Artificial neural networks (ANNs) have led to landmark changes in many fields, but they still differ significantly from the mechanisms of real biological neural networks and face problems such as high computing costs, excessive computing power, and so on. Spiking neural networks (SNNs) provide a new approach combined with brain-like science to improve the computational energy efficiency, computational architecture, and biological credibility of current deep learning applications. In the early stage of development, its poor performance hindered the application of SNNs in real-world scenarios. In recent years, SNNs have made great progress in computational performance and practicability compared with the earlier research results, and are continuously producing significant results. Although there are already many pieces of literature on SNNs, there is still a lack of comprehensive review on SNNs from the perspective of improving performance and practicality as well as incorporating the latest research results. Starting from this issue, this paper elaborates on SNNs along the complete usage process of SNNs including network construction, data processing, model training, development, and deployment, aiming to provide more comprehensive and practical guidance to promote the development of SNNs. Therefore, the connotation and development status of SNN computing is reviewed systematically and comprehensively from four aspects: composition structure, data set, learning algorithm, software/hardware development platform. Then the development characteristics of SNNs in intelligent computing are summarized, the current challenges of SNNs are discussed and the future development directions are also prospected. Our research shows that in the fields of machine learning and intelligent computing, SNNs have comparable network scale and performance to ANNs and the ability to challenge large datasets and a variety of tasks. The advantages of SNNs over ANNs in terms of energy efficiency and spatial-temporal data processing have been more fully exploited. And the development of programming and deployment tools has lowered the threshold for the use of SNNs. SNNs show a broad development prospect for brain-like computing.

Optical cavity has long been critical for a variety of applications ranging from precise measurement to spectral analysis. A number of theories and methods have been successful in describing the optical response of a stratified optical cavity, while the inverse problem, especially the inverse design of a displacement sensitive cavity, remains a significant challenge due to the cost of computation and comprehensive performance requirements. This paper reports a novel inverse design methodology combining the characteristic matrix method, mixed-discrete variables optimization algorithm, and Monte Carlo method-based tolerance analysis. The material characteristics are indexed to enable the mixed-discrete variables optimization, which yields considerable speed and efficiency improvements. This method allows arbitrary response adjustment with technical feasibility and gives a glimpse into the analytical characterization of the optical response. Two entirely different light-displacement responses, including an asymmetric sawtooth-like response and a highly symmetric response, are dug out and experimentally achieved, which fully confirms the validity of the method. The compact Fabry-Perot cavities have a good balance between performance and feasibility, making them promising candidates for displacement transducers. More importantly, the proposed inverse design paves the way for a universal design of optical cavities, or even nanophotonic devices.

Epilepsy is a neurological disorder characterized by abnormal neuronal discharges in the brain. As a rich source of biometric information, electroencephalography (EEG) provides favorable conditions for automated detection. Traditional algorithms and manual analysis possess solid theoretical foundations and good interpretability, however, these methods predominantly require extensive domain expertise and involve lengthy processing pipelines for complex data. The advent of artificial intelligence (AI) has facilitated the application of neural networks in the detection and prediction of epilepsy. Although such approaches heavily rely on high-quality annotated data, suffer from limited model interpretability, and involve complex training and parameter tuning, these efficient, real-time, end-to-end models still demonstrate significant potential in epilepsy analysis. This review systematically analyzes and summarizes the neural network technologies used in 341 papers published in the past three years, employing the PRISMA standard procedure. To facilitate readers’ related research, the review also summarizes the basic information of 16 publicly available datasets, common features, and metrics. Specifically, this review offers a comprehensive evaluation of diverse neural network architectures, concluding that convolutional neural networks have become a prevalent choice as classic neural networks. Furthermore, graph neural networks and transformers are experiencing a marked surge in popularity. The application of hybrid neural networks to fully extract information from EEG is also a growing trend. The review concludes with a comprehensive discussion and summary of the technical characteristics, research directions, and limitations of current methods, including patient-to-patient identification, explainable AI, dataset bias, and zone location.

Mismatches affect the dynamic performance of time-interleaved analog-to-digital converters (TIADCs). Linear mismatches can be calibrated by many mature methods, but if higher performance is required, nonlinearity mismatches have to be suppressed. The background calibration method based on input-free band (IFB) functions poorly for narrow-band signals. This brief proposes a correlation-based calibration method for nonlinearity mismatches in dual-channel TIADCs which behaves well for both wide-band and narrow-band signals. The output samples are calibrated by reducing the residual distortions which are approximated by multiplying the pseudo distortions and the estimated mismatch coefficients. The pseudo distortions are acquired by using a frequency-shifter, a differentiator, and multipliers. The coefficients which indicate the mismatch strength are estimated by eliminating the cross-correlation of the calibrated output samples and the calibrated pseudo distortions at zero lag. Simulations show that the proposed method can improve the SFDR by dozens of dBc for narrow-band input signals, compared with the IFB method. For the 16-QAM signal, the error vector magnitude improvement over the IFB method is 35.48 dB.

Memristor can store calculation results while computing, and utilizing memristor to realize in-memory computing is considered as one of the potential methods to break the bottleneck of Von Neumann architecture. To construct high performance computing systems, it is necessary to design memristor-based computing units with great advantages in delay, area, and integratability. This brief presents a high speed memristor-based $N$ -bit ripple carry adder (RCA), with special design to facilitate the implementation with 1T1R arrays. After carries are calculated and aligned efficiently, we adopt Memristor-Aided LoGIC (MAGIC) to calculate and store the outputs of RCA in parallel, where The proposed $N$ -bit RCA only consists of $5N+1$ memristors, $8N+1$ MOSFETs, and its delay is $N+4$ . Compared with the existing memristive RCA, the proposed RCA has great advantage in delay, while it is easy for integration in 1T1R array.