This paper is basically a survey to show a number of combinatorial optimization problems arising from VLSI circuit design. Some of them including the existence problem, minimax problem, net representation, bend minimi...This paper is basically a survey to show a number of combinatorial optimization problems arising from VLSI circuit design. Some of them including the existence problem, minimax problem, net representation, bend minimization, area minimization, placement problem, routing problem, etc. are especially discussed with new results and theoretical ideas for treating them. Finally, a number of problems for further research are mentioned.展开更多
Embedded systems used in real-time applications require low power, less area and high computation speed. For digital signal processing, image processing and communication applications, data are often received at a con...Embedded systems used in real-time applications require low power, less area and high computation speed. For digital signal processing, image processing and communication applications, data are often received at a continuously high rate. The type of necessary arithmetic functions and matrix operations may vary greatly among different applications. The RTL-based design and verification of one or more of these functions could be time-consuming. Some High Level Synthesis tools reduce this design and verification time but may not be optimal or suitable for low power applications. The design tool proposed in this paper can improve the design time and reduce the verification process. The design tool offers a fast design and verification platform for important matrix operations. These operations range from simple addition to more complex matrix operations such as LU and QR factorizations. The proposed platform can improve design time by reducing verification cycle. This tool generates Verilog code and its testbench that can be realized in FPGA and VLSI systems. The designed system uses MATLAB-based verification and reporting.展开更多
现代视频编码标准普遍采用变换与运动补偿预测混合型编码架构,该架构对运动补偿预测后的残差图像和运动矢量等信息进行变换编码,运动补偿预测的准确度对编码性能有显著影响.由于实际对象的运动精度是任意小的,允许运动矢量具有“分像素...现代视频编码标准普遍采用变换与运动补偿预测混合型编码架构,该架构对运动补偿预测后的残差图像和运动矢量等信息进行变换编码,运动补偿预测的准确度对编码性能有显著影响.由于实际对象的运动精度是任意小的,允许运动矢量具有“分像素”精度,可以有效地提高运动补偿预测准确度,为了得到“分像素”位置的像素值,需要参考其周围相邻的像素值进行插值滤波.文中提出了一种低空间复杂度1/4像素插值方法两步四抽头插值法(Two Steps Four Taps Interpolation,TSFT),该方法与目前国际上最先进的视频编码标准H.264/AVC相比,可以降低11%的空间复杂度,计算复杂度和编码效率相当,已经被国内制定的编码标准AVS1.0采纳.另外,分像素插值是解码端主要的访存和计算瓶颈,文中给出了一个基于多级流水线结构的VLSI实现结构,可以降低访存带宽,同时提高插值器的运算速度,满足高清视频实时解码的需要.展开更多
文摘This paper is basically a survey to show a number of combinatorial optimization problems arising from VLSI circuit design. Some of them including the existence problem, minimax problem, net representation, bend minimization, area minimization, placement problem, routing problem, etc. are especially discussed with new results and theoretical ideas for treating them. Finally, a number of problems for further research are mentioned.
文摘Embedded systems used in real-time applications require low power, less area and high computation speed. For digital signal processing, image processing and communication applications, data are often received at a continuously high rate. The type of necessary arithmetic functions and matrix operations may vary greatly among different applications. The RTL-based design and verification of one or more of these functions could be time-consuming. Some High Level Synthesis tools reduce this design and verification time but may not be optimal or suitable for low power applications. The design tool proposed in this paper can improve the design time and reduce the verification process. The design tool offers a fast design and verification platform for important matrix operations. These operations range from simple addition to more complex matrix operations such as LU and QR factorizations. The proposed platform can improve design time by reducing verification cycle. This tool generates Verilog code and its testbench that can be realized in FPGA and VLSI systems. The designed system uses MATLAB-based verification and reporting.
文摘现代视频编码标准普遍采用变换与运动补偿预测混合型编码架构,该架构对运动补偿预测后的残差图像和运动矢量等信息进行变换编码,运动补偿预测的准确度对编码性能有显著影响.由于实际对象的运动精度是任意小的,允许运动矢量具有“分像素”精度,可以有效地提高运动补偿预测准确度,为了得到“分像素”位置的像素值,需要参考其周围相邻的像素值进行插值滤波.文中提出了一种低空间复杂度1/4像素插值方法两步四抽头插值法(Two Steps Four Taps Interpolation,TSFT),该方法与目前国际上最先进的视频编码标准H.264/AVC相比,可以降低11%的空间复杂度,计算复杂度和编码效率相当,已经被国内制定的编码标准AVS1.0采纳.另外,分像素插值是解码端主要的访存和计算瓶颈,文中给出了一个基于多级流水线结构的VLSI实现结构,可以降低访存带宽,同时提高插值器的运算速度,满足高清视频实时解码的需要.
