We present a compact, single-shot photonic-integrated-circuit based RF spectrum analyzer that combines a loss-compensating speckle spectrometer design with an interferometric RF-to-optical encoding scheme to achieve 10 MHz resolution across a bandwidth of 10 GHz.

In this work, we propose a Nonlinear Pluggable Optic (NLPO) transceiver that combines the low latency and low power consumption of Linear Pluggable Optics (LPO) with the range and robustness of digital signal processing (DSP)-based transceivers. The proposed NLPO uses an analog photonic Next-Generation Reservoir Computing (NGRC) architecture, constructed on a photonic integrated circuit (PIC), to compensate for electrical-domain distortions as well as optical-channel impairments from chromatic dispersion and Kerr nonlinearity. Focusing on a simulated 50 GBd PAM-4 link, we find that the NGRC-based NLPO not only extends the range of LPO, but actually outperforms DSP-based solutions as well. Our simulations reveal two key advantages compared to DSP-based Intensity Modulation/Direct Detection (IM/DD) links: (1) the NGRC can take advantage of the optical phase information without requiring a local oscillator and (2) the NGRC can optically sample the transmitted data well above the symbol rate without requiring high-bandwidth electronics. This work showcases the potential for photonic NGRCs to outperform state-of-the-art digital solutions in real-world applications and opens a path to low-latency, lower-power IM/DD links at ranges of 10s of km.

Designing miniaturized optical spectrometers is an increasingly active area of research as spectrometers are crucial components for a wide range of applications including chemical and material analysis, medical diagnostics, classical and quantum sensing, characterization of light sources, and radio frequency (RF) spectrum analysis. Among these applications, designing on-chip spectrometers for RF spectrum analysis is particularly challenging since it requires combining high resolution and large bandwidth with a fast update rate. Existing chip-scale spectrometers cannot achieve the resolution required for RF analysis, setting aside challenges in maintaining a fast update rate and broad bandwidth. In this work, we address these challenges by introducing a silicon photonic integrated circuit (PIC)-based RF spectrum analyzer that combines an ultra-high-resolution speckle spectrometer with an interferometric RF-to-optical encoding scheme. The PIC-based speckle spectrometer uses a path-mismatched multimode interferometer with inverse designed splitters to compensate for waveguide loss, enabling a record-high resolution of 100 MHz (0.8 pm at a wavelength of 1550 nm). To further improve the resolution of the overall RF spectrum analyzer, we modify the RF-to-optical encoding scheme by directing the RF signal through a path mismatched interferometer and encoding the outputs of the RF interferometer on separate optical carriers. This further reduces the RF spectral correlation width of the combined system, enabling the RF spectrum analyzer to resolve RF tones separated by 10 MHz across a bandwidth of 10 GHz. Since this approach operates as a single-shot spectrometer, it can support fast update rates, providing a path to compact, persistent wideband RF spectrum analysis. Brandon Redding and colleagues report a silicon photonic integrated circuit-based RF spectrum analyzer that combines a speckle spectrometer with interferometric RF-to-optical encoding scheme. The device achieves a record-high optical resolution of 100 MHz across a bandwidth of 10 GHz enabling compact wideband RF sensing.

The need for real-time wideband radio frequency (RF) spectral analysis is driven by continued advances in modern wireless communications and RADAR systems used both for military and civilian applications. However, wideband RF sensing presents a challenge for typical high-speed analog to digital converters (ADC) since ADCs capable of operating continuously are typically limited to monitoring less than 1 GHz bands. Here, we leverage the high bandwidth of photonics to build a Nyquist folding receiver (NYFR) that uses an asymmetric optical frequency comb and a deep convolutional neural network to monitor a ∼5 GHz bandwidth using a 1 GS/s ADC with a 1 MHz update rate. We tested the deep-learning assisted NYFR on several signal classes, including linear chirps, nonlinear chirps, and continuous wave signals. The system presented here tackles many of the limitations of typical NYFR systems, including the ability to recover signals that cross Nyquist zones and the ability to detect multiple signals simultaneously. We also show that using a non-linear encoding to map the RF signal into the optical domain can improve the accuracy of the recovered RF spectrum.

In this work, we introduce and experimentally demonstrate a photonic frequency-multiplexed next generation reservoir computer (FM-NGRC) capable of performing real-time inference at GHz speeds. NGRCs apply a feed-forward architecture to produce a feature vector directly from the input data over a fixed number of time steps. This feature vector, analogous to the reservoir state in a conventional RC, is used to perform inference by applying a decision layer trained by linear regression. Photonic NGRC provides a flexible platform for real-time inference by forgoing the need for explicit feedback loops inherent to a physical reservoir. The FM-NGRC introduced here defines the memory structure using an optical frequency comb and dispersive fiber, while the sinusoidal response of electro-optic Mach–Zehnder interferometers controls the nonlinear transform applied to elements of the feature vector. A programmable waveshaper modulates each comb tooth independently to apply the trained decision layer weights in the analog domain. We apply the FM-NGRC to solve the benchmark nonlinear channel equalization task; after theoretically determining feature vectors that enable high-accuracy distortion compensation, we construct an FM-NGRC that generates these vectors to experimentally demonstrate real-time channel equalization at 5 GS/s with a symbol error rate of ∼2.5×10−3.

Stimulated Brillouin scattering (SBS) is often considered to be problematic, since it limits optical power in telecom and RF photonics applications. However, the highly efficient and narrowband SBS process also provides unique functionality that can be used for a variety of applications. In this talk, I will present recent work exploiting SBS for applications in distributed fiber sensing and high-resolution spectroscopy. First, I will present work showing how the complex Stokes and anti-Stokes interactions can be used to perform distributed temperature and strain measurements with immunity to cross-talk. To achieve higher sensitivity, we introduced a technique that operates by exciting up to 1000 Brillouin lasing modes simultaneously in a fiber ring cavity. We then showed that modifications to these schemes can enable high-resolution spectroscopy with unique combinations of bandwidth, resolution, and measurement rate.

The widespread adoption of machine learning and other matrix intensive computing algorithms has renewed interest in analog optical computing, which has the potential to perform large-scale matrix multiplications with superior energy scaling and lower latency than digital electronics. However, most optical techniques rely on spatial multiplexing, requiring a large number of modulators and detectors, and are typically restricted to performing a single kernel convolution operation per layer. Here, we introduce a fiber-optic computing architecture based on temporal multiplexing and distributed feedback that performs multiple convolutions on the input data in a single layer. Using Rayleigh backscattering in standard single mode fiber, we show that this technique can efficiently apply a series of random nonlinear projections to the input data, facilitating a variety of computing tasks. The approach enables efficient energy scaling with orders of magnitude lower power consumption than GPUs, while maintaining low latency and high data-throughput. Optical techniques adopted in optical computing rely on spatial multiplexing, requiring numerous integrated elements and restricting the architecture to perform a single kernel convolution per layer. The authors demonstrate a fiber-optic computing architecture based on temporal multiplexing that performs multiple convolutions in a single layer.

Modern lens designs are capable of resolving greater than 10 gigapixels, while advances in camera frame-rate and hyperspectral imaging have made data acquisition rates of Terapixel/second a real possibility. The main bottlenecks preventing such high data-rate systems are power consumption and data storage. In this work, we show that analog photonic encoders could address this challenge, enabling high-speed image compression using orders-of-magnitude lower power than digital electronics. Our approach relies on a silicon-photonics front-end to compress raw image data, foregoing energy-intensive image conditioning and reducing data storage requirements. The compression scheme uses a passive disordered photonic structure to perform kernel-type random projections of the raw image data with minimal power consumption and low latency. A back-end neural network can then reconstruct the original images with structural similarity exceeding 90%. This scheme has the potential to process data streams exceeding Terapixel/second using less than 100 fJ/pixel, providing a path to ultra-high-resolution data and image acquisition systems. The researchers showcase a silicon-photonics-based analog approach for large-scale image processing that can be deployed for high-speed image compression and de-noising using an auto-encoder framework with minimal power consumption.

We introduce a spectrometer that uses Brillouin lasing to perform scan-free measurements of a 4 THz optical band (1535-1570 nm) with 28 MHz (0.2 pm) resolution and an update rate of 5 ms.

Brillouin spectroscopy has become an important tool for mapping the mechanical properties of biological samples. Recently, stimulated Brillouin scattering (SBS) measurements have emerged in this field as a promising technology for lower noise and higher speed measurements. However, further improvements are fundamentally limited by constraints on the optical power level that can be used in biological samples, which effectively caps the gain and signal-to-noise ratio (SNR) of SBS biological measurements. This limitation is compounded by practical limits on the optical probe power due to detector saturation thresholds. As a result, SBS-based measurements in biological samples have provided minimal improvements (in noise and imaging speed) compared with spontaneous Brillouin microscopy, despite the potential advantages of the nonlinear scattering process. Here, we consider how a SBS spectrometer can circumvent this fundamental trade-off in the low-gain regime by leveraging the polarization dependence of the SBS interaction to effectively filter the signal from the background light via the polarization pulling effect. We present an analytic model of the polarization pulling detection scheme and describe the trade-space unique to Brillouin microscopy applications. We show that an optimized receiver design could provide >25× improvement in SNR compared to a standard SBS receiver in most typical experimental conditions. We then experimentally validate this model using optical fiber as a simplified test bed. With our experimental parameters, we find that the polarization pulling scheme provides 100× higher SNR than a standard SBS receiver, enabling 100× faster measurements in the low-gain regime. Finally, we discuss the potential for this proposed spectrometer design to benefit low-gain spectroscopy applications such as Brillouin microscopy by enabling pixel dwell times as short as 10 μs.

Reservoir computing (RC) is a machine learning paradigm that excels at dynamical systems analysis. Photonic RCs, which perform implicit computation through optical interactions, have attracted increasing attention due to their potential for low latency predictions. However, most existing photonic RCs rely on a nonlinear physical cavity to implement system memory, limiting control over the memory structure and requiring long warm-up times to eliminate transients. In this work, we resolve these issues by demonstrating a photonic next-generation reservoir computer (NG-RC) using a fiber optic platform. Our photonic NG-RC eliminates the need for a cavity by generating feature vectors directly from nonlinear combinations of the input data with varying delays. Our approach uses Rayleigh backscattering to produce output feature vectors by an unconventional nonlinearity resulting from coherent, interferometric mixing followed by a quadratic readout. Performing linear optimization on these feature vectors, our photonic NG-RC demonstrates state-of-the-art performance for the observer (cross-prediction) task applied to the Rössler, Lorenz, and Kuramoto-Sivashinsky systems. In contrast to digital NG-RC implementations, we show that it is possible to scale to high-dimensional systems while maintaining low latency and low power consumption.

We present an RF spectrum analyzer capable of monitoring a 15GHz band with MHz-level resolution and 385kHz update rate. We use Rayleigh-backscattering in single-mode fiber to produce frequency-dependent speckle patterns to recover the RF spectrum.

We review recent work at the US Naval Research Laboratory using stimulated Brillouin scattering in optical fiber for applications in distributed sensing, spectroscopy, and optical signal processing. In particular, we describe recent advances in distributed strain and temperature sensing enabled by simultaneously monitoring the complex Stokes and anti-Stokes Brillouin interactions. We then show how this scheme can be modified to enable high-speed, high-resolution spectroscopy. Finally, we describe how the narrow-linewidth of the SBS process can enable line-by-line optical frequency comb control for applications in RF photonics and optical arbitrary waveform generation.

We demonstrate an all-optical image compression technique using on-chip nanophotonic disordered media to perform local random transformations. Our compression technique is fast, scalable, and may consume about 50 times less energy than electronic compression.

Frequency shifting loops, consisting of a fiber optic ring cavity, a frequency modulator, and an amplifier to compensate for loss, enable high-speed frequency scanning with precise and easily controlled frequency steps. This platform is particularly attractive for applications in spectroscopy and optical ranging. However, amplified spontaneous emission noise accumulates due to the repeated amplification of light circulating in the cavity, limiting the frequency scanning range of existing frequency shifting loops (FSLs). Here, we introduce a cascaded approach which addresses this basic limitation. By cascading multiple FSLs in series with different frequency shifts we are able to dramatically increase the accessible scanning range. We present modeling showing the potential for this approach to enable scanning over ranges up to 1 THz—a tenfold increase compared with the state-of-the-art. Experimentally, we constructed a pair of cascaded FSLs capable of scanning a 200 GHz range with 100 MHz steps in 10 ms and used this platform to perform absorption spectroscopy measurements of an H13C14N cell. By increasing the operating bandwidth of FSLs, the cascaded approach introduced in this work could enable new applications requiring precise and high-speed frequency scanning.

We introduce an approach to enable high-resolution, wide-band, line-by-line manipulation of optical frequency combs. The technique relies on a single seed laser and a pair of frequency loops to produce both the comb and a series of control pulses. The control pulses are used to change the amplitude and phase of each line via a narrow –band (~100 MHz) Brillouin interaction in optical fiber. We generate and manipulate 50 comb lines spaced by 200 MHz with extinction as high as 30 dB and with speeds as high as 10 kHz.

Persistent wideband radio frequency (RF) surveillance and spectral analysis is increasingly important, driven by the proliferation of wireless communication and RADAR technology. However, conventional electronic approaches are limited by the ∼1 GHz bandwidth of real-time analog-to-digital converters (ADCs). While faster ADCs exist, high data rates prohibit continuous operation, limiting these approaches to acquiring short snapshots of the RF spectrum. In this work, we introduce an optical RF spectrum analyzer designed for continuous, wideband operation. Our approach encodes the RF spectrum as sidebands on an optical carrier and relies on a speckle spectrometer to measure these sidebands. To achieve the resolution and update rate required for RF analysis, we use Rayleigh backscattering in single-mode fiber to rapidly generate wavelength-dependent speckle patterns with MHz-level spectral correlation. We also introduce a dual-resolution scheme to mitigate the trade-off between resolution, bandwidth, and measurement rate. This optimized spectrometer design enables continuous, wideband (15 GHz) RF spectral analysis with MHz-level resolution and a fast update rate of 385 kHz. The entire system is constructed using fiber-coupled off-the-shelf-components, providing a powerful new approach for wideband RF detection and monitoring.