积分方程的多重网格解法及其在天线分析中的应用

本文将多重网格方法应用于积分方程，建立了Ｆｒｅｄｈｏｌｍ积分方程的多重网格解法，并将其应用于线天线的分析研究，对不同长度的线天线进行了数值分析，得到了较好的结果。数值实验表明，多重网格方法是一种十分有效的快速迭代方法，为分析大阵问题提供了一种新的数值方法。     

换能器阵列的传输线矩阵方法分析

介绍将传输线矩阵方法应用于声光换能器阵列的辐射特性分析，给出了与解析方法一致的数值结果。表明传输线矩阵方法可以用于对声学器件的工作过程进行时域数值模拟。     

电磁问题分析中的一种新型数值方法

提出了一种用于分析电磁问题的新型的数值方法，采用这种方法可以在降维了的空间里获得精度层次不同的解，从而大大地节省了分析电磁问题时的计算时间和内存要求，采用这个方法分析了几种线天线，并与用常规矩量法算出的结果或与相关文献给出的结果进行了比较，它们之间良好的一致性说明了本方法的正确性和有效性。     

发射极开关晶闸管正向特性分析

本文在对发射极开关晶闸管ＥＳＴ正向特性二维数值分析的基础上，提出了ＥＳＴ正向工作时的五个区域，从剖面图及等效电路图出发，分区、分阶段对其正向特性进行了解析分析，最后将实验结果、解析分析结果及数值分析结果进行比较，发现数值分析结果和解析分析结果吻合较好，同时说明了解析结果和实验结果之间存在差别的原因。     

并行化小波基FDTD方法的电磁场计算效率分析

为研究在大尺寸的光子器件中电磁场的传播特性,建立了一种基于小波基的时城有限差分(FDTD)数值计算方法,并在计算机集群系统中实现其并行化运算.该方法相较于标准的FDTD方法,具有更好的数值色散特性,大大减少了计算空间网格数,从而减少了内存的使用;且并行化计算大大减少了运算时间.验证了该方法的有效性,分析了影响并行运算效率的主要因素,并建立了一个估算模拟运算时间的数学模型.     

微带天线巴仑的准静态分析

提出了一种分析微带天线巴仑的数值方法。该方法具有计算简便，物理意义清晰，且不受巴仑的几何形状限制等优点，对实例计算结果和测量结果的比较表明，该方法是可行和有效的，并可加以推广，用来分析其它类型的多导体非均匀传输线。     

半导体器件瞬态模型渐近分析

本文采用奇异摄动方法分析了半导体器件瞬态模型解的渐近性态，给出了构造近似解的方法和步骤，获得了具有二阶修正精度的解的近似表达式．该近似结果与数值结果相吻合．     

高压JTE终端边界元电场分析

在半导体高压终端研究中，重要的工作之一是终端电场电位分析。国内外以往报道的JTE分析结果除采用解析方法外，数值方法主要采用差分或有限元法。作者利用边界元数值技术，采用“边界元临界电场分析法”，用自主开发的统一的边界元终端分析软件，从新的角度，以单区、双区JTE为例，详细讨论了新的边界元分析法，研究了极值电场分布等情况，得到了物理概念清晰，描述较准确而又简明直观的优化结果。边界元算法独特，优点显著。现有程序不但可用于JTE终端优化分析设计，还可直接用于场板、JTE加场板等终端结构场分析。     

基于FlexPDE的二维PSD数值分析

描述位置敏感探测器(PSD)的Lucovsky方程在一定简化条件下可求得解析解.提出了利用有限元分析软件FlexPDE对非简化Lucovsky方程进行二维区域数值求解的方法,并对四边型PSD进行了分析,结果表明该方法能够准确地分析二维PSD的光电特性,比解析法有更强的适应性.     

相位测量轮廓术中探测器非线性误差的分析

在相位测量轮廓术（PMP）中，探测器的非线性响应是导致测量误差的重要因素，本文对该误差进行了理论分析和计算机模拟，比较了各种相移算法对这一误差的抑制能力，数值模拟与理论分析的结果一致。本文可为实际应用中合理地选择算法提供理论依据。     

Subband-based blind signal separation for noisy speech recognition

A method for directly extracting clean speech features from noisy speech is proposed. This process is based on independent component analysis (ICA) and a new feature analysis technique for reducing the computational complexity of the frequency domain ICA. For noisy speech signals recorded in real environments, this method yielded a considerable performance improvement     

A proportional fair scheduling strategy using multiobjective gradient-based African buffalo optimization algorithm for effective resource allocation and interference minimization

The increased usage of Internet of Things (IoT) applications in several areas, like healthcare, agriculture, and business, has aggravated mobile traffic issues to a large extent. The deployment of 5G technology has resulted in increased traffic globally. These coherent devices, on the other hand, use the internet to fine-tune the quality of service in order to provide scalability, anonymity, and accessibility. Despite its numerous virtues, it is bound to encounter issues with interference management, fairness, throughput, and computational complexities. In this paper, a novel Multi-Objective Gradient-based African Buffalo Optimization (MOGABO) algorithm is developed to handle proportional fairness scheduling, improve interference management, increase throughput, and reduce computational complexities in cellular communication systems, particularly device-to-device (D2D) communication. The simulation analysis is done against other approaches such as the Hungarian technique, DDDPG technique, heuristic technique, and SC-FDMA technique demonstrates that this method has improved system fairness over the existing approaches. Our method maximizes throughput, and for 500 UE, the throughput is 10.4 Mbps, with an average queuing delay of 24 ms. As a result, when compared to the existing method, MOGABO method offers a 1.2% increase in throughput. Thus, our method offers better coverage and throughput for D2D cellular communication with the reduction of computational complexities.     

On the summation of double infinite series field computationsinside rectangular cavities

An accurate and efficient computational technique for the full-wave analysis of passive microstrip lines inside rectangular cavities is described. The unique feature of the technique is the transformation of the double infinite series expansion of the method of moments solution into a few single summations of fast converging series using the residue theorem and the contour-integration method. The technique offers improved convergence especially when the field and source points coincide. Examples of the field computations inside the cavity are presented to verify the technique and its usefulness     

A novel method for analysis of soliton propagation in opticalfibers

We have proposed a new technique to analyze the propagation behavior of solitons in an optical fiber. This technique is based on the Fourier series analysis technique (FSAT) which can consider fiber loss and third-order dispersion. Using this technique, the propagation of solitons in an optical fiber has been investigated. The results compare well with the well-known split-step Fourier method (SSFM). We found that for a given computational accuracy, in order to obtain the same results, we need fewer sampling points for FSAT compared with the SSFM     

一种宽带恒定束宽DBF技术实现方法

针对常规DBF方法不适用于宽带信号的问题提出一种宽带恒定束宽DBF方法.根据空间重采样理论将均匀线阵视为空间离散采样,利用内插定理得到各阵元在不同频点上的加权系数矩阵.利用维纳滤波理论设计宽带波束合成器拟合加权系数矩阵,实现宽带内波束指向一致,宽度相等.给出了理论分析依据及仿真结果,仿真结果表明了该方法具有计算复杂度低、稳健性强的优点.     

Digital System for Detection and Classification of Power Quality Disturbances

This paper presents a new method for detection and classification of power quality disturbances which achieves good performance with low computational cost. The event detection is based on the analysis of the statistical properties of the error signal which is defined as the difference between the sampled signal and its fundamental sinusoidal component, generated by the proposed algorithm. The classification is performed using a multi-rate technique for features extraction over the error signal and a multilayer feed-forward neural network. Numerical results from computational simulation showed the good performance of the proposed method.     

Fast MM-PO-based numerical modelling technique for wideband analysis of antennas near conducting objects

A highly efficient computational electromagnetics technique based upon method of moments-physical optics (MM-PO) hybrid formulation combined with the impedance matrix interpolation and a dynamic adaptive frequency sampling for wideband analysis of antennas radiating in the presence of conducting bodies is presented. Sample numerical results demonstrate the capability of the technique     

基于矩阵特征值分解的MIMO天线选择算法

天线子集选择技术是一种有效的低代价且能够获得MIMO好处的技术.文中提出了一种基于矩阵特征值分解的天线选择算法以最大化信道容量,它将大的搜索空间分成若干小区间,在较小的搜索空间中采用穷举法.分析了所提算法及相关算法的计算复杂度.仿真实验表明,该算法能够获得接近穷举法的中断容量性能而且具有较低的计算复杂度.     

基于高阶矩的Nakagami-m参数估计方法

衰落信道参数估计是无线信道传输系统中的关键技术。首先建立了噪声条件Nakagami-m分布模型,提出了一种基于高阶矩的Nakagami-m参数估计方法。该方法利用了高阶矩具有检测高斯分布特性的特点,用以识别衰落信道的属性。理论分析和仿真试验均证明了新方法的可行性和有效性,并对矩估计方法的计算复杂度进行了分析,从理论上证明其运算量低于最大似然估计方法。     

On the synthesis of very sharp decimators and interpolators using the frequency-response masking technique

Decimation and interpolation are very common multirate signal processing operations. Conventional decimation or interpolation technique using polyphase filters has the advantage that for a given transition-band sharpness, the filter's computational complexity decreases with increasing interpolation or decimation factor. Nevertheless, if the transition band of the decimation or interpolation filter is very sharp, the complexity of the filter may still be very high. The complexity of a very sharp filter may be reduced using the frequency-response masking (FRM) technique. However, as shown in this paper, for a given transition-band sharpness, the computational complexity of the classical FRM method does not reduce as rapidly as the increase in decimation or interpolation factor. In this paper, we present a novel variant of the FRM technique for interpolation or decimation application. In this new variant, the computational complexity reduces as rapidly as the interpolation or decimation factor increases. The reduction in computational complexity increases with decreasing transition width. Over an order of magnitude reduction in computational complexity is achieved when compared with conventional polyphase approach in a particular example presented in this paper.