In this paper, we propose a technique for lowering the latency of the communication in a NoC （network on chip）. The technique, which can support two qualities of service （QoS）, i.e., the guaranteed throughput （GT...In this paper, we propose a technique for lowering the latency of the communication in a NoC （network on chip）. The technique, which can support two qualities of service （QoS）, i.e., the guaranteed throughput （GT） and best effort （BE）, is based on splitting a wider link into narrower links to increase throughput and decrease latency in the NoC. In addition, to ease the synchronization and reduce the crosstalk, we use the l-of-4 encoding for the smaller buses. The use of the encoding in the proposed NoC architecture considerably lowers the latency for both BE and GT packets. In addition, the bandwidth is increased while the power consumption of the links is reduced.展开更多
We propose an accurate model to describe the I-V characteristics of a sub-90-nm metal-oxide-semiconductor field-effect transistor(MOSFET) in the linear and saturation regions for fast analytical calculation of the cur...We propose an accurate model to describe the I-V characteristics of a sub-90-nm metal-oxide-semiconductor field-effect transistor(MOSFET) in the linear and saturation regions for fast analytical calculation of the current.The model is based on the BSIM3v3 model.Instead of using constant threshold voltage and early voltage,as is assumed in the BSIM3v3 model,we define these voltages as functions of the gate-source voltage.The accuracy of the model is verified by comparison with HSPICE for the 90-,65-,45-,and 32-nm CMOS technologies.The model shows better accuracy than the nth-power and BSIM3v3 models.Then,we use the proposed I-V model to calculate the read static noise margin(SNM) of nano-scale conventional 6T static random-access memory(SRAM) cells with high accuracy.We calculate the read SNM by approximating the inverter transfer voltage characteristic of the cell in the regions where vertices of the maximum square of the butterfly curves are placed.The results for the SNM are also in excellent agreement with those of the HSPICE simulation for 90-,65-,45-,and 32-nm technologies.Verification in the presence of process variations and negative bias temperature instability(NBTI) shows that the model can accurately predict the minimum supply voltage required for a target yield.展开更多
We investigate the use of two integer inversion algorithms,a modified Montgomery modulo inverse and a Fermat's Little Theorem based inversion,in a prime-field affine-coordinate elliptic-curve crypto-processor.To p...We investigate the use of two integer inversion algorithms,a modified Montgomery modulo inverse and a Fermat's Little Theorem based inversion,in a prime-field affine-coordinate elliptic-curve crypto-processor.To perform this,we present a low-power/energy GF(p) affine-coordinate elliptic-curve cryptography(ECC) processor design with a simplified architecture and complete flexibility in terms of the field and curve parameters.The design can use either of the inversion algorithms.Based on the implementations of this design for 168-,192-,and 224-bit prime fields using a standard 0.13 μm CMOS technology,we compare the efficiency of the algorithms in terms of power/energy consumption,area,and calculation time.The results show that while the Fermat's theorem approach is not appropriate for the affine-coordinate ECC processors due to its long computation time,the Montgomery modulo inverse algorithm is a good candidate for low-energy implementations.The results also show that the 168-bit ECC processor based on the Montgomery modulo inverse completes one scalar multiplication in only 0.4 s at a 1 MHz clock frequency consuming only 12.92 μJ,which is lower than the reported values for similar designs.展开更多
We propose a modeling methodology for both leakage power consumption and delay of basic CMOS digital gates in the presence of threshold voltage and mobility variations. The key parameters in determining the leakage an...We propose a modeling methodology for both leakage power consumption and delay of basic CMOS digital gates in the presence of threshold voltage and mobility variations. The key parameters in determining the leakage and delay are OFF and ON currents, respectively, which are both affected by the variation of the threshold voltage. Additionally, the current is a strong function of mobility. The proposed methodology relies on a proper modeling of the threshold voltage and mobility variations, which may be induced by any source. Using this model, in the plane of threshold voltage and mobility, we determine regions for different combinations of performance (speed) and leakage. Based on these regions, we discuss the trade-off between leakage and delay where the leakage-delay-product is the optimization objective. To assess the accuracy of the proposed model, we compare its predictions with those of HSPICE simulations for both basic digital gates and ISCAS85 benchmark circuits in 45-, 65-, and 90-nm technologies.展开更多
基金Project supported by the Iranian National Science Foundation
文摘In this paper, we propose a technique for lowering the latency of the communication in a NoC （network on chip）. The technique, which can support two qualities of service （QoS）, i.e., the guaranteed throughput （GT） and best effort （BE）, is based on splitting a wider link into narrower links to increase throughput and decrease latency in the NoC. In addition, to ease the synchronization and reduce the crosstalk, we use the l-of-4 encoding for the smaller buses. The use of the encoding in the proposed NoC architecture considerably lowers the latency for both BE and GT packets. In addition, the bandwidth is increased while the power consumption of the links is reduced.
文摘We propose an accurate model to describe the I-V characteristics of a sub-90-nm metal-oxide-semiconductor field-effect transistor(MOSFET) in the linear and saturation regions for fast analytical calculation of the current.The model is based on the BSIM3v3 model.Instead of using constant threshold voltage and early voltage,as is assumed in the BSIM3v3 model,we define these voltages as functions of the gate-source voltage.The accuracy of the model is verified by comparison with HSPICE for the 90-,65-,45-,and 32-nm CMOS technologies.The model shows better accuracy than the nth-power and BSIM3v3 models.Then,we use the proposed I-V model to calculate the read static noise margin(SNM) of nano-scale conventional 6T static random-access memory(SRAM) cells with high accuracy.We calculate the read SNM by approximating the inverter transfer voltage characteristic of the cell in the regions where vertices of the maximum square of the butterfly curves are placed.The results for the SNM are also in excellent agreement with those of the HSPICE simulation for 90-,65-,45-,and 32-nm technologies.Verification in the presence of process variations and negative bias temperature instability(NBTI) shows that the model can accurately predict the minimum supply voltage required for a target yield.
基金supported in part by the Iran Telecommunication Research Center (ITRC) and the Research Council of University of Tehran
文摘We investigate the use of two integer inversion algorithms,a modified Montgomery modulo inverse and a Fermat's Little Theorem based inversion,in a prime-field affine-coordinate elliptic-curve crypto-processor.To perform this,we present a low-power/energy GF(p) affine-coordinate elliptic-curve cryptography(ECC) processor design with a simplified architecture and complete flexibility in terms of the field and curve parameters.The design can use either of the inversion algorithms.Based on the implementations of this design for 168-,192-,and 224-bit prime fields using a standard 0.13 μm CMOS technology,we compare the efficiency of the algorithms in terms of power/energy consumption,area,and calculation time.The results show that while the Fermat's theorem approach is not appropriate for the affine-coordinate ECC processors due to its long computation time,the Montgomery modulo inverse algorithm is a good candidate for low-energy implementations.The results also show that the 168-bit ECC processor based on the Montgomery modulo inverse completes one scalar multiplication in only 0.4 s at a 1 MHz clock frequency consuming only 12.92 μJ,which is lower than the reported values for similar designs.
文摘We propose a modeling methodology for both leakage power consumption and delay of basic CMOS digital gates in the presence of threshold voltage and mobility variations. The key parameters in determining the leakage and delay are OFF and ON currents, respectively, which are both affected by the variation of the threshold voltage. Additionally, the current is a strong function of mobility. The proposed methodology relies on a proper modeling of the threshold voltage and mobility variations, which may be induced by any source. Using this model, in the plane of threshold voltage and mobility, we determine regions for different combinations of performance (speed) and leakage. Based on these regions, we discuss the trade-off between leakage and delay where the leakage-delay-product is the optimization objective. To assess the accuracy of the proposed model, we compare its predictions with those of HSPICE simulations for both basic digital gates and ISCAS85 benchmark circuits in 45-, 65-, and 90-nm technologies.
