Machine learning algorithm partially reconfigured on FPGA for an image edge detection system

Unmanned aerial vehicles(UAVs)have been widely used in military,medical,wireless communications,aerial surveillance,etc.One key topic involving UAVs is pose estimation in autonomous navigation.A standard procedure for this process is to combine inertial navigation system sensor information with the global navigation satellite system(GNSS)signal.However,some factors can interfere with the GNSS signal,such as ionospheric scintillation,jamming,or spoofing.One alternative method to avoid using the GNSS signal is to apply an image processing approach by matching UAV images with georeferenced images.But a high effort is required for image edge extraction.Here a support vector regression(SVR)model is proposed to reduce this computational load and processing time.The dynamic partial reconfiguration(DPR)of part of the SVR datapath is implemented to accelerate the process,reduce the area,and analyze its granularity by increasing the grain size of the reconfigurable region.Results show that the implementation in hardware is 68 times faster than that in software.This architecture with DPR also facilitates the low power consumption of 4 mW,leading to a reduction of 57%than that without DPR.This is also the lowest power consumption in current machine learning hardware implementations.Besides,the circuitry area is 41 times smaller.SVR with Gaussian kernel shows a success rate of 99.18%and minimum square error of 0.0146 for testing with the planning trajectory.This system is useful for adaptive applications where the user/designer can modify/reconfigure the hardware layout during its application,thus contributing to lower power consumption,smaller hardware area,and shorter execution time.

A Novel Aquila Optimizer Based PV Array Reconfiguration Scheme to Generate Maximum Energy under Partial Shading Condition

This paper develops a real-time PV arrays maximum power harvesting scheme under partial shading condition(PSC)by reconfiguring PV arrays using Aquila optimizer(AO).AO is based on the natural behaviors of Aquila in capturing prey,which can choose the best hunting mechanism ingeniously and quickly by balancing the local exploitation and global exploration via four hunting methods of Aquila:choosing the searching area through high soar with the vertical stoop,exploring in different searching spaces through contour flight with quick glide attack,exploiting in convergence searching space through low flight with slow attack,and swooping through walk and grabbing prey.In general,PV arrays reconfiguration is a problem of discrete optimization,thus a series of discrete operations are adopted in AO to enhance its optimization performance.Simulation results based on 10 cases under PSCs show that the mismatched power loss obtained by AO is the smallest compared with genetic algorithm,particle swarm optimization,ant colony algorithm,grasshopper optimization algorithm,and butterfly optimization algorithm,which reduced by 4.34%against butterfly optimization algorithm.

Optimal PV array reconfiguration under partial shading condition through dynamic leader based collective intelligence

This paper applies the innovative idea of DLCI to PV array reconfiguration under various PSCs to capture the maxi-mum output power of a PV generation system.DLCI is a hybrid algorithm that integrates multiple meta-heuristic algo-rithms.Through the competition and cooperation of the search mechanisms of different metaheuristic algorithms,the local exploration and global development of the algorithm can be effectively improved to avoid power mismatch of the PV system caused by the algorithm falling into a local optimum.A series of discrete operations are performed on DLCI to solve the discrete optimization problem of PV array reconfiguration.Two structures(DLCI-I and DLCI-II)are designed to verify the effect of increasing the number of sub-optimizers on the optimized performance of DLCI by simulation based on 10 cases of PSCs.The simulation shows that the increase of the number of sub-optimizers only gives a relatively small improvement on the DLCI optimization performance.DLCI has a significant effect on the reduction in the number of power peaks caused by PSC.The PV array-based reconstruction system of DLCI-II is reduced by 4.05%,1.88%,1.68%,0.99%and 3.39%,when compared to the secondary optimization algorithms.

基于FMQL的双通道局部重构技术研究

针对双通道通信设备多波形同时独立运行的复杂需求,研究并提出了一种基于FMQL的双通道局部重构架构,该架构设计了多组件通信机制、重构波形加载流程、逻辑资源分配方法、波形接口和标识、重构解耦机制、时钟管理以及组件管理方法等。项目应用结果表明,双通道局部重构波形加载和卸载功能正常,相较于传统方法,所提方法的波形支持能力、波形加载速度、逻辑资源利用率均大幅提升。目前,该技术已在航空通信多型设备中得到应用,具有较强的实用性。

A novel reconfiguration of the solar array to enhance peak power and efficiency under partial shading conditions:experimental validation

Non-homogeneous irradiation patterns and temperature levels immensely affect the performance of solar photovoltaic arrays.Partial shading conditions on solar arrays reduce the peak power and efficiency.This paper provides a new remedy called a novel Ramanujan reconfiguration(NRR)to eliminate this physical shading problem in solar photovoltaic systems.NRR is a static-based reconfigured technique that is built using a three-diode model with the help of the MATLAB®/Simulink®tool.The special feature of the proposed NRR technique is that when shade occurs on the solar modules,it gets realigned in a particular row,column,diagonal,corner,centre and middle peripheral cages.This helps over a wide range of shade dispersion on the solar array.The novel topology is tested against the conventional total cross-tied(TCT)model and recently introduced advanced reconfigured models,namely odd–even topology(OET)and Kendoku topology(KDT).The results are tested under certain shading conditions.The proposed NRR technique increases the peak power by 4.45,2.15 and 2.17 W under the first shading condition regarding TCT,OET and KDT.Its efficiency is improved by 0.51–2.18%under the third shading condition compared with other considered models in this study.In addition,NRR leads to smooth output curves under the second,third and fourth shading conditions,effectively mitigating the local power peaks.The experimental results show the proposed enhanced performance of the novel model against the other models.Graphical Abstract Remedy for physical problem correlated with solar photovoltaics Comparison with traditional and recent solar models Conclusion:NRR has effectively handled the problem related with solar models.It has improved the efficiency up to 31.44%under S4.Also,smooth output curves under S2-S4 shows its effectiveness in mitigating the local power peaks.Greater power gain at 3.94%under S4 is achieved by novel model.Real-time verification proves the supremacy of novel proposed model over other considered models in this work.

Hardware Module Placement for Dynamically Reconfigurable Architecture

Partial Reconfigurable FPGAs （Field Programmable Gate Array） allow tasks to be placed and removed dynamically at runtime. One of the challenging problems is the placement of modules on reconfigurable resources. Several modules placement techniques have been introduced in the literature to solve the temporal placement problem. This paper presents a temporal placement approach that manages the resources of a reconfigurable device. In fact, the authors＇ contribution focuses on introducing a new temporal placement algorithm that aims to minimize the communication cost between modules. Results show an important improvement in communication cost compared with other approaches.

FPGA中基于空间连续性的碎片度量及任务放置

针对部分可重构现场可编程门阵列允许在运行时对芯片的各个部分进行配置导致的区域碎片,提出了一种新的基于被占用(或空闲)空间的连续性的碎片度量及在线任务放置方法;首先从一维结构出发,得到一个单元序列对一个单元流S的碎片度量F S的贡献值,进而得到一维碎片度量值,它不依赖于到达任务的大小;然后将一维结构得到的碎片度量值结果推广到二维及高维结构;最后在FPGA上的在线任务放置过程中采用这种碎片度量方法,从而减少芯片碎片;在二维结构的FPGA上的仿真实验结果表明,与通常采用的左下角、第一匹配和最佳匹配放置策略相比,采用提出的碎片度量及放置方法不仅在等待时间、分配时间和响应时间方面有所改善,而且提高了芯片的利用率,降低了失配率。

局部动态可重构FPGA进程式调度系统设计与实现

针对6G时代多样的边缘计算要求,基于FPGA上的可重构技术可以实现更低的时延同时提供多样性的服务。基于局部动态重配置的思路,使用ICAP接口对FPGA资源进行重新配置,从而实现FPGA逻辑上的局部动态可重构方案。借鉴操作系统中软件进程管理的思想,基于Linux操作系统中引入硬件进程的概念,这样可以将一整块FPGA资源划分为多个小的FPGA资源块,每一个小的可重构的FPGA资源块都可以抽象成为一个硬件进程,硬件进程实际并不运行在CPU上而是运行在FPGA逻辑资源区域中,在操作系统上只是硬件进程的软件语言描述。由此,设计出CPU加FPGA的硬件方案来实现局部可重构系统,并在Xilinx公司Zynq系列芯片上进行了验证,将FPGA硬件资源进行进程式调度以及资源分配,大大提高了FPGA硬件资源的利用率以及灵活性。

PV Array Reconfiguration Based on the Shaded Cells'Number for PV Modules

Reconfiguration can increase the output power for a PV array under partial shadows.However,traditional reconfiguration methods consider the PV module as either totally shaded or totally unshaded,and module-based simulation is employed to evaluate the reconfiguration effect.Actually,there is an unneglectable error when treating a partially shaded PV module as totally shaded,through using a more accurate cellbased simulation.Based on the analysis of the determinant factors on MPPs’power of a PV array,a new reconfiguration method is proposed based on the exact partial shadow shape projected on the PV array.This method restructures the electrical connection among PV modules of a PV array according to the shaded cells’number(SCN)of every PV module.Extensive cell-based simulations are carried out on a PV array to verify the effectiveness of the proposed SCN-based reconfiguration method.Comprehensive comparisons among various reconfiguration methods and shadow distributions clearly show its suitability to different irregular shadows and its superiority in PV output power enhancement.

Maximum Power Enhancement Under Partial Shadings Using a Modified Sudoku Reconfiguration

Partial shading is one of the important factors in reducing maximum power generation from PV(Photovoltaic)arrays.Maximum power generation can be improved by selecting a PV array through a Total-Cross-Tied(T-C-T)connection.However,maximum power generated from T-C-T can still be improved by distribution of shading over various rows.Due to distribution of shading,the current entering the node increases and results in improved maximum power generation.This can be done effectively by using Sudoku reconfiguration techniques.These techniques are economical,since they don’t require any sensors and switching networks.This technique only changes the physical location of the PV panel but the electrical connection between the panels remains the same.This paper proposes a Modified Sudoku reconfiguration pattern which enhances the maximum power from the T-C-T connected PV array.Furthermore,the theoretical calculation of row current and power output have been done for existing and proposed topologies under various shading patterns.The performance of the proposed pattern has been analyzed and compared using specifications,such as Global Maximum Power(GMP),Fill Factor(FF),mismatch losses,and efficiency.From the results,it can be concluded that the Modified Sudoku reconfiguration enhances the GMP under all shading conditions.

一种基于动态部分重构的FPGA自修复控制器

由于FPGA具有灵活性高、开发周期短和性能强等优点,被广泛应用于航空航天自动化电子设备中。以航空航天强辐射环境下SRAM型FPGA易受到单粒子效应的影响、会导致电子设备发生故障为研究背景,以提高SRAM型FPGA的可靠性、硬件资源利用率和减少故障修复时间为目标展开研究。基于FPGA动态部分重构技术设计了一种自修复控制器,能够根据系统的故障自检测信号,自主完成FPGA故障区域中部分位流的调度和配置刷新,有效缓解了单粒子应对FPGA器件造成的影响。实验证明:该自修复控制器的可行性和高效性可提高航空航天自动化电子设备的安全性和可靠性。

智能超表面辅助边缘计算任务卸载增强技术

针对无线信道的随机性和不确定性导致的MEC网络任务卸载时传输速率低的问题,提出一种RIS辅助MEC方案,利用RIS增强链路的能力来支持任务卸载。通过对用户发射功率、任务卸载量、MEC计算资源调度、RIS相移矩阵等进行联合优化,最小化服务总时延。由于原问题是具有高度耦合变量的非凸问题,借助块坐标下降法将原问题分解为三个子问题,并利用拉格朗日对偶法、逐次凸近似法和交替方向乘子法进行求解。最后设计了一种交替算法,迭代求解近似最优解。仿真结果表明,与其他基准方案相比,该方案在降低服务时延方面具有更好的性能。

FPGA动态可重构技术原理及实现方法分析

FPGA动态重构技术主要是指对于特定结构的FPGA芯片 ,在一定的控制逻辑的驱动下 ,对芯片的全部或部分逻辑资源实现在系统的高速的功能变换 ,从而实现硬件的时分复用 ,,节省逻辑资源。本文以SRAMFPGA的可重配置特性为基础 ,结合几种新型的动态重构FPGA芯片结构 ,分析了FPGA动态重构技术的原理 ,讨论了其实现方法。

基于FPGA的动态可重构系统实现

介绍了Xilinx公司VirtexTMFPGA芯片的配置原理,采用模块化设计实现FPGA的动态部分重构,并设计出CPU加CPLD配置FPGA的硬件方案来实现可重构系统.FPGA采用Select MAP配置方式,实现配置逻辑的快速重构和动态部分重构.同时给出了可重构系统配置的软件流程,并计算出相应的重构时间.

基于遗传算法的可重构系统软硬件划分

在考虑动态部分重构及重构延时等特征的基础上,采用遗传算法及其与爬山算法的融合实现可重构系统软硬件任务的划分,并采用动态优先级调度算法进行划分结果的评价。实验表明,在可重构系统的资源约束等条件下,算法能够有效地实现应用任务图到可重构系统的时空映射。

动态部分重配置及其FPGA实现

动态部分重配置充分利用了FPGA芯片提供的可重配置功能,提高了FPGA芯片的利用率,减小了FPGA芯片的配置时间,有效地提高了系统的整体性能。该文介绍了动态部分重配置的两种实现方法,并在Spartan-ⅡFPGA上进行了验证。

基于FPGA的动态可重构系统设计与实现

近年来,随着计算机技术的发展,尤其是现场可编程门阵列FPGA的出现,使实时电路重构成为研究热点。基于FPGA的重构系统具有自适应、自主修复特性,在空间应用中具有非常重要的作用。介绍FPGA可重构技术的分类以及动态可重构技术的原理,并在此基础之上选取Virtex-4系列FPGA给出一种动态重构的应用以及具体实现,即通过微处理器(ARM)结合多个FPGA,并采用一种新的边界扫描链方法对多个FPGA进行配置,从而实现局部动态可重构。这种实现方法具有较强通用性和适于模块化设计等优点。

基于局部动态可重构技术的多通道数据采集系统

可重构技术既具有硬件任务的高效性又具有软件任务的灵活性,近年已经成为研究热点,并越来越广泛的应用到实际系统之中.应用在化工、电力、冶金等行业的大型监控系统中的多通道数据采集系统具有高可靠性、强实时性和高并行性特点,典型的多通道数据采集系统普遍采用ASIC集成电路设计,可配置性、灵活性和扩展性差,本文提出一种将局部动态可重构技术应用到多通道数据采集系统的设计方案,提高了系统的可配置性和应用范围,并且系统具有更高的实时性和并行性.

电力系统恢复控制的网络重构智能优化策略

作为现代电力系统恢复控制的核心研究内容之一,该文对恢复控制中的网络重构问题进行探讨,提出最优送电路径的通用模型和相应的智能优化算法解算模式。以寻找最短的加权送电路径为优化目标,将网络重构建模为一个寻找图的局部最小树问题,并计及各种约束。利用遗传算法易于处理离散变量且具有全局收敛性的特点,对该优化问题进行求解。求解过程中,对算法寻优性能进行研究以提高求解速度、算法稳定性和寻优效率。所提方法能较好地解决解算精度与速度的矛盾。最后以IEEE30节点系统作为算例,验证所提模型和算法的有效性。

基于模块化设计方法实现FPGA动态部分重构

介绍了Xilinx FPGA的配置原理和FPGA模块化设计流程以及划分重构模块的原则。通过一个实例介绍了采用模块化设计方法实现Virtex-E FPGA动态部分重构的过程,能使重构模块在系统运行时改变其逻辑功能,而固定模块逻辑功能不中断,同时器件的重构时间大大减少。