Enhancing Signal Recognition Accuracy in Delay-Based Optical Reservoir Computing: A Comparative Analysis of Training Algorithms

Aiming at the problem that the ranging accuracy of the frequency-modulated continuous wave (FMCW) lidar decreases in the strong noise environment, experiments are carried out to verify the feasibility of the Faraday anomalous dispersion filter (FADOF) in the FMCW lidar system. The transmitter of the system uses the potassium atom vapor cell to stabilize the frequency and uses the field programmable gate array (FPGA) as the servo system. K-FADOF is used to suppress the additive noise at the receiving end of the lidar system. The results of this experiment show that the use of FADOF can successfully eliminate the interference of noise at the receiver. This experiment verifies the effectiveness of FADOF in suppressing noise in the field of FMCW lidar.

Frequency-modulated continuous-wave (FMCW) LIDAR is a promising technology for next-generation integrated 3D imaging systems. However, it has been considered difficult to apply FMCW LIDAR for long-distance (> 100m) targets, such as those in automotive and airborne applications. Maintaining coherence between the reflected beam from the target and locally forwarded beam becomes a significant challenge for tunable laser design. This paper demonstrates the possibility of extending the detection range of FMCW LIDAR beyond the coherence range of its laser by improving the spectral estimation algorithm. By exploiting the Lorentzian prior of the received signal in the spectral domain, > 10x improvement in ranging accuracy is achieved compared to traditional algorithms that do not consider phase noise in the signal model. In light of this finding, the end-to-end modeling framework is presented to examine true system-level trade-offs of FMCW LIDAR and the feasibility of long -distance measurement.

In industrial sites, large-scale advanced manufacturing equipment such as optical diamond cutting lathes require stable operation during processing and are very sensitive to operating speed. However, the vibration generated by the environment or equipment will interfere with its operation and machining accuracy. A precise laser interference measurement system is required to monitor the running speed and position of the three-axis tool of the lathe in real time, providing accurate parameters for the closed-loop control system of the equipment, and improving the stability of lathe operation. Frequency-modulated continuous-wave (FMCW) lidar is widely used in the field of industrial measurement due to its non-contact, high accuracy, and fast dynamic response. The basic principle of FMCW lidars is to measure the velocity of a moving object through the Doppler frequency shift phenomenon. But the vibration generated by the moving object will cause the spectrum to broaden and the precision and repeatability of measurement to decrease. Therefore, we propose a large-bandwidth triangular wave-modulated lidar structure with Fabry-Pérot(F-P) cavity to achieve real-time high-precision measurement of the speed and distance of moving targets. This structure is based on the F–P resonance peaks generated by changing the length of the F-P cavity to accurately split the beat signal generated in the sweep frequency range of 1545-1565nm to obtain 40 sets of data with equal frequency intervals of 62GHz, effectively solve the problem of excessive data volume when measuring the continuous moving target speed and reduces the complexity of the algorithm. Splitting the measured beat signal based on the resonance peaks signal of the F-P cavity, which reduces the phase delay of the beat signal corresponding to the up- and down-scanning, thus reducing the signal spectrum broadening caused by frequency deviation and nonlinear, and increasing the target measurement resolution, accuracy and range. The experimental results show that for speeds of up to 250mm/s, the mean standard deviation was less than 152μm/s, the mean error was less than 183μm/s, the relative error of the mean value of speed measurement does not exceed 0.22%, and the standard deviation of the distance measurement results within a range of 4m under various speed conditions does not exceed 17μm, the error does not exceed 14μm, it has good accuracy and repeatability for the speed and distance measurement. The lidar architecture solution we proposed has important application value for large-scale industrial equipment measurement and operation monitoring.

We demonstrate a closed-loop Electro-Optical Phase-Locked Loop (EO-PLL) designed to increase the linearity of frequency chirps in Frequency-Modulated Continuous-Wave (FMCW) LiDAR systems, resulting in improved metrological accuracy. The system is fully implemented on an easily accessible FPGA-based digital electronic platform (Red Pitaya STEMlab 125-14). A PC-based UI was developed to facilitate remote real-time control over key system parameters. The proposed closed-loop control reduces the Full-Width Half Maximum (FWHM) in frequency domain to 11 kHz, corresponding to a range resolution of 19.9 mm at a beat frequency of 2.45 MHz. Additionally, the system effectively suppresses disturbances in the modulation signal for frequencies up to 150 kHz, with stabilization settling within a single modulation period. This approach enables the use of lasers with nonlinear modulation slopes in FMCW LiDAR applications, even under fluctuating ambient conditions.

This work presents a novel scheme for a 4-dimensional (D) frequency modulated continuous wave (FMCW) LiDAR, which demonstrates an unprecedentedly high velocity sensitivity. This is achieved by utilizing the driving waveform to minimize the phase noise originating from the distributed feedback (DFB) laser during the wavelength sweeping process on the transmitting side, while on the receiving end, we use a combination of a self-injection-locked oscillator (SILO) and an avalanche photodiode (APD) with a cascaded multiplication layer, to ensure excellent responsivity and saturation current performance. Comparison is made with a reference LiDAR system comprised of the traditional radio-frequency (RF) oscillator with a p-i-n PD at the receiving-end. The results show that our SILO + APD design provides much better quality 4-D images of slow-moving objects (5 μm/sec) and state-of-the-art velocity sensitivity. This can be attributed to a reduction in the phase noise of the down-converted baseband signals, which is beneficial to the resolution of a small Doppler frequency shift. Furthermore, the incorporation of a high-performance APD also improves the contrast ratio between the object and the background pixels. These improvements offer new possibilities for the development of the next generation of 4-D FMCW LiDARs capable of simultaneously providing the absolute 3-D geometric size and small vibration information.

<div class="section abstract"><div class="htmlview paragraph">In pursuit of safety validation of automated driving functions, efforts are being made to accompany real world test drives by test drives in virtual environments. To be able to transfer highly automated driving functions into a simulation, models of the vehicle’s perception sensors such as lidar, radar and camera are required. In addition to the classic pulsed time-of-flight (ToF) lidars, the growing availability of commercial frequency modulated continuous wave (FMCW) lidars sparks interest in the field of environment perception. This is due to advanced capabilities such as directly measuring the target’s relative radial velocity based on the Doppler effect. In this work, an FMCW lidar sensor simulation model is introduced, which is divided into the components of signal propagation and signal processing. The signal propagation is modeled by a ray tracing approach simulating the interaction of light waves with the environment. For this purpose, an ASAM Open Simulation Interface (OSI) object list referencing virtual 3D objects provides the input for the ray tracer. The divergence of the continuous laser beam is approximated by super-sampling the beam with multiple rays, the calculation of the received power is supported by the future ASAM OpenMATERIAL standard. Subsequently, the output of the ray tracer serves as the input of the signal processing that adapts the so-called Fourier tracing from the field of radar sensor simulation. This approach uses the range and velocity information of the individual rays to estimate the frequency spectrum of the intermediate frequency signal. A subsequent peak detection algorithm determines the output of the model, which is provided in the form of OSI lidar detections. Verification scenarios are tested to check the plausibility of the output and the source code of the signal processing is made available as open source.</div></div>

Physical reservoir computing (RC) systems have emerged as a prominent research frontier due to their exceptional efficiency in temporal information processing. However, existing implementations, predominantly utilizing resistive devices, face challenges pertaining to power efficiency and dynamic richness. Here, we propose a ferroelectric capacitor-linear capacitor (FC-LC) series device for RC implementation. By leveraging nonlinear polarization switching and back-switching, the FC-LC series device realizes two essential reservoir properties: nonlinearity and fading memory. In addition, the device exhibits an ultralow power consumption, which, along with its direct voltage readout capability, marks a significant advance over resistive reservoir devices. Moreover, the device features bidirectional operation and widely tunable time constants, thereby enhancing reservoir space dimensionality and state richness. Building upon these FC-LC series devices, a ferroelectric capacitive RC system is developed, which demonstrates superior performance in various benchmark tasks. By exploiting the bidirectional operation of the device, the RC system not only delivers enhanced performance in waveform classification but also enables high-accuracy multimodal digit recognition. Through strategically hybridizing the FC-LC series devices with varying time constants, the RC system achieves remarkable performance in Mackey-Glass time-series prediction. Our study paves the way for power-efficient, dynamic-rich RC systems capable of handling diverse temporal tasks.

In‐sensor computing hardware with signal sensing and dynamic signal processing capabilities—inspired by the human sensory system—have attracted interest as the rapid proliferation of data sensing and computation in the era of big data. Here, an in‐sensor reservoir computing (RC) system with integrated functions of signal sensing, preprocessing, and RC in a single device is developed, by constructing a full quantum dot optoelectronic memristor (FQDOM) based on a ZnO QDs/CdSe QDs/ZnO QDs heterojunction. The device shows nonlinear and short‐term memory behavior in response to both electrical stimuli and optical signals covering the spectra from ultraviolet to red light. The FQDOM can achieve color perception. The in‐sensor RC system can sense varied types of input signal, exhibiting multisensory fusion capability by fashion‐MNIST classification. More importantly, these proposed FQDOMs can preprocess signal and then send for RC in sensor, which effectively reduces the bit error rate of RC systems. The processing capability of the in‐sensor RC system for temporal signals is further demonstrated by a dynamic gesture perception recognition task, showing an accuracy of up to 92.59%. This research provides an effective way to design advanced in‐sensor computational systems.

Frequency-modulated continuous-wave (FMCW) lidar offers high precision and strong interference resistance, capable of synchronously measuring target motion speed and vibration information. However, extracting target information using single-photon signal levels poses a critical challenge in extreme conditions. In this study, we propose a single-photon level FMCW lidar scheme. Quantum compressed sensing (QCS) is employed to simultaneously extract target distance and vibration information. Experimental results demonstrate successful synchronous detection at a beat frequency of 27.304 kHz and a vibration frequency of 500 Hz within an integration time of 0.125 s at a photon counting rate of 9 kcps. This approach provides a new, to the best of our knowledge, solution for FMCW radar application in extreme environments or long-range scenarios.

Analysis and fully memristor-based reservoir computing for temporal data classification

Simultaneously sensing and recognizing motion information and time‐varying data are crucial in machine vision applications. Reservoir computing (RC) systems provide a promising framework for processing spatiotemporal signals efficiently. However, a bottleneck of fixed spatiotemporal scales restricts the dynamic application scenarios of RC systems. Here we demonstrate a photosensor based on silicon telluride (SiTe) that has gate‐tunable Schottky barrier under optical stimuli. The photocurrent and relaxation time were customized at various levels by tuning gate voltage without changing light intensity. The programmable response spatialscale and relaxation timescale are attributed to the gate modulation on Schottky barrier height and trap filling level, respectively. Controllable dynamics and time‐dependent response are leveraged to construct a multiscale optoelectronic RC system, delivering efficient and robust perception of the car motion mode and achieving a recognition accuracy of 90%. This work provides an insight of material and device design into RC system for real‐time motion recognition.

Frequency tuning nonlinearities in semiconductor lasers constitute a critical factor that degrades measurement precision and spectral resolution in frequency-modulated continuous-wave (FMCW) LiDAR systems. This study systematically investigates the influence of nonlinear beat signal phase distortions on spectral peak broadening and develops a phase-fitting-based pre-correction algorithm. To further enhance system performance, an electro-optic phase-locked loop architecture combined with non-uniform discrete Fourier transform signal processing is implemented, establishing a comprehensive solution for tuning nonlinearity suppression. Experimental validation demonstrates a sub-18 µm standard deviation in absolute distance measurements at a 19 m target range. This integrated approach represents a significant advancement in coherent frequency-sweep detection methodologies, offering considerable potential for high-precision photonic radar applications.

Frequency-modulated continuous-wave (FMCW) LiDAR systems are drawing increasing interest due to their potential applications in autonomous driving, machine perception, rapid prototyping, and medical diagnostics. The nonlinearity of a laser’s input-output transfer function can degrade the performance of an FMCW LiDAR. However, traditional discrete-time electro-optical phase-locked loops (DT-EOPLLs) face an unfavorable trade-off between chirp bandwidth and Mach-Zehnder delay. We present an integrated continuous-time electro-optic phase-locked loop (CT-EOPLL) to address this problem. The proposed EOPLL is very wideband, with its loop bandwidth equal to its reference frequency. This feature enables it to relax the trade-off between chirp bandwidth and Mach-Zehnder (MZ) delay by 10× in dB scale, which consequently reduces the area and loss associated with the silicon photonic delay implementation. It also does not suffer from the challenging issue of spurs in wideband PLLs because it features image and harmonic spur suppression in the loop using single-sideband (SSB) and harmonic-reject (HR) mixing techniques. The electrical part of this EOPLL is implemented in 65nm CMOS technology, and its optical integrated circuit is fabricated using a silicon photonic process. Featuring more than 25dB of suppression of the highest spur, this EOPLL is utilized in a high precision LiDAR sensor that shows an RMS depth precision of 558μm at 2m distance, and a 9.4mm RMS depth resolution at ranges exceeding 3.3m.

Fusion of diverse data sources for flood extent mapping and risk assessment in Sindh: A comparative study of inundation mapping approaches

In the course of our research work, the American, Russian and Turkish sign languages were analyzed. The program of recognition of the Kazakh dactylic sign language with the use of machine learning methods is implemented. A dataset of 5000 images was formed for each gesture, gesture recognition algorithms were applied, such as Random Forest, Support Vector Machine, Extreme Gradient Boosting, while two data types were combined into one database, which caused a change in the architecture of the system as a whole. The quality of the algorithms was also evaluated. The research work was carried out due to the fact that scientific work in the field of developing a system for recognizing the Kazakh language of sign dactyls is currently insufficient for a complete representation of the language. There are specific letters in the Kazakh language, because of the peculiarities of the spelling of the language, problems arise when developing recognition systems for the Kazakh sign language. The results of the work showed that the Support Vector Machine and Extreme Gradient Boosting algorithms are superior in real-time performance, but the Random Forest algorithm has high recognition accuracy. As a result, the accuracy of the classification algorithms was 98.86 % for Random Forest, 98.68 % for Support Vector Machine and 98.54 % for Extreme Gradient Boosting. Also, the evaluation of the quality of the work of classical algorithms has high indicators. The practical significance of this work lies in the fact that scientific research in the field of gesture recognition with the updated alphabet of the Kazakh language has not yet been conducted and the results of this work can be used by other researchers to conduct further research related to the recognition of the Kazakh dactyl sign language, as well as by researchers, engaged in the development of the international sign language