并行遗传算法的FPGA硬件实现研究

提出基于FPGA的并行遗传算法的硬件实现系统,从硬件实现角度提高遗传算法的收敛速度.硬件系统划分4个子系统,每个子系统同步而单独地运行一个群体大小为M的简单遗传算法,在简单遗传算法每代结束时,总控制器从4个子系统中选取1个最佳个体,然后复制到与其物理相邻的2个子系统中,实现子系统之间的信息交换.每个子系统采用5段流水线处理技术,即将子系统划分为解码操作、适应度计算、预选操作、随机地址比例选择操作以及交叉-变异操作5个单元.为了解决各段速度瓶颈,适应度计算采用4个具有加速模块的NiosⅡ处理器,预选操作采用M个取整电路,交叉-变异操作采用1个交叉部件和1个变异部件,解码操作采用2个解码部件的内部并行处理方式.用遗传算法标准测试函数Ⅱ测试该硬件系统,实验数据表明,由FPGA硬件实现的并行遗传算法同由软件实现的遗传算法相比,收敛速度大幅度提高,约2个数量级.     

遗传算法的FPGA硬件实现

针对遗传算法软件实现速度慢、效率低的缺点,提出了便于算法实现的串行和流水线两种硬件实现方案。详细描述了设计方案,选择算子、交叉变异算子结构,种群以及适应度的存储和流水线结构,并在流水线中引入并行机制。利用函数极值和旅行商问题分别对方案的资源耗费、运行速度的有效性进行了验证。实验结果显示,这两种硬件实现方法结构简单,资源耗费少,运算速度和运行效率较软件实现大大提高,运行速度平均提升2~3个数量级,为算法在一些实时性要求较高的场合得到应用提供了良好基础。     

适用于演化硬件的遗传算法

为提高演化硬件在演化过程中的收敛速度,实现复杂的演化硬件,研究以Xilinx公司的Virtex-5 Pro系列开发板作为硬件平台的基于SOPC的自演化系统.分析简单遗传算法与量子遗传算法对种群的适应度以及收敛速度的影响;实验中通过全加器电路和2位乘法器电路实现了自演化系统的验证.结合实例,对2种算法分别进行仿真,仿真结果表明,相对于标准遗传算法而言,量子遗传算法效率更高、更适应于进化复杂的大规模电路.     

基于FPGA的遗传算法实现

针对基于软件实现的遗传算法在求解问题的规模与复杂性不断扩大时,往往会速度慢、效率低下的缺点,提出了一种基于现场可编程门阵列的实现方法,并利用测试函数对算法的实现进行效果验证。实际效果显示,这种硬件实现方法,不仅结构简单,而且有效地减少了运算时间、提高了运行效率,为遗传算法能在一些实时、高速的场合得到应用提供了依据。     

智能车车速检测与控制模块设计

论述了基于单片机控制的车速检测与控制模块设计,从硬件设计和软件设计两个方面详细的介绍了系统的构成及工作原理.在硬件电路上,系统包括了车速检测电路设计、车速控制电路设计、系统电源设计,介绍了各个模块元器件的选型及其工作原理.在软件上,勾勒了设计大致思路,利用C语言进行了模块化的设计,使得程序简单易懂.系统通过小车的实际运行证明利用速度传感器可以实现对智能车速度的实时检测,通过对理想速度的设置可以实现对智能车速度的控制,达到了本设计的目的.     

基于FPGA的信息检索技术实现研究

本文首先讨论了基于关键词的信息检索的基本原理,然后介绍了一种基于FPGA的硬件哈希算法来加速信息检索过程中的关键词计数工作.最后比较了硬件实现和软件实现完成关键词计数工作所需的时间,结果表明在检索速度方面FP-GA技术优于软件实现.     

基于IM遗传算法和神经网络的风险管理模型及应用

根据企业多类风险权重评估管理策略,提出一种基于投资理论的多层神经网络模型,并结合IM（Insertion Mutation）算子的遗传算法,以提高精确度和收敛速度。设计并实现基于IM神经网络和遗传算法的软件包。运行结果表明该系统能满足客户需求,对它的应用前景作了展望。     

基于遗传算法的伺服系统参数自整定

针对直流伺服电机的速度控制和位置跟踪问题,根据直流电机的数学模型,搭建了电机的三闭环控制系统模型;依据遗传算法对PID参数的寻优过程及方法,编写了基于遗传算法优化伺服参数的软件,并利用软件对电流环和速度环的控制参数做了寻优实验。实验证明:软件运行良好,基于遗传算法的PI参数整定方法可行,控制效果良好。     

一种KNN算法的可重构硬件加速器设计

设计并实现了一种可快速运算基于哈尔小波变换的KNN(Knearest neighbors)算法且具备可重构能力的硬件结构.该硬件结构通过增减哈尔小波变换组件即可适应不同维度样本的哈尔小波变换;对同样维度样本的计算则可以通过调整并行度满足对逻辑资源和处理时间的不同需求,克服了现有软件KNN计算速度慢、硬件实现的KNN不够灵活的缺陷.通过在Xilinx VC707 FPGA开发板上实现该硬件结构,实验结果展示了不同维度及并行度下算法实现在逻辑资源耗费及运算时间方面的变化.此外,将该硬件结构作为一种高质量轮廓提取算法硬件加速器的纹理分类模块时,在保持计算准确度的情况下获得了远高于软件运行的速度.     

SoPC构架中散列查找用户自定义逻辑设计与实现

针对嵌入式应用系统软件运行速度较低、难以实现快速查找的特点,在Altera公司提供的SoPC片上系统设计方案中,采用用户自定义逻辑实现散列查找。结果表明这种基于FPGA的硬件实现散列查找的方法比软件实现具有更好的性能。     

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。     

快速中值滤波算法的改进及其FPGA实现

针对传统中值滤波算法排序量大、速度慢且处理效果模糊的问题,在快速中值滤波算法的基础上,提出了一种加入阈值比较、且具有更高并行流水结构的改进算法,并在现场可编程门阵列(FPGA)硬件平台上实现了该算法。实验结果表明,改进的快速中值滤波算法不仅减少了比较的次数,还更好地保护了图像的细节,可满足图像预处理对实时性的要求。     

基于最小延时启发式搜索的TTA代码优化

针对传输触发架构下代码生成中指令调度的流水线冲突、调度死锁、资源冲突等问题,给出一种基于最小延时的遗传搜索算法模型,将软件旁路优化和资源动态分配优化整合到该模型中。实验结果表明,该算法能产生较高质量的并行代码,90%以上测试用例的指令级并行度高于表调度算法获得的结果。     

基于FPGA的FESH分组密码算法高速实现

FESH分组密码算法为2019年全国密码算法竞赛中进入第二轮评选的密码算法。文章对该算法的FESH-128-128型进行Verilog HDL高速实现,在有限状态机的基础上对顶层模块采用流水线设计方法进行优化,通过寄存器存储中间数据,提高运行效率。实验结果表明,在软件Quartus Ⅱ 15.0上使用5CEFA7F31C6芯片进行综合设计,采用流水线设计方法进行优化后,算法最高运行速率达到296.74 MHz,相较于有限状态机实现提高了98.28%;吞吐率达到37.98 Gbps,相较于有限状态机实现提升了约33倍。     

Applying genetic algorithm for the development of the components-based embedded system

The embedded system is primarily designed for a particular piece of equipment and it varies on a case-by-case basis. The functionality is required to be specific to the equipment and consequently the application domain is limited. The software embedded in the system also faces problem due to the limitation of the hardware capacity. It is necessary for the designers to consider the hardware capacity and software specification simultaneously while an embedded system is developed. If hardware and software are taken into account concurrently, the design applicability and efficiency are decreased. The evolutionary computing (EC), which comprises techniques of evolutionary programming, evolution strategies, genetic algorithms, and genetic programming has been widely used to solve optimization problems for large scale and complex systems. It is capable to escape not only from local optima due to population based approach, but also from unbiased nature, which enables it to perform well in a situation with little domain knowledge. Therefore, this study proposes an evolutionary approach that applies the characteristics of software reuse, the metrics for the object-oriented concept, and the genetic algorithm to effectively manage and optimize the embedded system. This approach is implemented in the World Wide Web environment. Numerous results associated with performance enhancements of the algorithm are presented in this paper.     

AES和SMS4算法的可重构设计与高效实现

分析AES和SMS4算法的原理及可重构性,给出系统的整体结构,综合应用可重构技术、并行处理及流水线技术对算法进行高效实现。与传统设计方案相比,该设计在保证运行速度的同时大大减少了资源的消耗,因此,适用于面积受限且有多种密码需求的安全 系统。     

Computer-aided pipeline operation using genetic algorithms and rule learning. PART II: Rule learning control of a pipeline under normal and abnormal conditions

In this two-paper series, techniques connected with artificial intelligence and genetics are applied to the problem of gas pipeline control. In the first paper, genetic algorithms were applied to two pipeline optimization problems. In this, the second paper, genetic algorithms are used as a basic learning mechanism in a larger rule learning system called a learning classifier system. The learning classifier system is developed and applied to the control of a gas pipeline under normal summer and winter operations as well as abnormal operations during leak events.A learning classifier system is a software system that learns rules called classifiers to guide its behavior in arbitrary environments. Environmental information comes in through sensors and decodes to a finite length message. Messages on a message list match and fire rules called classifiers with explicit pattern recognition capability. Classifiers, once matched, send messages to the message list. These messages may in turn match other classifiers, or they may fire action triggers called effectors. Using environmental reward, the system selects good rules through a reward allocation system based on a competitive service economy. Competition encourages the survival of good rules, those that set up environmental reward. Furthermore, the system tries and learns new rules using a genetic algorithm similar to the one presented in the previous paper.Together, the learning classifier system with its complete rule and message system and powerful learning heuristic is capable of learning how to operate a pipeline under normal and abnormal conditions alike. Computational experiments are presented that demonstrate the systems learning ability under summer and winter conditions starting from a random state of mind. These results compare favorably with a random walk through the decision space. Additionally, the learning classifier system is trained to detect leaks. Repeated exposure to simulated leak events results in the development of rules that permit a high percentage of detected leaks and a low percentage of false alarms.     

Generating depth of-field effects in virtual reality applications

Computer-generated images without depth-of-field effects always look surreal. A new technique and algorithm are presented which solve the problem of simulating these effects in VR and other real-time applications. The adaptive convolution approach can generate depth-of-field effects in real time using commercially available hardware filters. The picture-processing algorithm proposed can be easily implemented in a visualization pipeline. The final effect can be obtained by iterative filtering with a single hardware convolution filter or with identical 3×3 filters placed as pipeline stages. The speed of this approach makes it suitable for generating images in real-time applications, particularly virtual reality systems. As a postprocessing technique, it also has the advantage of being easily implemented as a final stage of the visualization pipeline in an existing rendering system     

基于改进遗传算法的可重构计算任务划分

为实现可重构计算中的软硬件任务自动划分,引入了遗传算法来搜寻最优解。为解决标准遗传算法可能出现种群早熟和种群进化后期收敛速度慢的问题,使用了小生境技术来保护种群中基因的多样性。设计了能够随适应度自动改变的自适应遗传算子（杂交算子和变异算子）。对算法进行了50次随机实验,并对结果进行分析。实验表明,改进后的遗传算法搜寻到全局最优任务划分的概率和搜寻到最优任务划分时的进化代数都要优于标准遗传算法。