By applying switch-signal theory, the interaction between MOS transmission switch-ing transistor and current signal in current-mode CMOS circuits is analyzed, and the theory oftransmission current-switches which is su...By applying switch-signal theory, the interaction between MOS transmission switch-ing transistor and current signal in current-mode CMOS circuits is analyzed, and the theory oftransmission current-switches which is suitable to current-mode CMOS circuits is proposed. Thecircuits, such as ternary full-adder etc., designed by using this theory have simpler circuit struc-tures and correct logic functions. It is confirmed that this theory is efficient in guiding the logicdesign of current-mode CMOS circuits at switch level.展开更多
By applying switch-signal theory, the theory of transmission current-switches based on symmetric ternary logic is proposed, this theory is suitable to design symmetric ternary current-mode CMOS circuits. The symmetric...By applying switch-signal theory, the theory of transmission current-switches based on symmetric ternary logic is proposed, this theory is suitable to design symmetric ternary current-mode CMOS circuits. The symmetric ternary current-mode CMOS circuits designed by using this theory not only have simpler circuit structures and correct logic functions, but also can process bidirectional signals.展开更多
This paper proposes a simplification method for realization of current-mode multivalued CMOS circuits. The key of this method is to find a cover on the K-map for a given multivalued function, which fits to the realiza...This paper proposes a simplification method for realization of current-mode multivalued CMOS circuits. The key of this method is to find a cover on the K-map for a given multivalued function, which fits to the realization of current-mode CMOS circuits. The design example shows that the design presented in this paper is better than the design proposed by G. W. Dueck et al. (1987).展开更多
Based on the physical characteristics of SiGe material,a new three-dimensional （3D） CMOS IC structure is proposed,in which the first device layer is made of Si material for nMOS devices and the second device layer i...Based on the physical characteristics of SiGe material,a new three-dimensional （3D） CMOS IC structure is proposed,in which the first device layer is made of Si material for nMOS devices and the second device layer is made of Six Ge1- x material for pMOS. The intrinsic performance of ICs with the new structure is then limited by Si nMOS.The electrical characteristics of a Si-SiGe 3D CMOS device and inverter are all simulated and analyzed by MEDICI. The simulation results indicate that the Si-SiGe 3D CMOS ICs are faster than the Si-Si 3D CMOS ICs. The delay time of the 3D Si-SiGe CMOS inverter is 2-3ps,which is shorter than that of the 3D Si-Si CMOS inverter.展开更多
基于忆阻器阵列的类脑电路为实现高能效神经网络计算提供了极具潜力的技术路线.然而,现有方案通常需要使用大量的模数转换过程,成为计算电路能效进一步提升的瓶颈.因此,提出了一种基于1T1R(1 Transistor 1 Resistor)忆阻器交叉阵列与CMO...基于忆阻器阵列的类脑电路为实现高能效神经网络计算提供了极具潜力的技术路线.然而,现有方案通常需要使用大量的模数转换过程,成为计算电路能效进一步提升的瓶颈.因此,提出了一种基于1T1R(1 Transistor 1 Resistor)忆阻器交叉阵列与CMOS(Complementary Metal-Oxide-Semiconductor)激活函数的全模拟神经网络架构,以及与其相关的训练优化方法 .该架构采用1T1R忆阻器交叉阵列来实现神经网络线性层中的模拟计算,同时利用CMOS非线性电路来实现神经网络激活层的模拟计算,在全模拟域实现神经网络大幅减少了模数转换器的使用,优化了能效和面积成本.实验结果验证了忆阻器作为神经网络权重层的可行性,同时设计多种CMOS模拟电路,在模拟域实现了多种非线性激活函数,如伪ReLU(Rectified Linear Unit)、伪Sigmoid、伪Tanh、伪Softmax等电路.通过定制化训练方法来优化模拟电路神经网络的训练过程,解决了实际非线性电路的输出饱和条件下的训练问题.仿真结果表明,即使在模拟电路的激活函数与理想激活函数不一致的情况下,全模拟神经网络电路在MNIST(Modified National Institute of Standards and Technology)手写数字识别任务中的识别率仍然可以达到98%,可与基于软件的标准网络模型的结果相比.展开更多
To improve the accuracy and speed in cycle-accurate power estimation, this paper uses multiple dimensional coefficients to build a Bayesian inference dynamic power model. By analyzing the power distribution and intern...To improve the accuracy and speed in cycle-accurate power estimation, this paper uses multiple dimensional coefficients to build a Bayesian inference dynamic power model. By analyzing the power distribution and internal node state, we find the deficiency of only using port information. Then, we define the gate level number computing method and the concept of slice, and propose using slice analysis to distill switching density as coefficients in a special circuit stage and participate in Bayesian inference with port information. Experiments show that this method can reduce the power-per-cycle estimation error by 21.9% and the root mean square error by 25.0% compared with the original model, and maintain a 700 ＋ speedup compared with the existing gate-level power analysis technique.展开更多
Simulating the total ionizing dose(TID)of an electrical system using transistor-level models can be difficult and expensive,particularly for digital-integrated circuits(ICs).In this study,a method for modeling TID eff...Simulating the total ionizing dose(TID)of an electrical system using transistor-level models can be difficult and expensive,particularly for digital-integrated circuits(ICs).In this study,a method for modeling TID effects in complementary metaloxide semiconductor(CMOS)digital ICs based on the input/output buffer information specification(IBIS)was proposed.The digital IC was first divided into three parts based on its internal structure:the input buffer,output buffer,and functional area.Each of these three parts was separately modeled.Using the IBIS model,the transistor V-I characteristic curves of the buffers were processed,and the physical parameters were extracted and modeled using VHDL-AMS.In the functional area,logic functions were modeled in VHDL according to the data sheet.A golden digital IC model was developed by combining the input buffer,output buffer,and functional area models.Furthermore,the golden ratio was reconstructed based on TID experimental data,enabling the assessment of TID effects on the threshold voltage,carrier mobility,and time series of the digital IC.TID experiments were conducted using a CMOS non-inverting multiplexer,NC7SZ157,and the results were compared with the simulation results,which showed that the relative errors were less than 2%at each dose point.This confirms the practicality and accuracy of the proposed modeling method.The TID effect model for digital ICs developed using this modeling technique includes both the logical function of the IC and changes in electrical properties and functional degradation impacted by TID,which has potential applications in the design of radiation-hardening tolerance in digital ICs.展开更多
在用于人脸检测的图像传感器中,帧率和功耗均是关键指标。针对人脸检测智能图像传感器帧率较低和功耗较高的问题,设计了一种高速低功耗读出电路,利用卷积神经网络进行人脸检测,将原始图像输出的数据量压缩了66.6%,减少了模数转换的次数...在用于人脸检测的图像传感器中,帧率和功耗均是关键指标。针对人脸检测智能图像传感器帧率较低和功耗较高的问题,设计了一种高速低功耗读出电路,利用卷积神经网络进行人脸检测,将原始图像输出的数据量压缩了66.6%,减少了模数转换的次数。所使用的网络可以实现3×3卷积、修正线性单元、2×2最大池化和1×1全连接层。读出电路支持可配置时序,通过配置不同的时序,读出电路适用于灰度成像模式和卷积模式。基于55 nm CMOS工艺进行设计,电源供电电压为1.2 V,卷积模式下输出帧率达到603帧/s,读出电路的总功耗为137μW。在Labeled Faces in the Wild数据集中,人脸检测的准确率达到了98.3%。展开更多
A simple and successful method for the stability enhancement of integrated circuits is presented. When the process parameters, temperature, and supply voltage are changed, according to the simulation results, this met...A simple and successful method for the stability enhancement of integrated circuits is presented. When the process parameters, temperature, and supply voltage are changed, according to the simulation results, this method yields a standard deviation of the transconductance of MOSFETs that is 41.4% less than in the uncompensated case. This method can be used in CMOS LC oscillator design.展开更多
New methodologies for l-Bit XOR-XNOR full- adder circuits are proposed to improve the speed and power as these circuits are basic building blocks for ALU circuit implementation. This paper presents comparative study o...New methodologies for l-Bit XOR-XNOR full- adder circuits are proposed to improve the speed and power as these circuits are basic building blocks for ALU circuit implementation. This paper presents comparative study of high-speed, low-power and low voltage full adder circuits. Simulation results illustrate the superiority of the proposed adder circuit against the conventional complementary metal-oxide-semiconductor （CMOS）, complementary pass-transistor logic （CPL）, TG, and Hybrid adder circuits in terms of delay, power and power delay product （PDP）. Simulation results reveal that the proposed circuit exhibits lower PDP and is more power efficient and faster when compared with the best available 1-bit full adder circuits. The design is implemented on UMC 0.18 μm process models in Cadence Virtuoso Schematic Composer at 1.8 V single ended supply voltage and simulations are carried out on Spectre S.展开更多
This paper analyses the transient characteristics of high temperature CMOS inverters and gate circuits, and gives the computational formulas of their rise time, fall time and delay time. It may be concluded that the t...This paper analyses the transient characteristics of high temperature CMOS inverters and gate circuits, and gives the computational formulas of their rise time, fall time and delay time. It may be concluded that the transient characteristics of CMOS inverters and gate circuits deteriorate due to the reduction of carrier mobilities and threshold voltages of MOS transistors and the increase of leakage currents of MOS transistors drain terminal pn junctions. The calculation results can explain the experimental phenomenon.展开更多
Based on transmission function theory,the synthesis technique for multivaluedCMOS circuits is discussed.By comparing the CMOS circuits based on transmission functiontheory with the T gate,it is shown that their action...Based on transmission function theory,the synthesis technique for multivaluedCMOS circuits is discussed.By comparing the CMOS circuits based on transmission functiontheory with the T gate,it is shown that their action principles are identical.Based on it,thesynthesis method for multivalued CMOS circuits with many variables by using function decom-position is proposed.展开更多
基金Supported by National Natural Science Foundation of China
文摘By applying switch-signal theory, the interaction between MOS transmission switch-ing transistor and current signal in current-mode CMOS circuits is analyzed, and the theory oftransmission current-switches which is suitable to current-mode CMOS circuits is proposed. Thecircuits, such as ternary full-adder etc., designed by using this theory have simpler circuit struc-tures and correct logic functions. It is confirmed that this theory is efficient in guiding the logicdesign of current-mode CMOS circuits at switch level.
基金National Natural Science Foundation of ChinaNatural science Foundation of Zhejiang Province
文摘By applying switch-signal theory, the theory of transmission current-switches based on symmetric ternary logic is proposed, this theory is suitable to design symmetric ternary current-mode CMOS circuits. The symmetric ternary current-mode CMOS circuits designed by using this theory not only have simpler circuit structures and correct logic functions, but also can process bidirectional signals.
基金Supported by the National Natural Science Foundation of China
文摘This paper proposes a simplification method for realization of current-mode multivalued CMOS circuits. The key of this method is to find a cover on the K-map for a given multivalued function, which fits to the realization of current-mode CMOS circuits. The design example shows that the design presented in this paper is better than the design proposed by G. W. Dueck et al. (1987).
文摘Based on the physical characteristics of SiGe material,a new three-dimensional （3D） CMOS IC structure is proposed,in which the first device layer is made of Si material for nMOS devices and the second device layer is made of Six Ge1- x material for pMOS. The intrinsic performance of ICs with the new structure is then limited by Si nMOS.The electrical characteristics of a Si-SiGe 3D CMOS device and inverter are all simulated and analyzed by MEDICI. The simulation results indicate that the Si-SiGe 3D CMOS ICs are faster than the Si-Si 3D CMOS ICs. The delay time of the 3D Si-SiGe CMOS inverter is 2-3ps,which is shorter than that of the 3D Si-Si CMOS inverter.
文摘基于忆阻器阵列的类脑电路为实现高能效神经网络计算提供了极具潜力的技术路线.然而,现有方案通常需要使用大量的模数转换过程,成为计算电路能效进一步提升的瓶颈.因此,提出了一种基于1T1R(1 Transistor 1 Resistor)忆阻器交叉阵列与CMOS(Complementary Metal-Oxide-Semiconductor)激活函数的全模拟神经网络架构,以及与其相关的训练优化方法 .该架构采用1T1R忆阻器交叉阵列来实现神经网络线性层中的模拟计算,同时利用CMOS非线性电路来实现神经网络激活层的模拟计算,在全模拟域实现神经网络大幅减少了模数转换器的使用,优化了能效和面积成本.实验结果验证了忆阻器作为神经网络权重层的可行性,同时设计多种CMOS模拟电路,在模拟域实现了多种非线性激活函数,如伪ReLU(Rectified Linear Unit)、伪Sigmoid、伪Tanh、伪Softmax等电路.通过定制化训练方法来优化模拟电路神经网络的训练过程,解决了实际非线性电路的输出饱和条件下的训练问题.仿真结果表明,即使在模拟电路的激活函数与理想激活函数不一致的情况下,全模拟神经网络电路在MNIST(Modified National Institute of Standards and Technology)手写数字识别任务中的识别率仍然可以达到98%,可与基于软件的标准网络模型的结果相比.
文摘To improve the accuracy and speed in cycle-accurate power estimation, this paper uses multiple dimensional coefficients to build a Bayesian inference dynamic power model. By analyzing the power distribution and internal node state, we find the deficiency of only using port information. Then, we define the gate level number computing method and the concept of slice, and propose using slice analysis to distill switching density as coefficients in a special circuit stage and participate in Bayesian inference with port information. Experiments show that this method can reduce the power-per-cycle estimation error by 21.9% and the root mean square error by 25.0% compared with the original model, and maintain a 700 ＋ speedup compared with the existing gate-level power analysis technique.
基金This work was supported by the special fund of the State Key Laboratory of Intense Pulsed Radiation Simulation and Effect(No.SKLIPR2011).
文摘Simulating the total ionizing dose(TID)of an electrical system using transistor-level models can be difficult and expensive,particularly for digital-integrated circuits(ICs).In this study,a method for modeling TID effects in complementary metaloxide semiconductor(CMOS)digital ICs based on the input/output buffer information specification(IBIS)was proposed.The digital IC was first divided into three parts based on its internal structure:the input buffer,output buffer,and functional area.Each of these three parts was separately modeled.Using the IBIS model,the transistor V-I characteristic curves of the buffers were processed,and the physical parameters were extracted and modeled using VHDL-AMS.In the functional area,logic functions were modeled in VHDL according to the data sheet.A golden digital IC model was developed by combining the input buffer,output buffer,and functional area models.Furthermore,the golden ratio was reconstructed based on TID experimental data,enabling the assessment of TID effects on the threshold voltage,carrier mobility,and time series of the digital IC.TID experiments were conducted using a CMOS non-inverting multiplexer,NC7SZ157,and the results were compared with the simulation results,which showed that the relative errors were less than 2%at each dose point.This confirms the practicality and accuracy of the proposed modeling method.The TID effect model for digital ICs developed using this modeling technique includes both the logical function of the IC and changes in electrical properties and functional degradation impacted by TID,which has potential applications in the design of radiation-hardening tolerance in digital ICs.
文摘在用于人脸检测的图像传感器中,帧率和功耗均是关键指标。针对人脸检测智能图像传感器帧率较低和功耗较高的问题,设计了一种高速低功耗读出电路,利用卷积神经网络进行人脸检测,将原始图像输出的数据量压缩了66.6%,减少了模数转换的次数。所使用的网络可以实现3×3卷积、修正线性单元、2×2最大池化和1×1全连接层。读出电路支持可配置时序,通过配置不同的时序,读出电路适用于灰度成像模式和卷积模式。基于55 nm CMOS工艺进行设计,电源供电电压为1.2 V,卷积模式下输出帧率达到603帧/s,读出电路的总功耗为137μW。在Labeled Faces in the Wild数据集中,人脸检测的准确率达到了98.3%。
文摘A simple and successful method for the stability enhancement of integrated circuits is presented. When the process parameters, temperature, and supply voltage are changed, according to the simulation results, this method yields a standard deviation of the transconductance of MOSFETs that is 41.4% less than in the uncompensated case. This method can be used in CMOS LC oscillator design.
文摘New methodologies for l-Bit XOR-XNOR full- adder circuits are proposed to improve the speed and power as these circuits are basic building blocks for ALU circuit implementation. This paper presents comparative study of high-speed, low-power and low voltage full adder circuits. Simulation results illustrate the superiority of the proposed adder circuit against the conventional complementary metal-oxide-semiconductor （CMOS）, complementary pass-transistor logic （CPL）, TG, and Hybrid adder circuits in terms of delay, power and power delay product （PDP）. Simulation results reveal that the proposed circuit exhibits lower PDP and is more power efficient and faster when compared with the best available 1-bit full adder circuits. The design is implemented on UMC 0.18 μm process models in Cadence Virtuoso Schematic Composer at 1.8 V single ended supply voltage and simulations are carried out on Spectre S.
基金Supported by the National Native Science Foundation of China
文摘This paper analyses the transient characteristics of high temperature CMOS inverters and gate circuits, and gives the computational formulas of their rise time, fall time and delay time. It may be concluded that the transient characteristics of CMOS inverters and gate circuits deteriorate due to the reduction of carrier mobilities and threshold voltages of MOS transistors and the increase of leakage currents of MOS transistors drain terminal pn junctions. The calculation results can explain the experimental phenomenon.
基金Project supported by the National Natural Science Fund of China
文摘Based on transmission function theory,the synthesis technique for multivaluedCMOS circuits is discussed.By comparing the CMOS circuits based on transmission functiontheory with the T gate,it is shown that their action principles are identical.Based on it,thesynthesis method for multivalued CMOS circuits with many variables by using function decom-position is proposed.
