CN117095010A - Chiplet data communication transmission authenticity verification method and system - Google Patents

Chiplet data communication transmission authenticity verification method and system Download PDF

Info

Publication number
CN117095010A
CN117095010A CN202311361125.2A CN202311361125A CN117095010A CN 117095010 A CN117095010 A CN 117095010A CN 202311361125 A CN202311361125 A CN 202311361125A CN 117095010 A CN117095010 A CN 117095010A
Authority
CN
China
Prior art keywords
temperature
temperature change
change curve
core particle
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202311361125.2A
Other languages
Chinese (zh)
Other versions
CN117095010B (en
Inventor
王嘉诚
张少仲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhongcheng Hualong Computer Technology Co Ltd
Original Assignee
Zhongcheng Hualong Computer Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhongcheng Hualong Computer Technology Co Ltd filed Critical Zhongcheng Hualong Computer Technology Co Ltd
Priority to CN202311361125.2A priority Critical patent/CN117095010B/en
Publication of CN117095010A publication Critical patent/CN117095010A/en
Application granted granted Critical
Publication of CN117095010B publication Critical patent/CN117095010B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/56—Extraction of image or video features relating to colour
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10048—Infrared image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Quality & Reliability (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The application discloses a method and a system for verifying authenticity of chip data communication transmission, which relate to the field of core particle communication and comprise the following steps: s1: when starting data communication transmission, starting an infrared camera to shoot; s2: pretreatments with trained convolutional neural networksMeasuring the temperature change curve of each core particle; s3: acquiring a temperature change curve of each core particle according to the infrared image; s4: selecting core particle with maximum temperature variation amplitude, comparing the captured temperature variation curve with predicted temperature variation curve, and measuring the difference between the two curvesIf the transmission rate is smaller than the threshold value, the transmission is successful; s5: and (5) ending. The application calculates the comprehensive difference measurementAnd further judging whether the transmission is successful or not, and realizing the automatic verification judgment of the authenticity of the data communication transmission of the core particles.

Description

Chiplet data communication transmission authenticity verification method and system
Technical Field
The application relates to the field of core particle communication, in particular to a chip data communication transmission authenticity verification and system.
Background
With the continuous progress of semiconductor technology, the integration level of chips is higher and higher, but the problems of power consumption and heat dissipation are also serious. Conventional systems on chip (SoC) have grown in core granulation (chip) when faced with process technology bottlenecks. By decomposing the different functional sub-modules into separate core particles and then combining them together by high-speed interconnect technology, the Chiplet technology provides greater flexibility and scalability for chip design.
However, with the widespread use of core pelletization technology, the transfer of data between cores becomes a critical issue. The authenticity and integrity of the data transmission is critical to the stability and performance of the overall system. Conventional verification methods, such as electrical testing and logic verification, while capable of ensuring the authenticity of data to some extent, often require additional hardware support, adding to the complexity and cost of the system.
Furthermore, as technology advances, each die on the chip may operate at different frequencies, voltages, and modes of operation, which makes the power consumption and temperature variations of data transmission more complex. Conventional verification methods have difficulty accurately predicting these changes, resulting in instability and performance degradation of the system.
In order to solve the above problems, researchers have begun to explore new verification methods. The temperature distribution of the chip monitored by the infrared camera is a popular research direction. By analyzing the infrared image, the temperature change between the core particles can be intuitively observed, thereby judging the authenticity and the integrity of data transmission. The method not only can monitor the state of data transmission in real time, but also can provide valuable feedback for the optimization of the system.
In addition, the resolution and accuracy of the infrared image requires the use of high-accuracy camera technology to accurately capture small temperature variations. Secondly, how to combine the infrared image data with other related data, such as power consumption, operating frequency and voltage, etc., for comprehensive analysis is also a challenge, as the temperature distribution of the chip is affected by various factors, such as the size of the core particle, the size of the transmission file, the transmission time, etc. How to overcome these challenges, improving the accuracy and efficiency of verification is an important direction of current research.
Disclosure of Invention
Aiming at the problems mentioned in the prior art, the application provides a verification and system for verifying authenticity of chip data communication transmission, which predicts temperature change curves of each chip by adopting a trained convolutional neural network according to the size of the transmitted data; acquiring temperature change curves of all the core grains according to the infrared images captured by the infrared cameras; comparing the actually captured temperature change curve with the predicted temperature change curve, and calculating the comprehensive difference measure of the actually captured temperature change curve and the predicted temperature change curveAnd further judging whether the transmission is successful or not, and realizing the automatic verification judgment of the authenticity of the data communication transmission of the core particles.
A Chiplet data communication transmission authenticity verification method comprises the following steps:
s1: when the Chiplet starts data communication transmission, an infrared camera is started to shoot the data in real time; simultaneously, parameters of the infrared camera are adjusted according to the ambient temperature and the ambient humidity, and an infrared image is obtained through calibration shooting;
s2: according to the transmitted data size, the working frequency and the voltage of the chip, predicting the temperature change curve of each core chip by adopting a trained convolutional neural network;
s3: acquiring temperature change curves of all the core grains according to the infrared images captured by the infrared cameras;
s4: selecting core particle with maximum temperature variation amplitude obtained by infrared image, and selecting coreComparing the actual captured temperature profile with the predicted temperature profile of the selected core, if the two are measured for a combined differenceIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->If the data transmission rate is greater than the set threshold value, considering that the data transmission is problematic, and retransmitting;
s5: and closing the infrared camera, and finishing verification.
Preferably, the step of predicting the temperature change curve of each core chip by using the trained convolutional neural network according to the size of the transmitted data further comprises the step of training and predicting the convolutional neural network according to multi-mode input characteristics, wherein the multi-mode input characteristics comprise the size of the transmitted data, the size of the core chip, the working frequency, the voltage and the temperature at the previous moment.
Preferably, the infrared image captured by the infrared camera acquires a temperature change curve of each core particle according to the infrared image, the method comprises the steps of carrying out Gaussian filtering on the infrared image, finding out the edges of the core particles in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, and the temperature change curve of each core particle is drawn by using the data sequence.
Preferably, the actually captured temperature change curve is compared with the predicted temperature change curve, and if the difference between the actually captured temperature change curve and the predicted temperature change curve is smaller than a set threshold value, the data transmission is considered to be successful and real; if the phase difference is larger than the set threshold value, considering that the data transmission has a problem, and retransmitting; specifically, the difference between the two curvesThe method comprises the following steps:
wherein,for the actually captured temperature profile, +.>Is a predicted temperature change curve; integral value of difference +.>The method comprises the following steps:
indicating the moment when Chiplet starts data communication transmission,/->The moment when the Chiplet completes data communication transmission is represented;
temperature gradientThe expression is as follows:
integral value of temperature gradientThe expression is as follows:
comprehensive difference metricThe expression is as follows:
、/>the weight coefficients of the integral value of the difference value and the integral value of the temperature gradient are respectively represented.
The application also provides a system for verifying authenticity of the Chiplet data communication transmission, which comprises:
when the Chiplet starts data communication transmission, the infrared camera is started to shoot the data in real time; simultaneously, parameters of the infrared camera are adjusted according to the ambient temperature and the ambient humidity, and an infrared image is obtained through calibration shooting;
the convolutional neural network prediction module predicts the temperature change curve of each core chip by adopting a trained convolutional neural network according to the transmitted data size, the working frequency and the voltage of the chip;
the temperature change curve acquisition module of the core particles acquires the temperature change curve of each core particle according to the infrared image captured by the infrared camera;
comprehensive difference metricThe calculation module selects the core particle with the largest temperature change amplitude obtained by the infrared image, compares the temperature change curve actually captured by the selected core particle with the temperature change curve predicted by the selected core particle, and if the temperature change curve is the comprehensive difference measure +.>If the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two are combined to measure the differenceIf the data transmission rate is greater than the set threshold value, the data transmission is considered to be problematic, and the data transmission is performed againTransmitting;
and the ending module is used for closing the infrared camera and ending the verification.
Preferably, the step of predicting the temperature change curve of each core chip by using the trained convolutional neural network according to the size of the transmitted data further comprises the step of training and predicting the convolutional neural network according to multi-mode input characteristics, wherein the multi-mode input characteristics comprise the size of the transmitted data, the size of the core chip, the working frequency, the voltage and the temperature at the previous moment.
Preferably, the infrared image captured by the infrared camera acquires a temperature change curve of each core particle according to the infrared image, the method comprises the steps of carrying out Gaussian filtering on the infrared image, finding out the edges of the core particles in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, and the temperature change curve of each core particle is drawn by using the data sequence.
Preferably, the actually captured temperature change curve is compared with the predicted temperature change curve, and if the difference between the actually captured temperature change curve and the predicted temperature change curve is smaller than a set threshold value, the data transmission is considered to be successful and real; if the phase difference is larger than the set threshold value, considering that the data transmission has a problem, and retransmitting; specifically, the difference between the two curvesThe method comprises the following steps:
wherein,for the actually captured temperature profile, +.>Is a predicted temperature change curve; integral value of difference +.>The method comprises the following steps:
indicating the moment when Chiplet starts data communication transmission,/->The moment when the Chiplet completes data communication transmission is represented;
temperature gradientThe expression is as follows:
integral value of temperature gradientThe expression is as follows:
comprehensive difference metricThe expression is as follows:
、/>respectively represent the weight of the integral value of the difference and the integral value of the temperature gradientCoefficients.
The application provides a Chiplet data communication transmission authenticity verification method, which has the following beneficial technical effects:
1. the application compares the actual captured temperature change curve with the predicted temperature change curve, if the two are measured by the comprehensive differenceIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->When the power consumption is different, the temperature is different in the communication transmission process according to the difference of the size and the transmission time of the transmission file in the core particle data communication process, so that the direct measurement of hardware is avoided, and the true and false accuracy judgment of the data transmission is realized through the non-contact infrared image measurement.
2. The application is based on the integral value of the difference between two curvesIntegral value of temperature gradient->Combined judgment of temperature comprehensive difference measurement>The reference of multi-dimensional data is realized by combining the temperature difference and considering the temperature difference change rate, so that the accuracy of difference judgment can be greatly improved.
3. According to the application, the temperature change curve of each core chip is predicted by adopting the trained convolutional neural network according to the size of the transmitted data, the convolutional neural network is trained and predicted according to the multi-mode input characteristics, wherein the multi-mode input characteristics comprise the size of the transmitted data, the size of the core chip, the working frequency and the voltage, and the prediction rate and the accuracy of the temperature curve are greatly improved by considering a plurality of characteristics.
4. The application obtains the temperature change curve of each core particle according to the infrared image, comprising Gaussian filtering the infrared image, finding out the edge of the core particle in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, the temperature change curve of each core particle is drawn by using the data sequence, and the drawing speed of the non-contact temperature curve is greatly improved according to the color-temperature mapping table.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of steps of a Chiplet data communication transmission authenticity verification method according to the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Example 1:
in order to solve the above-mentioned problems mentioned in the prior art, as shown in fig. 1: a Chiplet data communication transmission authenticity verification method comprises the following steps:
s1: when the Chiplet starts data communication transmission, an infrared camera is started to shoot the data in real time; simultaneously, parameters of the infrared camera are adjusted according to the ambient temperature and the ambient humidity, and an infrared image is obtained through calibration shooting;
s2: according to the transmitted data size, the working frequency and the voltage of the chip, predicting the temperature change curve of each core chip by adopting a trained convolutional neural network; other input characteristics, such as operating frequency, voltage, temperature at the previous time, etc., can be considered in addition to the size of the data to be transmitted, and these characteristics can be combined into a vector，
Wherein the method comprises the steps ofFor data transmission size, +.>For the working frequency +.>Is voltage, < >>The temperature is the moment before the core particle;
weight of full connection layer, +.>For biasing (I)>Y is the output of the full connection layer for activating the function;
a one-dimensional vector converted for the averaging pooling or flattening operation;
s3: acquiring temperature change curves of all the core grains according to the infrared images captured by the infrared cameras;
1. image preprocessing:
noise removal: a median filter or gaussian filter is used to remove noise from the image.
Contrast enhancement: the contrast of the image is enhanced using histogram equalization or other methods, making the temperature difference more pronounced.
2. Core particle positioning:
edge detection: and (5) using edge detection algorithms such as Sobel, canny and the like to find the edge of the core particle in the image.
Morphological operations: corrosion and expansion operations are used to remove small noise and strengthen the edges of the core.
Contour detection: the outline of the core particles in the image is found, and the position and the size of each core particle are determined.
3. And (3) temperature extraction:
color mapping: the color of an infrared image is typically temperature dependent. We can use a color-temperature mapping table to convert each pixel value in the image to a corresponding temperature value.
Regional analysis: for each located pellet, the average or maximum temperature in its region is calculated.
4. Drawing a temperature change curve:
time series data: the above temperature extraction step is repeated for each core particle in successive infrared images, resulting in a time-temperature data sequence.
And (3) drawing a curve: using the above data sequence, a temperature change curve was plotted for each core.
S4: selecting a temperature for infrared image acquisitionThe core particle with the largest degree variation amplitude is compared with the temperature variation curve actually captured by the selected core particle and the temperature variation curve predicted by the selected core particle, if the two are comprehensively measuredIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->If the data transmission rate is greater than the set threshold value, considering that the data transmission is problematic, and retransmitting;
s5: and closing the infrared camera, and finishing verification.
Preferably, the step of predicting the temperature change curve of each core chip by using the trained convolutional neural network according to the size of the transmitted data further comprises the step of training and predicting the convolutional neural network according to multi-mode input characteristics, wherein the multi-mode input characteristics comprise the size of the transmitted data, the size of the core chip, the working frequency, the voltage and the temperature at the previous moment.
Preferably, the infrared image captured by the infrared camera acquires a temperature change curve of each core particle according to the infrared image, the method comprises the steps of carrying out Gaussian filtering on the infrared image, finding out the edges of the core particles in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, and the temperature change curve of each core particle is drawn by using the data sequence.
Preferably, the actually captured temperature change curve is compared with the predicted temperature change curve, and if the difference between the actually captured temperature change curve and the predicted temperature change curve is smaller than a set threshold value, the data transmission is considered to be successful and real; if the phase difference is larger than the set threshold value, considering that the data transmission has a problem, and retransmitting; specifically, the difference between the two curvesThe method comprises the following steps:
wherein,for the actually captured temperature profile, +.>Is a predicted temperature change curve; integral value of difference +.>The method comprises the following steps:
indicating the moment when Chiplet starts data communication transmission,/->The moment when the Chiplet completes data communication transmission is represented;
temperature gradientThe expression is as follows:
integral value of temperature gradientThe expression is as follows:
comprehensive difference metricThe expression is as follows:
、/>the weight coefficients of the integral value of the difference value and the integral value of the temperature gradient are respectively represented.
The application also provides a system for verifying authenticity of the Chiplet data communication transmission, which comprises:
when the Chiplet starts data communication transmission, the infrared camera is started to shoot the data in real time; simultaneously, parameters of the infrared camera are adjusted according to the ambient temperature and the ambient humidity, and an infrared image is obtained through calibration shooting;
the convolutional neural network prediction module predicts the temperature change curve of each core chip by adopting a trained convolutional neural network according to the transmitted data size, the working frequency and the voltage of the chip;
the temperature change curve acquisition module of the core particles acquires the temperature change curve of each core particle according to the infrared image captured by the infrared camera;
comprehensive difference metricThe calculation module selects the core particle with the largest temperature change amplitude obtained by the infrared image, compares the temperature change curve actually captured by the selected core particle with the temperature change curve predicted by the selected core particle, and if the temperature change curve is the comprehensive difference measure +.>If the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two are combined to measure the differenceIs greater than the designDetermining a threshold value, considering that the data transmission has a problem, and retransmitting;
and the ending module is used for closing the infrared camera and ending the verification.
Preferably, the step of predicting the temperature change curve of each core chip by using the trained convolutional neural network according to the size of the transmitted data further comprises the step of training and predicting the convolutional neural network according to multi-mode input characteristics, wherein the multi-mode input characteristics comprise the size of the transmitted data, the size of the core chip, the working frequency, the voltage and the temperature at the previous moment.
Preferably, the infrared image captured by the infrared camera acquires a temperature change curve of each core particle according to the infrared image, the method comprises the steps of carrying out Gaussian filtering on the infrared image, finding out the edges of the core particles in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, and the temperature change curve of each core particle is drawn by using the data sequence.
When we use 3 ℃ as the scale variation range of temperature, we can assign a specific RGB value for each 3 ℃ temperature range. The following is a more detailed example of a color-temperature mapping table, where colors are represented using RGB values:
preferably, the actually captured temperature change curve is compared with the predicted temperature change curve, and if the difference between the actually captured temperature change curve and the predicted temperature change curve is smaller than a set threshold value, the data transmission is considered to be successful and real; if the phase difference is larger than the set threshold value, considering that the data transmission has a problem, and retransmitting; specifically, the difference between the two curvesThe method comprises the following steps:
wherein,for the actually captured temperature profile, +.>Is a predicted temperature change curve; integral value of difference +.>The method comprises the following steps:
indicating the moment when Chiplet starts data communication transmission,/->The moment when the Chiplet completes data communication transmission is represented;
temperature gradientThe expression is as follows:
integral value of temperature gradientThe expression is as follows:
comprehensive difference metricThe expression is as follows:
、/>the weight coefficients of the integral value of the difference value and the integral value of the temperature gradient are respectively represented.
The application provides a method and a system for verifying authenticity of Chiplet data communication transmission, which can realize the following beneficial technical effects:
1. the application compares the actual captured temperature change curve with the predicted temperature change curve, if the two are measured by the comprehensive differenceIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->When the power consumption is different, the temperature is different in the communication transmission process according to the difference of the size and the transmission time of the transmission file in the core particle data communication process, so that the direct measurement of hardware is avoided, and the true and false accuracy judgment of the data transmission is realized through the non-contact infrared image measurement.
2. The application is based on the integral value of the difference between two curvesIntegral value of temperature gradient->Combined judgment of temperature comprehensive difference measurement>I.e. by difference in temperature and taking into account the temperature at the same timeThe combination of the difference change rates of the degrees realizes the reference of multi-dimensional data, and can greatly improve the accuracy of difference judgment.
3. According to the application, the temperature change curve of each core chip is predicted by adopting the trained convolutional neural network according to the size of the transmitted data, the convolutional neural network is trained and predicted according to the multi-mode input characteristics, wherein the multi-mode input characteristics comprise the size of the transmitted data, the size of the core chip, the working frequency and the voltage, and the prediction rate and the accuracy of the temperature curve are greatly improved by considering a plurality of characteristics.
4. The application obtains the temperature change curve of each core particle according to the infrared image, comprising Gaussian filtering the infrared image, finding out the edge of the core particle in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, the temperature change curve of each core particle is drawn by using the data sequence, and the drawing speed of the non-contact temperature curve is greatly improved according to the color-temperature mapping table.
The foregoing describes in detail a method and a system for verifying authenticity of a chip data communication transmission, and specific examples are applied to illustrate the principle and implementation of the present application, and the above description of the examples is only used to help understand the core idea of the present application; also, as will be apparent to those skilled in the art in light of the present teachings, the present disclosure should not be limited to the specific embodiments and applications described herein.

Claims (10)

1. The Chiplet data communication transmission authenticity verification method is characterized by comprising the following steps of:
s1: when the Chiplet starts data communication transmission, an infrared camera is started to shoot the data in real time; simultaneously, parameters of the infrared camera are adjusted according to the ambient temperature and the ambient humidity, and an infrared image is obtained through calibration shooting;
s2: according to the transmitted data size, the working frequency and the voltage of the chip, predicting the temperature change curve of each core chip by adopting a trained convolutional neural network;
s3: acquiring temperature change curves of all the core grains according to the infrared images captured by the infrared cameras;
s4: selecting core particle with maximum temperature variation amplitude obtained by infrared image, comparing the temperature variation curve actually captured by the selected core particle with the temperature variation curve predicted by the selected core particle, if the two are combined with each other to obtain a difference measureIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->If the data transmission rate is greater than the set threshold value, considering that the data transmission is problematic, and retransmitting;
s5: and closing the infrared camera, and finishing verification.
2. The method for verifying authenticity of chip data communication transmission according to claim 1, wherein the step of predicting the temperature change curve of each chip by using a trained convolutional neural network according to the size of the transmitted data, and the step of training and predicting the convolutional neural network according to a multi-mode input characteristic, wherein the multi-mode input characteristic comprises the size of the transmitted data, the size of the chip, the working frequency, the voltage and the temperature of the chip at the previous moment.
3. The method for verifying authenticity of chip data communication transmission according to claim 1, wherein the acquiring the temperature change curve of each core particle according to the infrared image by the infrared image captured by the infrared camera comprises performing gaussian filtering on the infrared image, finding out the edge of the core particle in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, and the temperature change curve of each core particle is drawn by using the data sequence.
4. The method for verifying authenticity of chip data communication transmission according to claim 1, wherein comparing the actual captured temperature change curve of the selected core with the predicted temperature change curve of the selected core comprisesThe method comprises the following steps:
；
wherein,for the actually captured temperature profile, +.>Is a predicted temperature change curve; integral value of difference +.>The method comprises the following steps:
；
indicating the moment when Chiplet starts data communication transmission,/->The moment when the Chiplet completes data communication transmission is represented;
temperature gradientThe expression is as follows:
；
integral value of temperature gradientThe expression is as follows:
；
comprehensive difference metricThe expression is as follows:
；
、/>weight coefficients respectively representing the integral value of the difference value and the integral value of the temperature gradient;
if the two are combined to measure the differenceIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->If the data transmission rate is greater than the set threshold value, the data transmission is considered to be problematic, and the data transmission is carried out again.
5. The method for verifying authenticity of a chip data communication transmission according to claim 1, wherein in S3, an infrared image is captured by a micro-focal infrared camera.
6. A Chiplet data communication transmission authenticity verification system, comprising:
when the Chiplet starts data communication transmission, the infrared camera is started to shoot the data in real time; simultaneously, parameters of the infrared camera are adjusted according to the ambient temperature and the ambient humidity, and an infrared image is obtained through calibration shooting;
the convolutional neural network prediction module predicts the temperature change curve of each core particle Chiplet by adopting a trained convolutional neural network according to the size of transmitted data;
the temperature change curve acquisition module of the core particles acquires the temperature change curve of each core particle according to the infrared image captured by the infrared camera;
comprehensive difference metricThe calculation module selects the core particle with the largest temperature change amplitude obtained by the infrared image, compares the temperature change curve actually captured by the selected core particle with the temperature change curve predicted by the selected core particle, and if the temperature change curve is the comprehensive difference measure +.>If the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->If the data transmission rate is greater than the set threshold value, considering that the data transmission is problematic, and retransmitting;
and the ending module is used for closing the infrared camera and ending the verification.
7. The system for verifying authenticity of a chip data communication transmission according to claim 6, wherein said predicting a temperature change curve of each chip using a trained convolutional neural network according to a size of data transmitted, further comprises training the convolutional neural network according to a multi-modal input characteristic, the multi-modal input characteristic including a size of data transmitted, a size of a chip, a frequency of operation, a voltage, and a temperature at a previous time.
8. The system for verifying authenticity of a chip data communication transmission according to claim 6, wherein the infrared image captured by the infrared camera is used for obtaining a temperature change curve of each core particle according to the infrared image, the system comprises the steps of performing Gaussian filtering on the infrared image, finding out the edges of the core particles in the image by adopting a Canny edge detection algorithm, and determining the position and the size of each core particle; according to the color-temperature mapping table, each pixel value in the image is converted into a corresponding temperature value, the temperature extraction step is repeated for each core particle, a time-temperature data sequence is obtained, and the temperature change curve of each core particle is drawn by using the data sequence.
9. The system for verifying authenticity of a chip data communication transmission according to claim 6, wherein the temperature change curve actually captured by the selected core particle is compared with the temperature change curve predicted by the selected core particle, and specifically comprises a difference value between the two curvesThe method comprises the following steps:
；
wherein,for the actually captured temperature profile, +.>Is a predicted temperature change curve; integral value of difference +.>The method comprises the following steps:
；
indicating the moment when Chiplet starts data communication transmission,/->The moment when the Chiplet completes data communication transmission is represented;
temperature gradientThe expression is as follows:
；
integral value of temperature gradientThe expression is as follows:
；
comprehensive difference metricThe expression is as follows:
；
、/>weight coefficients respectively representing the integral value of the difference value and the integral value of the temperature gradient;
if the two are combined to measure the differenceIf the data transmission rate is smaller than the set threshold value, the data transmission is considered to be successful and real; if the two comprehensive difference measures->If the data transmission rate is greater than the set threshold value, the data transmission is considered to be problematic, and the data transmission is carried out again.
10. The system for verifying authenticity of a chip data communication transmission according to claim 6, wherein an infrared image is captured by a micro-focal infrared camera in the temperature change curve acquisition module of the core particle.
CN202311361125.2A 2023-10-20 2023-10-20 Chiplet data communication transmission authenticity verification method and system Active CN117095010B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311361125.2A CN117095010B (en) 2023-10-20 2023-10-20 Chiplet data communication transmission authenticity verification method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311361125.2A CN117095010B (en) 2023-10-20 2023-10-20 Chiplet data communication transmission authenticity verification method and system

Publications (2)

Publication Number Publication Date
CN117095010A true CN117095010A (en) 2023-11-21
CN117095010B CN117095010B (en) 2023-12-19

Family

ID=88783340

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311361125.2A Active CN117095010B (en) 2023-10-20 2023-10-20 Chiplet data communication transmission authenticity verification method and system

Country Status (1)

Country Link
CN (1) CN117095010B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111174917A (en) * 2019-12-26 2020-05-19 国网山东省电力公司莱芜供电公司 Online temperature measurement camera shooting alarm system of electricity distribution room
CN115552355A (en) * 2020-06-26 2022-12-30 英特尔公司 Performance level control in a data processing apparatus
US20230122214A1 (en) * 2021-10-19 2023-04-20 Johnson & Johnson Vision Care, Inc. Defect detection using synthetic data and machine learning
CN116054405A (en) * 2023-01-29 2023-05-02 保定云鹰能源科技有限公司 Double-light monitoring analysis system of photovoltaic power station based on computer vision and machine learning
CN116401137A (en) * 2023-06-06 2023-07-07 中诚华隆计算机技术有限公司 Core particle health state prediction method and device, electronic equipment and storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111174917A (en) * 2019-12-26 2020-05-19 国网山东省电力公司莱芜供电公司 Online temperature measurement camera shooting alarm system of electricity distribution room
CN115552355A (en) * 2020-06-26 2022-12-30 英特尔公司 Performance level control in a data processing apparatus
US20230122214A1 (en) * 2021-10-19 2023-04-20 Johnson & Johnson Vision Care, Inc. Defect detection using synthetic data and machine learning
CN116054405A (en) * 2023-01-29 2023-05-02 保定云鹰能源科技有限公司 Double-light monitoring analysis system of photovoltaic power station based on computer vision and machine learning
CN116401137A (en) * 2023-06-06 2023-07-07 中诚华隆计算机技术有限公司 Core particle health state prediction method and device, electronic equipment and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
杨晖;: "后摩尔时代Chiplet技术的演进与挑战", 集成电路应用, no. 05 *

Also Published As

Publication number Publication date
CN117095010B (en) 2023-12-19

Similar Documents

Publication Publication Date Title
US10929945B2 (en) Image capture devices featuring intelligent use of lightweight hardware-generated statistics
US9020195B2 (en) Object tracking device, object tracking method, and control program
Zhang et al. Multi-class weather classification on single images
Zakaria et al. Object shape recognition in image for machine vision application
JP2004062605A (en) Scene identification method and device, and program
Shuo et al. Object tracking method based on SURF
CN111127384A (en) Strong reflection workpiece vision measurement method based on polarization imaging
Liu et al. Iterating tensor voting: A perceptual grouping approach for crack detection on EL images
Hua et al. Pedestrian-and vehicle-detection algorithm based on improved aggregated channel features
CN101320477B (en) Human body tracing method and equipment thereof
Song et al. Blind image quality assessment for authentic distortions by intermediary enhancement and iterative training
CN117095010B (en) Chiplet data communication transmission authenticity verification method and system
CN106373177A (en) Design method used for optimizing image scene illumination estimation
Song et al. A Single Image Dehazing Method Based on End-to-End CPAD-Net Network in Deep Learning Environment
Tan et al. Gesture segmentation based on YCb'Cr'color space ellipse fitting skin color modeling
Yang et al. Contrast limited adaptive histogram equalization for an advanced stereo visual slam system
Prakhya et al. On creating low dimensional 3D feature descriptors with PCA
Tsai et al. Deep Learning Based AOI System with Equivalent Convolutional Layers Transformed from Fully Connected Layers
CN113793372A (en) Optimal registration method and system for different-source images
CN110532892B (en) Method for detecting road vanishing point of single image of unstructured road
Huang et al. Multi-task deep CNN model for no-reference image quality assessment on smartphone camera photos
He et al. Research on Digital Image Intelligent Recognition Method for Industrial Internet of Things Production Data Acquisition.
Ye et al. BIVS: block image and voting strategy for weather image classification
Zhang et al. A unified saliency detection framework for visible and infrared images
Cheng et al. Research on Fast Target Detection And Classification Algorithm for Passive Millimeter Wave Imaging

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant