一种基于半监督AdaBoost模型树的FPGA性能表征方法  

An FPGA performance characterization approach based on semi-supervised Ada Boost model tree

在线阅读下载全文

作  者:杨立群[1,2] 李威[1] 黄志洪[1] 孙嘉斌[1] 杨海钢[1] YANG Liqun;LI Wei;HUANG Zhihong;SUN Jiabin;YANG Haigang(System on Programmable Chip Research Department,Institute of Electronics,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区:[1]中国科学院电子学研究所可编程芯片与系统研究室,北京100190 [2]中国科学院大学,北京100049

出  处:《太赫兹科学与电子信息学报》2016年第4期647-652,共6页Journal of Terahertz Science and Electronic Information Technology

基  金:国家自然科学基金资助项目(61271149)

摘  要:提出了一种基于半监督自适应增强(Ada Boost)模型树的建模方法,用于现场可编程门阵列(FPGA)的性能表征。该方法以半监督学习方式,构建了FPGA性能关于FPGA架构参数的解析模型,同时采用Ada Boost算法提高FPGA性能模型的预测精确度。使用VTR(Verilog To Routing)电路集,基于该方法构建的性能模型在预测FPGA上实现的应用电路面积时,平均相对误差(MRE)为4.42%;预测延时的MRE为1.63%;预测面积延时积时,MRE为5.06%。与全监督模型树算法以及现有的半监督模型树算法相比较,该方法构建的FPGA实现面积模型的预测精确度分别提高了39%,26%。实验结果显示,该方法在确保较少的时间开销前提下,构建了具有高预测精确度的FPGA性能模型,提供了一种高效的FPGA性能表征方法。A semi-supervised Adaptive Boosting(AdaBoost) model tree based modeling approach isproposed for Field Programmable Gate Array(FPGA) performance characterization. The proposed approach,which adopts AdaBoost to improve the prediction accuracy, constructs an analytical performance modelwith regard to the FPGA architecture parameters in semi-supervised learning way. The FPGA performancemodel built through the proposed approach estimates the area, delay and area-delay product with MeanRelative Errors(MREs) of 4.42%, 1.62% and 5.06%, respectively. Compared to the supervised model treeand the previous semi-supervised model tree algorithm, the proposed approach boosts the estimationaccuracy by 39% and 26% respectively. Experimental results show that the proposed approach is proved tobe an efficient FPGA characterization approach, building FPGA performance models with high accuracy inless time cost. The proposed modeling approach can be applied to explore the FPGA architecture designspace effectively and efficiently.

关 键 词:FPGA性能表征 半监督模型树 AdaBoost模型树 

分 类 号:TN791[电子电信—电路与系统] TP301.6[自动化与计算机技术—计算机系统结构]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象