simultaneous excitatory and inhibitory dynamics in a graphene excitable laser. This technology potentially opens novel spike processing functionality for future neuromorphic photonic systems.
We present an integrated, multi-channel laser processor. It utilizes a novel photodetector-to-laser O/E/O receiverless link to receive multiple wavelength inputs. To our knowledge, this is the first laser neuron compatible with a wavelength-based networking scheme.
Excitable laser dynamics is proposed to realize spike coded bit sequence generation. Numerical simulation based on integrated two-section excitable laser and experimental investigation using graphene excitable fiber laser are both demonstrated.
Brain-inspired distributed computing has attracted attention for its energy-efficient processing, and photonic neuromorphic hardware can overcome latency vs. fan-in tradeoffs from neuromorphic electronics. Here, we introduce system-level, physically detailed modeling tools for photonic neural networks, and use them to study the behavior of attractor networks.
Photonic integrated circuit technology has the potential to revolutionize optical information processing, beyond conventional binary-logic approaches, granting the capacity of complex, ultrafast categorization and decision making. We will discuss the progress and requirements of scalable and reconfigurable emerging brain-inspired photonic hardware platforms.
Silicon photonic integration could enable high-performance brain-inspired photonic processors. We demonstrate a 3 node recurrent photonic neural network. Cusp and Hopf bifurcations induced by synaptic reconfiguration are shown as proof-of-concept. The prototype represents an early step towards network-based models of physical computing with integrated photonics.
We experimentally observe and simulate coherent inter-resonator effects specific to microring weight banks. An analysis based on this effect results in quantitative performance limits of weight banks, a key subcircuit for multivariate analog signal processing and scalable analog interconnect approaches in silicon photonics.
We demonstrate 4-channel weighted addition in a silicon microring filter bank with 3.8 bit accuracy. Scalable calibration techniques are developed for thermal cross-talk and cross-gain saturation. Practical weight control is essential for large-scale photonic processing based on microrings.
Analog Interconnection networks are configured setting connection weights. Microring weight banks are a key device for making such networks in silicon photonic circuits. We demonstrate a small analog network in silicon using a single node with simple dynamics, showing dynamics parameterized by microring weight banks.
We propose and experimentally demonstrate a microwave photonic system that iteratively performs principal component analysis on partially correlated, 8-channel, 13 Gbaud signals. The system that is presented is able to adapt to oscillations in interchannel correlations and follow changing principal components. The system provides advantages in bandwidth performance and fan-in scalability that are...
Spiking neural networks (SNN) have inherent advantages over traditional computing architectures for many computational problems such as adaptive control, sensory processing, and pattern recognition. Recently, a graphene-based fiber laser has been shown that demonstrates all the key properties of spike processing: logic-level restoration, cascadability and input-output isolation, in one device[1]....
We experimentally demonstrate a compact optical steganography method using chirped fiber Bragg gratings. The stealth signals are carried by wide band amplified spontaneous emission noise, which has strong dispersion effect for pulse stretching.
We propose an integrated optoelectronic circuit capable of tracking on real-time the first principal component of an array of wideband analog RF signals. Preliminary results warrants the suitability of photonic components for wideband PCA.
Many-to-one connections are difficult to implement in excitable laser neurons. We design and simulate an O/E/O receiver-less link from photodetector to laser that accepts many spiking inputs (large fan-in) without significant bandwidth degradation.
We experimentally demonstrate resonant switching and pulse regeneration using a graphene-based excitable fiber ring laser and simulate an analogous integrated device structure. Such devices could find use in pulse regeneration or optical computing.
We demonstrate a simple photonic spatiotemporal pattern recognition (polychronization) circuit enabled by cascading two graphene excitable lasers. This technology is a potential candidate for information processing and computing.
We demonstrate a photonic bistable spiking circuit with a graphene excitable laser for cascadable logic. This technology could be a potential candidate for applications in novel all optical devices for information processing and computing.
We discuss a novel application of a photonic circuit for integrated high-performance neuromorphic signal processing. Large fan-in is an especially important capability in distributed systems; however, electronic physics impose tradeoffs between bandwidth performance and fan-in degree. A circuit developed in the field of radio frequency (RF) photonics, wavelength(λ)-fan-in does not exhibit a corresponding...
Sillion photonic platform development has revolved around point-to-point links for multi-core computing systems. We examine an opportunity for this technology to extend to unconventional architectures that rely heavily on interconnect performance. Broadcast-and-weight is a new approach for joining neuron-inspired processing and optical interconnect physics.
We demonstrate a photonic coincidence detection circuit with a graphene excitable laser. This technology is a potential candidate for applications in novel all-optical devices for information processing and computing.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.