The analysis of infrared spectroscopy of substances is a non-invasive measurement technique that can be used in analytics. Although the main objective of this study is to provide a review of machine learning (ML) algorithms that have been reported for analyzing near-infrared (NIR) spectroscopy from traditional machine learning methods to deep network architectures, we also provide different NIR measurement modes, instruments, signal preprocessing methods, etc. Firstly, four different measurement modes available in NIR are reviewed, different types of NIR instruments are compared, and a summary of NIR data analysis methods is provided. Secondly, the public NIR spectroscopy datasets are briefly discussed, with links provided. Thirdly, the widely used data preprocessing and feature selection algorithms that have been reported for NIR spectroscopy are presented. Then, the majority of the traditional machine learning methods and deep network architectures that are commonly employed are covered. Finally, we conclude that developing the integration of a variety of machine learning algorithms in an efficient and lightweight manner is a significant future research direction.

Transformers were initially introduced for natural language processing, leveraging the self-attention mechanism. They require minimal inductive biases in their design and can function effectively as set-based architectures. Additionally, transformers excel at capturing long-range dependencies and enabling parallel processing, which allows them to outperform traditional models, such as long short-term memory (LSTM) networks, on sequence-based tasks. In recent years, transformers have been widely adopted in computer vision, driving remarkable advancements in the field. Previous surveys have provided overviews of transformer applications across various computer vision tasks, such as object detection, activity recognition, and image enhancement. In this survey, we focus specifically on image classification. We begin with an introduction to the fundamental concepts of transformers and highlight the first successful Vision Transformer (ViT). Building on the ViT, we review subsequent improvements and optimizations introduced for image classification tasks. We then compare the strengths and limitations of these transformer-based models against classic convolutional neural networks (CNNs) through experiments. Finally, we explore key challenges and potential future directions for image classification transformers.

A novel compact multiple-input multiple-output (MIMO) antenna is proposed with wideband gain enhancement. It consists of two identical antenna elements spacing about half wavelength. Firstly, a 1-to-2 Y-shaped power divider with wideband anti-phase outputs is designed to feed two modified Vivaldi radiators. The currents on the two radiating surfaces keep the same direction in a wide band to increase radiation without being cancelled. Secondly, the square-ring unit cell is analyzed to form metasurface, placed in the front of the radiator. It can guide the forward electromagnetic radiation while reducing the backward radiation. Thirdly, a U-shaped slot is etched between the two modified Vivaldi radiators. It is approximated as an effective electrically small dipole radiator. Such arrangement further effectively enhances the radiation performance in the direction of the main lobe and destructively interferes with the side lobe radiation. These features make the antenna far-field radiation pattern re-shaped. Thereby, the gain is improved in a wide bandwidth. Meanwhile, the MIMO antenna diversity performance is analyzed with low mutual coupling. Wideband gain enhancement contributes to improving the detection capability of the frequency modulated continuous wave (FMCW) radar sensor. As a proof of concept, the proposed MIMO antenna prototype is fabricated. The measured impedance bandwidth ranges from 11.5 to 21.3 GHz (59.76% FBW) with isolation of ≥ 24.92 dB. The gain is up to 10.6 dBi, the radiation efficiency is 88.01-90.02%, and the envelope correlation coefficient (ECC) is ≤ 0.00122. Integrated with the transceiver chip fabricated on the 65-nm CMOS process, the proposed MIMO antenna is applied in the FMCW radar sensor. The test system is built up, and the experiments on different breathing conditions are conducted for different human subjects. The proposed MIMO antenna could reduce measurement error, thereby improving measurement accuracy for human respiratory monitoring.

A bionic artificial device commonly integrates various distributed functional units to mimic the functions of biological sensory neural system, bringing intricate interconnections, complicated structure, and interference in signal transmission. Here we show an all-in-one bionic artificial nerve based on a separate electrical double-layers structure that integrates the functions of perception, recognition, and transmission. The bionic artificial nerve features flexibility, rapid response (10,000 tests), personalized cutability, and no energy consumption when no mechanical stimulation is being applied. The response signals are highly regionally differentiated for the mechanical stimulations, which enables the bionic artificial nerve to mimic the spatiotemporally dynamic logic of a biological neural network. Multifunctional touch interactions demonstrate the enormous potential of the bionic artificial nerve for human-machine hybrid perceptual enhancement. By incorporating the spatiotemporal resolution function and algorithmic analysis, we hope that bionic artificial nerves will promote further development of sophisticated neuroprosthetics and intelligent robotics.

A compact dual-polarized multiple-input-multiple- output (MIMO) Vivaldi antenna with a wide bandwidth, high gain, and high isolation is proposed for wireless communications in intelligent Internet of Vehicles (IoV). It consists of four identical orthogonally-crossed Vivaldi radiators, crossed-fishbone-shaped slot, metasurface lens, parasitic stubs, electromagnetic (EM) wave reflector, and three-dimensional (3D) printed radome with dielectric resonator (DR) array. These functional modules jointly contribute to optimized antenna performance. Four identical antenna elements are rotationally symmetrical. To reduce the inter-element mutual coupling, the crossed-fishbone-shaped slot is etched to reduce the surface current flowing in adjacent radiators. To raise the gain, the metasurface lens, parasitic stubs, and DR array are loaded to guide the EM wave in the frontward radiation. An EM wave reflector located underneath the Vivaldi radiators could increase the boresight gain and reduce the influence of the vehicle body on the antenna performance. Diversity performance analysis indicates the proposed antenna could offer better MIMO capability. Besides, the impact of weather conditions on the antenna is analyzed. As a proof of concept, an expanded 8-element antenna array prototype is designed, fabricated, and tested. A good agreement is observed between measured and simulated results. It achieves an -10-dB impedance bandwidth of 7.55-22.85 GHz (100.66% FBW) with port isolation of higher than 18.05 dB. The realized gain is 6.98-11.54 dBi, and the radiation efficiency is 85.05-90.27%. The proposed antenna could be used in wireless communications between vehicles and the satellite/infrastructure.