Near-threshold circuit delay estimation method based on polynomial chaos Kriging element model

A kriging element and circuit delay technology, applied in CAD circuit design, special data processing applications, etc., can solve the problems that sub-threshold circuits cannot meet the requirements of calculation speed, increase leakage power consumption, and weaken energy consumption advantages. Achieve the effects of saving simulation time, improving accuracy, and increasing model accuracy

Active Publication Date: 2021-06-08
SOUTHEAST UNIV
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are some shortcomings in the design of sub-threshold circuits: on the one hand, because the performance of chips near the sub-threshold voltage decreases exponentially compared with conventional voltages, this makes sub-threshold circuits unable to meet the calculation speed requirements of some applications; on the other hand , at extremely low voltage, the leakage power consumption increases greatly and becomes the main source of power consumption at the subthreshold voltage, which weakens the power consumption advantage of the circuit at the subthreshold voltage
As the supply voltage decreases, the delay fluctuation increases under process deviation, and the traditional timing analysis method will cause serious performance loss, which is no longer suitable for the timing analysis of near-threshold circuits under process deviation.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Near-threshold circuit delay estimation method based on polynomial chaos Kriging element model
  • Near-threshold circuit delay estimation method based on polynomial chaos Kriging element model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] Under normal voltage, the delay fluctuation caused by the process deviation is small, and the performance loss caused by the pessimistic estimation of the path delay fluctuation by the traditional timing analysis method is small. However, as the supply voltage decreases, the delay fluctuation increases under the process deviation, and the traditional timing analysis method will cause serious performance loss, which is no longer suitable for the near-threshold circuit timing analysis under the process deviation. Different from traditional timing analysis methods, the present invention's near-threshold circuit delay estimation method based on polynomial chaotic Kriging element model models the delay of the path as a distribution, and determines the delay by solving the distribution and its statistics. time fluctuation magnitude and worst case delay. Compared with the traditional timing analysis method, the statistical timing analysis method used in the present invention h...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses and protects a near-threshold circuit delay estimation method based on a polynomial chaos Kriging element model, which considers the influence of process parameter fluctuation increase under a near-threshold voltage, constructs a model from process parameters to circuit delay through the polynomial chaos Kriging element model, and further performs yield evaluation on circuit path delay. The accurate and effective time sequence analysis method under the near-threshold voltage is realized, and guidance is provided for circuit design. Firstly, a key path of a circuit is extracted through a tool PrimeTime, efficient sampling is carried out on the circuit through Latin hypercube, then delay of the key path is obtained through SPICE simulation under different technological parameter conditions, delay data of non-Gaussian distribution are converted into data conforming to Gaussian distribution through generalized power transformation and maximum likelihood estimation, and a path delay model under low voltage is built by using a polynomial chaos Kriging element model; and finally the time sequence yield evaluation of the circuit is finished.

Description

technical field [0001] The invention belongs to the field of integrated circuit design automation (EDA), in particular to a near-threshold circuit delay estimation method based on a polynomial chaotic Kriging element model. Background technique [0002] Modern Internet of Things and mobile terminal devices have greatly increased the demand for chip processing data volume and standby time. Therefore, how to reduce the chip's data processing power consumption and improve energy efficiency has become a key issue in chip design. Traditional low power consumption technology utilizes the law that power consumption decreases with the square relationship of voltage, and reduces power consumption by reducing the power supply voltage. Subthreshold circuit design greatly reduces circuit power consumption by reducing the supply voltage below the threshold voltage. This method is applied to devices that do not require high performance but require extremely low power consumption, such as...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F30/33
CPCG06F30/33
Inventor 闫浩骈续喜宣城镇刘玉涛时霄宋慧滨
Owner SOUTHEAST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products