Reconstruction Capabilities Research Articles

Effect of Freeze–Thaw Cycles on Microstructure and Hydraulic Characteristics of Claystone: A Case Study of Slope Stability from Open-Pit Mines in Wet Regions

The action of freeze–thaw (F–T) cycles of claystone exerts a profound impact on the slope stability of open-pit mines in water-rich regions. Microstructural changes are observed as a crucial factor in determining the hydraulic characteristics and mechanical behaviors of claystone. The present work integrates a micro-X-ray computed tomography (Micro-CT) scanner, equipped with image processing and three-dimensional (3D) reconstruction capabilities, employed to observe the microstructure of claystone under varying numbers of F–T cycles (0, 10, 20, 30, and 50). Furthermore, seepage numerical simulations based on Micro-CT measurements are conducted to evaluate the hydraulic characteristics. Through meticulous microscopic observation and mechanical analysis, the damage mechanism induced by F–T cycles is revealed and the evolutionary characteristics are analyzed. The two-dimensional (2D) images of 3D reconstructed models unveil the gradual initiation propagation and coalescence of an intricate fissuring network in claystone during the F–T cycles. As the number of F–T cycles increases from 0 to 50, the 3D porosity exhibits exponential growth. Additionally, the influence of F–T cycles substantially enhances the connectivity of fissures. The seepage numerical simulations demonstrate that the evolutionary progression of fissures substantially augments the number of flow paths and enhances permeability. The increase in permeability follows an exponential trend, reflecting the distribution and evolution of fissures under F–T cycles. The impact on permeability arises from a combination of micromechanical properties and the microstructure of claystones. The present research tries to elucidate the microscopic evolution of fissures and their corresponding hydraulic properties in water-saturated claystone, offering significant insights for investigating the slope stability of open-pit mines in regions.

Infrared colorization with cross-modality zero-shot learning

Existing infrared colorization methods based on image-to-image translation in a GAN framework require paired image data for network learning and are limited by data quality. They are also restricted in their colorization ability to perform spectral translation between infrared bands, specifically, long-wave infrared colorization methods cannot be directly applied to the near-infrared modality. This paper proposes a novel infrared colorization algorithm that leverages similarities in high-frequency features between visible and infrared images. The method achieves infrared colorization through frequency domain feature decoupling and reconstruction. By decoupling high-frequency and low-frequency features in the frequency domain, it retains similar high-frequency features and reconstructs removed low-frequency features. The network is trained using visible images, enabling cross-modality zero-shot learning, eliminating the need for infrared datasets during training. During inference, the low-frequency information of the infrared image’s frequency domain is removed, and the visible low-frequency information is supplemented using the trained network’s reconstruction capability, resulting in the desired coloring effect. The method is equipped with multi-modality cross-spectral colorization capability and performs well on multiple infrared spectral colorization tasks. The experimental results fully demonstrate the excellent cross-modality adaptability and broad spectral colorization capability of the method.

Mechanism and Measurement of the Effects of Industrial Agglomeration on Agricultural Economic Resilience

This study investigates the potential of agricultural industrial agglomeration to bolster agricultural economic resilience and identifies the underlying pathways. We developed an analytical framework for agricultural economics that integrates the concept of “resilience”. This framework facilitates an examination of the influence of agricultural industrial agglomeration on agricultural economic resilience, focusing on two key aspects: enhancement of income and reduction of costs. Utilizing panel data from 30 provincial-level regions in China covering the period from 2006 to 2021, this research empirically assesses the impact, underlying mechanisms, and regional variations of agricultural industrial agglomeration on agricultural economic resilience. The findings reveal that agricultural industrial agglomeration significantly boosts agricultural economic resilience. This positive influence manifests through two primary channels: firstly, “agricultural industrial agglomeration → enhancement of socialized services → agricultural economic resilience” and secondly, “agricultural industrial agglomeration → improvement of agricultural production efficiency → agricultural economic resilience”. The contribution of agricultural industrial agglomeration to agricultural economic resilience is particularly pronounced in major grain-producing regions, notably enhancing capabilities for reconstruction and reinvention, as well as adjustment and adaptation. The study concludes with recommendations aimed at strengthening agricultural economic resilience. These recommendations emphasize the critical role of agricultural industrial agglomeration in fostering agricultural economic resilience, its contribution to the growth of rural economies and the enhancement of socialized services, and the need to consider regional disparities in the process of developing agricultural economic resilience.

An integrated group decision-making framework for evaluating the urban fire emergency management capability based on PDHLWPHM and PDHLWPGHM operator

Urban fire accident is a common dangerous accident in urban sudden accidents, which threatens the safety of people’s lives and property. For this reason, in recent years, all cities have incorporated the prevention and emergency management of urban fire accidents into their urban development planning, and actively improved their fire accident emergency management capabilities. However, how to evaluate the urban fire accident emergency management capacity of each city to ensure that people’s lives and property are protected to the greatest extent is an urgent problem to be considered and solved. Therefore, this paper defines a class of probabilistic double hierarchy linguistic Heronian mean (PDHLHM) operator, probabilistic double hierarchy linguistic Power Heronian mean (PDHLPHM) operators, and their dual operators that can reflect the relationship between two attributes during aggregation. Taking urban fire accident risk monitoring and early warning capability, fire infrastructure and communication system, fire-fighting and rescue capability, recovery and reconstruction capability as evaluation attributes, the probabilistic double hierarchy linguistic weight Power Heronian mean (PDHLWPHM) operator model and the probabilistic double hierarchy linguistic weight Power geometric Heronian mean (PDHLWPGHM) operator model are constructed for group decision-making. In addition, the idempotence, boundedness and monotonicity of these operators are studied, and the sensitivity of the parameters involved in the operator model is analyzed. Finally, the new model proposed in this paper is compared with the existing model to verify its scientificity.

An efficient impact force identification methodology via a single sensor utilizing the concept of generalized transmissibility

The issue of localizing and reconstructing the impact force through double-inverse approaches is widely recognized as an ill-posed and computationally demanding problem. Such a challenge has emerged as a prominent research topic of interest and great concern in the field of structural health monitoring. This study proposes a novel and efficient hierarchical methodology that is adept at ascertaining both the location and time history of an impact force through data obtained from a single accelerometer. The proposed methodology entails an efficient retrieval of uncorrected modal constants with a ratio function referred to as generalized transmissibility. The generalized transmissibility symbolizes the ratio of responses from identical accelerometer subjected to two distinct impacting scenarios, one excited at a yet-to-be-identified location and the other at a pre-determined reference location. The theoretical disclosure of the correlation between the generalized transmissibility at pole frequencies and retrieved uncorrected modal constants is demonstrated in details herein. These modal constants are then served as a signature to identify the impact location. Following localization, the force reconstruction problem is tackled by fitting a parametric model, wherein a gaussian basis function is employed to effectively approximate the time history of impact force. Experimental demonstrations on a metallic plate and a composite wing were carried out to validate the efficient and accurate localization as well as the rapid reconstruction capabilities in response to an impact force applied anywhere on a 2D structure.

Optimization of a Tsunami Gauge Configuration for Pseudo‐Super‐Resolution of Wave Height Distribution

AbstractIn this study, we present an optimization method for determining a cost‐effective sparse configuration for tsunami gauges to realize the reconstruction of high‐resolution wave height distribution throughout the target region based on the concept of super‐resolution. This optimization method consists of three procedures. First, we generate time series data of tsunami wave heights at synthetic gauges by conducting numerical simulations of various earthquake and tsunami scenarios at the target site. Next, we apply proper orthogonal decomposition to the synthetic tsunami data to extract the spatial features of the wave height distribution. Finally, according to these spatial features, an optimization process is performed to determine a sparse configuration of synthetic gauges. In the optimization, the optimal gauges are sequentially selected from the set of synthetic gauges to reconstruct the wave height distribution with the highest accuracy. Targeting hypothetical Nankai Trough earthquakes and tsunamis, we determine the optimal configuration near Shikoku and demonstrate the wave height reconstruction capability of the approach by comparing the performance of networks with optimally and randomly placed gauges. The results indicate that coastal gauges contribute more to improving the reconstruction accuracy and that a configuration with 21 optimal gauges has satisfactory performance. In addition, we assess the performance of the existing NOWPHAS network installed in the Shikoku region and find that the reconstruction performance of the existing network is equivalent to that of the optimal gauge network.

Advanced hierarchical planning and design for low-carbon landscape architecture leveraging virtual reality technology

Abstract The incorporation of sophisticated artificial intelligence technologies and data mining methodologies is critical for the planning and design of low-carbon landscape architectural environments. To enhance the three-dimensional reconstruction capabilities and visual esthetics of hierarchical planning and design within low-carbon landscape architectural environments, this study introduces a novel methodology predicated on virtual reality (VR) technology. This approach harnesses the diverse pictorial structure feature distribution inherent to low-carbon landscape architectural environments, amalgamated with VR's feature structure reorganization and enhancement techniques. Furthermore, we propose methodologies for joint measurement and multi-level integration to refine the hierarchical planning and design process within these environments. Simulation outcomes demonstrate the efficacy of this method in achieving superior three-dimensional reconstruction and visual representation, thereby augmenting the planning and design quality within the domain of low-carbon landscape architecture.

The Perception of Depth and Form in Fine Arts, Video Games and Medicine.

Medical imaging technology has greatly improved over the last 40 years. A good example of this is the improved 3D reconstruction capabilities of CTs, which enable photorealistic reconstructions. This technological advancement has also taken place in the computer industry, and the development of modern graphics cards has fueled progress in video games. In art, many of the techniques used to improve three-dimensionality have been known for centuries. At first sight, these fields seem unrelated, but there is actually considerable confluence. This paper focuses on three areas: fine arts, video games and radiology. The art section illustrates different drawing techniques used to create three-dimensionality. The section on video games shows the development of video games from the 1980s to the present day. Finally, the medical imaging section focuses on the first 3D radiography reconstructions in the 1980s and the improvements to the present day. Contemporary video games and 3D reconstructions of CTs enable photorealistic impressions of 3D anatomical structures. An important factor is shadow casting and light, which has been known in fine arts for centuries. The effect on the observer of photorealistic 3D reconstructions in video games and in medical imaging can be largely explained by the knowledge of 3D techniques in drawings and paintings.

Physics-informed deep learning for scattered full wavefield reconstruction from a sparse set of sensor data for impact diagnosis in structural health monitoring

This paper presents a physics-informed deep learning framework for the reconstruction of full scattered spatiotemporal Lamb wavefields (video images) in plate-like structures from a sparse set of time-series sensor data. The reconstructed scattered wavefield contains a wealth of information about the wave propagation phenomenon including any interactions of the propagating wave with damage in the structure. This information is paramount for damage diagnosis as is demonstrated in this paper via impact diagnosis—a key structural health monitoring application. A physics-informed neural network (PINN) that encodes the underlying elastodynamic field equations into the learning/training process in the neural network is proposed for this purpose. This prior wavefield physics knowledge embedded in the loss function acts as a regularization agent for the minimization problem in the neural network training, thereby enabling the extrapolation of a sparse set of one-dimensional time-series signals into two-dimensional scattered wavefield. The wavefield reconstruction capabilities of the proposed supervised forward PINN framework are first verified both numerically and experimentally for a stiffened aluminum panel under a couple of narrowband ultrasonic-frequency excitations, and the results confirm its robustness to low spatial resolution and substantial noise in the measured sensor data. The PINN requires far fewer sensors for scattered wavefield reconstruction, thereby permitting for a higher sensor spacing or lower spatial sampling. To this end, it is shown that a sensor spacing of 5λ generates good wavefield reconstruction accuracy, which is a 10-fold increase over the Nyquist–Shannon sampling limit (λ/2). Two sets of experiments are then conducted on a long-stiffened aluminum panel to validate the proposed framework via low-velocity impact diagnosis in the near-ultrasonic frequency range. The first set of experiments, with the known excitation force incorporated into the PINN, allows the wavefields to be accurately reconstructed with the sensor spacing up to 5λ as expected. The second set of experiments assumes unknown impact force history—a classical case of impact diagnosis where the impact force history is not known a priori. It is shown that the wavefield reconstruction through PINN still provides good accuracy albeit with a less generous sensor spacing of 2λ. A convolutional neural network long short-term memory (CNN-LSTM) model then solves the mathematically inverse problem of inferring the impact location and impact force history by analyzing the reconstructed impact generated wavefield. The impact location is predicted well with 93% accuracy, and the impact force history is reconstructed with 90% accuracy, further validating the proposed framework.

Full-field eigenstrain reconstruction for the investigation of residual stresses in finite length weldments

The reconstruction of welding residual stresses has been performed based on the assumption of continuous processing conditions, owing to the limitations of reconstruction methods. This assumption prevents the attainment of a comprehensive understanding of mechanical deformations in the entire weld zone. The voxel-based eigenstrain (inherent strain) reconstruction method emerges as an appropriate approach for the full-field reconstruction of residual stresses in finite length weldments without such assumption. The innovative reconstruction capability of this advanced method is presented in this study, aiming to comprehend all components of residual stresses and eigenstrains in finite length weldments with discontinuous processing properties using only a single component of experimental data belonging to a limited portion of the specimen. Numerical experiments were conducted to develop the tools. These tools were validated using a case study employing experimental data obtained through contact profilometry measurements from the surface of discharge machine cutting (minimally disturbing cutting) in a bead-on-plate weldment of Inconel alloy 740H designed for ultra-supercritical power generation. The reconstructed two components of normal residual stresses that do not align with the available experimental data were validated by independent neutron diffraction strain scanning quantifications. Furthermore, the reconstruction of residual stresses in the intact component prior to cutting was carried out using the eigenstrains associated with the post-cut state of the weldment.

EUSO-Offline: A comprehensive simulation and analysis framework

The complexity of modern cosmic ray observatories and the rich data sets they capture often require a sophisticated software framework to support the simulation of physical processes, detector response, as well as reconstruction and analysis of real and simulated data. Here we present the EUSO-Offline framework. The code base was originally developed by the Pierre Auger Collaboration, and portions of it have been adopted by other collaborations to suit their needs. We have extended this software to fulfill the requirements of Ultra-High Energy Cosmic Ray detectors and very high energy neutrino detectors developed for the Joint Exploratory Missions for an Extreme Universe Observatory (JEM-EUSO). These path-finder instruments constitute a program to chart the path to a future space-based mission like POEMMA. For completeness, we describe the overall structure of the framework developed by the Auger collaboration and continue with a description of the JEM-EUSO simulation and reconstruction capabilities. The framework is written predominantly in modern C++ (compliled against C++17) and incorporates third-party libraries chosen based on functionality and our best judgment regarding support and longevity. Modularity is a central notion in the framework design, a requirement for large collaborations in which many individuals contribute to a common code base and often want to compare different approaches to a given problem. For the same reason, the framework is designed to be highly configurable, which allows us to contend with a variety of JEM-EUSO missions and observation scenarios. We also discuss how we incorporate broad, industry-standard testing coverage which is necessary to ensure quality and maintainability of a relatively large code base, and the tools we employ to support a multitude of computing platforms and enable fast, reliable installation of external packages. Finally, we provide a few examples of simulation and reconstruction applications using EUSO-Offline.

First operation of an ACHINOS-equipped spherical proportional counter with individual anode read-out

The multi-anode sensor ACHINOS revolutionised the capabilities of the spherical proportional counter by enabling large-size, high-pressure, operation and TPC-like reconstruction capabilities through individual anode read-out. First measurements with an individually read out ACHINOS are performed, which enables improved calibration and response homogenisation. Experimental results demonstrating the improvement in energy resolution brought by the individual anode calibration are presented. These are complemented by detailed simulation studies on the effect of sensor design and manufacturing imperfections, and how they may be corrected both in hardware and analysis.

Constraining below-threshold radio source counts with machine learning

We propose a machine-learning-based technique to determine the number density of radio sources as a function of their flux density, for use in next-generation radio surveys. The method uses a convolutional neural network trained on simulations of the radio sky to predict the number of sources in several flux bins. To train the network, we adopt a supervised approach wherein we simulate training data stemming from a large domain of possible number count models going down to fluxes a factor of 100 below the threshold for source detection. We test the model reconstruction capabilities as well as benchmark the expected uncertainties in the model predictions, observing good performance for fluxes down to a factor of ten below the threshold. This work demonstrates that the capabilities of simple deep learning models for radio astronomy can be useful tools for future surveys.

Deep convolutional autoencoders for the time–space reconstruction of liquid rocket engine flames

The integration of high-fidelity numerical simulations into the rocket engine design-cycle promises to cut costs in an ever more competitive launch market. Although still being intractable in the optimization of most of combustion devices, their potential can be harnessed via appropriate reduced order models (ROMs). In recent years, significant progress has been made in the ROM literature through the adoption of non-linear methods, coming from the realm of deep learning. These methods promise to deliver where conventional linear techniques, such as Proper Orthogonal Decomposition (POD), fall short due to their inherent qualities.In the current work, we carry out a first attempt to develop purely data-driven emulators for a shear-coaxial liquid rocket engine injector. In particular, a Convolutional Neural Network autoencoder (CNN-AE) with a Multi-Layer Perceptron (MLP) latent dynamics encoder architecture is considered. The main objective is to assess the capacity of autoencoders based models to reconstruct, in time and space, the instantaneous temperature field of a non-canonical, complex, reactive, flow configuration. The data used is made of Large Eddy Simulations (LES) of three distinct, real-life inspired, injector configurations.The autoencoder based models show superior reconstruction capabilities compared to POD with similar latent-space dimension. Moreover, the time–space reconstructed fields preserve the spectral content of the snapshots, with only a small loss in gain at high frequencies. These results serve as proof-of-concept for further developments on surrogate models of liquid rocket engine injectors.

Future Prospects for Constraining Black Hole Spacetime: Horizon-scale Variability of Astrophysical Jets

The Event Horizon Telescope (EHT) Collaboration has recently published the first horizon-scale images of the supermassive black holes M87* and Sgr A* and provided some first information on the physical conditions in their vicinity. The comparison between the observations and the three-dimensional general relativistic magnetohydrodynamic (GRMHD) simulations has enabled the EHT to set initial constraints on the properties of these black hole spacetimes. However, accurately distinguishing the properties of the accretion flow from those of the spacetime, most notably, the black hole mass and spin, remains challenging because of the degeneracies the emitted radiation suffers when varying the properties of the plasma and those of the spacetime. The next-generation EHT (ngEHT) observations are expected to remove some of these degeneracies by exploring the complex interplay between the disk–jet dynamics, which represents one of the most promising tools for extracting information on the black hole spin. By using GRMHD simulations of magnetically arrested disks and general relativistic radiative transfer (GRRT) calculations of the emitted radiation, we have studied the properties of the jet and the accretion disk dynamics on spatial scales that are comparable with the horizon. In this way, we are able to highlight that the radial and azimuthal dynamics of the jet are well correlated with the black hole spin. Based on the resolution and image reconstruction capabilities of the ngEHT observations of M87*, we can assess the detectability and associated uncertainty of this correlation. Overall, our results serve to assess the prospects for constraining the black hole spin with future EHT observations.

BiVAE를 활용한 MBTI 기반 OTT 서비스 개인화 추천 시스템

The OTT platform, which provides various media contents in the modern society, seeks to improve the user experience by utilizing a personalized recommendation system. However, current recommendation systems are having difficulty in making accurate recommendations because they do not fully understand user's tendencies. This study proposes the use of a personalized recommendation algorithm using BiVAE (bilateral variational autoencoder) to solve this problem, and considers group-specific applications through MBTI. BiVAE is a probabilistic generation model that compresses and generates data considering of bidirectional information flow, providing richer expression and reconstruction capabilities by encoding and decoding in both directions. As a result of applying four types of data to the model, when applying BiVAE for each group using MBTI, the accuracy of OTT genre recommendation was shown to be superior to that of other methods. In addition, when considering MBTI, a personal recommendation using BiVAE showed higher accuracy compared to the existing recommendation system. This result means that the personalized recommendation system considering the user's personal characteristics and personality is important.

ACVAE: A novel self-adversarial variational auto-encoder combined with contrast learning for time series anomaly detection

Deep generative models have advantages in modeling complex time series and are widely used in anomaly detection. Nevertheless, the existing deep generative approaches mainly concentrate on the investigation of models’ reconstruction capability rather than customizing a model suitable for anomaly detection. Meanwhile, VAE-based models suffer from posterior collapse, which can lead to a series of undesirable consequences, such as high false positive rate etc. Based on these considerations, in this paper, we propose a novel self-adversarial variational auto-encoder combined with contrast learning, short for ACVAE, to address these challenges. ACVAE consist of three parts 〈T,E,G〉, wherein the transformation network T is employed to generate abnormal latent representations similar to those normal latent representations encoded by the encoder E, and the decoder G is used to distinguish the two representations. In the framework of this model, the normal reconstructions are considered as positive samples and abnormal reconstructions as negative samples, and the contrast learning is executed on the part E to measure the similarities between inputs and positive samples, dissimilarities between inputs and negative samples. Thus, an improved objective function is proposed by integrating two novel regularizers, one refers to adversarial mechanism and the other involves contrast learning, in which the encoder E and decoder G hold the capability to distinguish, and decoder G is constrained to mitigate the posterior collapse. We perform several experiments on five datasets, whose results show ACVAE outperforms state-of-the-art methods.

개인중심접근 기반 개별화교육계획 수립을 위한 2022 개정 특수교육 기본교육과정 성취기준 분석

Objectives This study is a study that analyzes the achievement standards of all subjects of the 2022 revised spe-cial education curriculum based on the curriculum area of the COACH program to establish an individualized edu-cation plan through an person-centered approach.
 Methods To this end, keyword analysis was conducted based on the curriculum area of the COACH program, and the analysis subjects were selected from all grades from elementary to high school, including integrated curricu-lum (right life, wise life, fun life), Korean, math, science, practical course, career and job, physical, music, art, and information and communication, health.
 Results As a result of the analysis, first, when looking at the analysis results of each curriculum area of the COACH program, it was analyzed that academic preparation skills had the most achievement standards and jobs had the least achievement standards. Second, looking at the results of the analysis by school level, all elementary, middle, and high school levels included the highest achievement standards for academic preparation skills. On the other hand, elementary and middle school levels had the least job-related achievement standards, and high school lev-els had the least community-related achievement standards. Third, according to the results of the analysis of subjects by curriculum area, the communication area was the highest, and the social area was analyzed as society, personal management as science, leisure as science, academic preparation skills as science, functional learning skills as math, community as well as career.
 Conclusions Based on the results of this study, the implications for establishing an individualized education plan based on an person-centered approach and follow-up research are as follows. First, in order to design an in-dividualized education plan based on an person-centered approach, special teachers' high curriculum literacy and curriculum reconstruction capabilities are required. Second, in order to design an individualized education plan based on an person-centered approach, it is necessary to establish an individualized education plan so that the learned content can be linked to real life.

A full-scale hierarchical encoder-decoder network with cascading edge-prior for infrared and visible image fusion

Most existing deep learning-based infrared and visible image fusion methods always fail to consider the full-scale long-range correlation and the prior knowledge, resulting in the fused images with low-contrast salient objects and blurred edge details. To overcome these drawbacks, a full-scale hierarchical encoder-decoder network with cascading edge-prior for infrared and visible image fusion is proposed. First, a top-down encoder extracts the hierarchical representations from source image. Then, to inject edge priors into the network and capture the progressive semantic correlations, a triple fusion mechanism is proposed including edge image fusion based on maximum fusion strategy, single-scale shallow layer fusion and full-scale semantic layer fusion based on dual-attention fusion (DAF) strategy. The fused full-scale semantic features (F2SF) are obtained by capturing the long-range affinities of the full-scale. At the same time, a cascading edge-prior branch (CEPB) is designed to embed the fused edge knowledge into fused single-scale shallow features, jointly guiding the decoder to focus on abundant details layer-by-layer on the basis of F2SF, thus recovering the edge and texture details of the fused image well. Finally, a novel loss function consisting of SSIM, intensity and edge loss is constructed to further maintain the network with better edge representation and reconstruction capability. Compared with existing state-of-the-art fusion methods, the proposed method has better performance in terms of both visual evaluation and objective evaluation on public datasets. The source code is available at https://github.com/lxq-jnu/FSFusion.

Error correction algorithm of array time-varying amplitude and phase based on autoencoder

As array antennas are widely used in various mobile platforms, the time-varying amplitude and phase error has become an important factor affecting the application of array signal processing technology. A deep learning-based algorithm for the correction of time-varying amplitude and phase errors in arrays is proposed in terms of the idea of autoencoder. The algorithm makes full use of the data feature extraction and reconstruction capability of the autoencoder network, designs a deep learning network for the correction of time-varying amplitude phase error of the channel, gives a double-driven learning mechanism without time-varying amplitude phase error data (unperturbed data) and time-varying amplitude phase error data (perturbed data), completes the extraction of the array stream shape hidden features based on the principle of minimising the mean square error of the desired output and the ideal model. The simulated experiments show that the algorithm can effectively correct the time-varying amplitude and phase errors of each channel, and the mean square error of the corrected amplitude and phase errors are within 0.5% and 1.5% respectively when there are ±80% random time-varying amplitude errors and ±5° random time-varying phase errors. The effectiveness of the proposed algorithm is verified.