Methods, systems, and devices for detecting side-channel attacks on security chips; security chips.

By incorporating a multi-sensor acquisition module and an attack identification model into the security chip, a fusion feature vector is generated to identify the type and intensity of side-channel attacks. This solves the problems of insufficient detection coverage and overreaction in existing technologies, achieving higher identification accuracy and ensuring normal use.

CN123069463APending Publication Date: 2026-11-10BEIJING TONGFANG MICROELECTRONICS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611168143.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-03
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

When faced with multi-dimensional and complex attack methods, existing technologies have insufficient coverage for side-channel attack detection in security chips, low identification accuracy, and a tendency to overreact to minor attacks, affecting normal use.

Method used

A multi-sensor acquisition module is set up inside the security chip to collect multi-dimensional signal parameters. After preprocessing, a fusion feature vector is generated, and an attack identification model is used to determine whether a side-channel attack exists, its type and strength level, and to execute corresponding protection response measures.

Benefits of technology

It improves the side-channel attack identification accuracy of security chips in complex attack environments, avoids overreaction under minor attacks, and ensures normal chip operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123069463A_ABST
    Figure CN123069463A_ABST
Patent Text Reader

Abstract

This application relates to the field of chip attack detection technology, and discloses a method, system, and device for detecting side-channel attacks on security chips, as well as a security chip itself. The security chip internally incorporates a multi-sensor acquisition module. The detection method includes: receiving multi-dimensional signal parameters of the security chip acquired by the multi-sensor acquisition module; preprocessing the multi-dimensional signal parameters to generate a fused feature vector; inputting the fused feature vector into an attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level; and, in the case of a side-channel attack, causing the security chip to execute protective response measures corresponding to the attack type and attack strength level. This application can improve the accuracy of side-channel attack identification in complex attack environments while avoiding overreaction by the security chip under minor attacks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of chip attack detection technology, such as a detection method, detection system and device, and security chip for detecting side-channel attacks on security chips. Background Technology

[0002] Currently, security chips, as a core carrier in the field of information security, are widely used in critical scenarios such as financial payments, identity authentication, data encryption, and IoT terminals. The security of security chips directly determines the security of the storage, transmission, and computation of sensitive information. Among these threats, side-channel attacks are one of the main security threats facing the security chip field. Attackers monitor the physical leakage information of security chips during operation and, combined with statistical analysis methods, deduce sensitive information such as the chip's keys, thereby breaking the chip's security protection mechanisms.

[0003] In related technologies, side-channel attack detection and protection schemes for security chips typically involve collecting a certain signal parameter (e.g., power consumption or temperature) during the operation of the security chip and comparing this parameter with a benchmark value to determine whether a side-channel attack exists. If a side-channel attack is detected, the security chip will be shut down directly.

[0004] In the process of implementing the embodiments of this disclosure, it was found that the related technology has at least the following problems: Existing solutions suffer from insufficient attack detection coverage and inaccurate attack identification when facing multi-dimensional and complex attack methods. Furthermore, the immediate shutdown of the security chip upon detecting a side-channel attack indicates an overreaction under minor attacks, impacting the chip's normal operation. Therefore, improving the accuracy of side-channel attack identification in complex attack environments while avoiding overreaction under minor attacks has become a pressing technical challenge.

[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments, but rather as a prelude to the detailed description that follows.

[0007] This disclosure provides a method, system, and apparatus for detecting side-channel attacks on security chips, as well as a security chip itself. It can improve the accuracy of side-channel attack identification in complex attack environments while preventing the security chip from overreacting to minor attacks.

[0008] In some embodiments, the security chip is internally equipped with a multi-sensor acquisition module for a method of detecting side-channel attacks on the security chip, including: receiving multi-dimensional signal parameters of the security chip acquired by the multi-sensor acquisition module; preprocessing the multi-dimensional signal parameters to generate a fused feature vector; inputting the fused feature vector into an attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level; and, if a side-channel attack exists, causing the security chip to execute protective response measures corresponding to the attack type and attack strength level.

[0009] Optionally, the multidimensional signal parameters include: the external power supply voltage of the security chip, the internal core voltage domain voltage of the security chip, the temperature of the packaging surface or core die of the security chip, the light intensity received by the security chip, the external input clock frequency of the security chip, the external power supply glitches of the security chip, the internal power supply glitches of the security chip, the current consumption of the entire or some functional modules of the security chip, and the packaging integrity of the security chip.

[0010] Optionally, the multidimensional signal parameters are preprocessed to generate a fused feature vector, including: performing time synchronization processing on the multidimensional signal parameters; and performing min-max normalization processing on the time-synchronized multidimensional signal parameters to generate a fused feature vector.

[0011] Alternatively, the expression for min-max normalization is as follows:

[0012] in These are the original signal parameters. , These are the historical extreme values ​​of this signal parameter. These are the normalized eigenvalues.

[0013] Optionally, the security chip is made to perform protection response measures corresponding to the attack type and attack strength level, including: calculating a strength score based on the attack type and attack strength level; and performing target protection response measures corresponding to the strength score.

[0014] Optionally, target protection response measures corresponding to the strength score are executed, including: when the strength score is in the low score range, logging, internal alarms and dynamic noise injection are executed; when the strength score is in the medium score range, non-core functions are suspended, key fragmentation migration and access permission isolation are executed; when the strength score is in the high score range, sensitive data erasure and chip security locking are executed.

[0015] Optionally, the detection method further includes: when the strength score is greater than or equal to the lower limit of the intermediate score range, recording the multidimensional signal parameters of the side-channel attack, the attack type, the attack strength level, and the target protection response measures implemented, and generating an attack tracing report; storing the attack tracing report in a non-volatile secure storage area.

[0016] In some embodiments, the security chip includes a multi-sensor acquisition module for detecting side-channel attacks on the security chip, comprising: a signal acquisition module for receiving multi-dimensional signal parameters of the security chip acquired by the multi-sensor acquisition module; a data preprocessing module for preprocessing the multi-dimensional signal parameters to generate a fused feature vector; an attack identification module for inputting the fused feature vector into an attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level; and a graded response module for enabling the security chip to execute protective response measures corresponding to the attack type and attack strength level in the event of a side-channel attack.

[0017] In some embodiments, a detection device for security chip side-channel attacks includes a processor and a memory storing program instructions, the processor being configured to execute the detection device for security chip side-channel attacks as described above when the program instructions are executed.

[0018] In some embodiments, a security chip includes: a chip body; and a detection system for security chip side-channel attacks as described above, or a detection device for security chip side-channel attacks as described above, integrated inside the chip body.

[0019] The detection method, detection system, and apparatus for side-channel attacks on security chips provided in this disclosure can achieve the following technical effects: In this embodiment, during the operation of the security chip, it first receives multi-dimensional signal parameters collected by a multi-sensor acquisition module and preprocesses these parameters into a fused feature vector. This fused feature vector is then input into an attack identification model to determine the presence of a side-channel attack, as well as the attack type and intensity level. Compared to related technologies that only compare a single signal parameter, this approach allows for the detection of attack traces in other dimensions even when a single signal parameter is spoofed, improving the accuracy of side-channel attack identification. After confirming the existence of a side-channel attack, the security chip is controlled to execute protective response measures corresponding to the attack type and intensity level, ensuring uninterrupted operation even under minor side-channel attacks. Therefore, this embodiment improves the accuracy of side-channel attack identification in complex attack environments while preventing overreaction by the security chip under minor attacks.

[0020] The above general description and the description below are exemplary and illustrative only and are not intended to limit this application. Attached Figure Description

[0021] One or more embodiments are illustrated by way of example with reference to the accompanying drawings. These illustrations and drawings do not constitute a limitation on the embodiments. Elements having the same reference numerals in the drawings are shown as similar elements. The drawings are not to be scaled. And wherein: Figure 1 This is a schematic diagram of a security chip provided in an embodiment of this disclosure; Figure 2 This is a schematic diagram of another security chip provided in an embodiment of this disclosure; Figure 3 This is a schematic diagram of a detection method for side-channel attacks on security chips provided in an embodiment of this disclosure; Figure 4 This is a schematic diagram of another detection method for side-channel attacks on security chips provided in this disclosure embodiment; Figure 5 This is a schematic diagram of another detection method for side-channel attacks on security chips provided in this disclosure embodiment; Figure 6 This is a schematic diagram of a detection system for side-channel attacks on security chips provided in an embodiment of this disclosure; Figure 7 This is a schematic diagram of a detection device for side-channel attacks on security chips provided in an embodiment of this disclosure. Detailed Implementation

[0022] To provide a more detailed understanding of the features and technical content of the embodiments of this disclosure, the implementation of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for illustrative purposes only and are not intended to limit the embodiments of this disclosure. In the following technical description, for ease of explanation, several details are used to provide a full understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and devices may be simplified in their depiction to simplify the drawings.

[0023] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this disclosure described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0024] Unless otherwise stated, the term "multiple" means two or more.

[0025] In this embodiment of the disclosure, the character " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B means: A or B.

[0026] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0027] The term "correspondence" can refer to an association or binding relationship. The correspondence between A and B means that there is an association or binding relationship between A and B.

[0028] It should be noted that, unless otherwise specified, the embodiments and features described in the present disclosure can be combined with each other.

[0029] The terminal device provided in this disclosure requires near-field communication (NFC) functionality and can be an electronic device such as a mobile phone, tablet, or smartwatch. The following description uses a mobile phone as an example to illustrate the terminal device.

[0030] Combination Figure 1 As shown, this embodiment of the disclosure provides a security chip 100, including: a chip body 110 and a detection system 600 or a detection device 700 for security chip side-channel attacks. The detection system 600 or the detection device 700 for security chip side-channel attacks is integrated within the chip body 110.

[0031] Optionally, the security chip 100 also includes a multi-sensor acquisition module 120 for acquiring multi-dimensional signal parameters.

[0032] Optionally, the multi-sensor acquisition module includes: an external voltage detection unit (EVD), an internal voltage detection unit (IVD), a temperature detection unit (TD), a digital light detection unit (LD), a frequency detection unit (FD), an external power supply glitch detection unit (EGS), an internal power supply glitch detection unit (IGS), a current detection unit (ID), and a physical anti-tamper detection unit (ASH).

[0033] Specifically, the external voltage detection unit is used to acquire the external power supply voltage of the safety chip through the ANATOP analog front end. The acquisition frequency is 100Hz, and the voltage value and voltage fluctuation rate ΔV / Δt are recorded in real time. When the voltage exceeds the normal range of [2.7V, 3.6V], an interrupt is triggered and reported.

[0034] Specifically, the internal voltage detection unit is used to collect the voltage of the internal core voltage domain of the security chip, simultaneously record the difference characteristics between the core voltage and the external voltage, and identify internal voltage coercion attacks.

[0035] Specifically, the temperature detection unit is used to collect the temperature of the packaging surface or core die of the security chip. The collection frequency is 50Hz, and the normal range is [-40℃, 85℃]. If the temperature exceeds the range, an abnormality flag will be triggered.

[0036] Specifically, there are multiple digital light detection units, ranging from dozens to hundreds, deployed in key areas of the security chip such as storage, security registers, and algorithm modules. The sampling frequency is 50Hz, used to identify the light intensity received by the security chip in order to identify laser fault injection attacks.

[0037] Specifically, the frequency detection unit is used to sample the external input clock frequency of the security chip using an internal high-precision 100MHz crystal oscillator, and an anomaly is marked when the external clock frequency changes beyond ±10%.

[0038] Specifically, the external power supply glitch detection unit and the internal power supply glitch detection unit are used to detect external and internal power supply glitch through the ANATOP power monitoring module. When the glitch amplitude exceeds 50mV and the duration exceeds 10ns, an abnormality report is triggered.

[0039] Specifically, the current detection unit is used to collect the current consumption of the entire or part of the functional modules of the security chip. The sampling frequency is 100Hz, and abnormal current jumps can indicate power consumption side-channel attacks.

[0040] Specifically, the physical anti-tamper detection unit is used to monitor the packaging integrity of the security chip, the connectivity of the top metal shielding mesh, and probe intrusion signals. When physical intrusion behaviors such as opening the security chip, micro probe contact, and FIB modification are detected, an interrupt report is triggered.

[0041] For example, the arrangement of each detection unit in the multi-sensor acquisition module within the chip is as follows: Figure 2 As shown.

[0042] Optionally, the detection device 700 for side-channel attacks on security chips includes a processor. The processor can receive multi-dimensional signal parameters of the security chip acquired by a multi-sensor acquisition module and preprocess the multi-dimensional signal parameters into a fused feature vector. After generating the fused feature vector, the processor can input the fused feature vector into a pre-built attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level. After determining that a side-channel attack exists, the processor can execute protective response measures corresponding to the attack type and attack strength level, and record the multi-dimensional signal parameters of the side-channel attack, the attack type, the attack strength level, and the executed target protective response measures, generating an attack tracing report; the attack tracing report is stored in a non-volatile secure storage area.

[0043] In conjunction with the aforementioned security chip, this disclosure provides a method for detecting side-channel attacks on security chips, wherein the processor described above can be the executing entity of this detection method. Figure 3 As shown, the detection method includes: S301, the processor receives multi-dimensional signal parameters from the security chip acquired by the multi-sensor acquisition module.

[0044] Specifically, the security chip is equipped with a multi-sensor acquisition module that integrates various types of sensors. After the security chip is powered on, the processor can receive the multi-dimensional signal parameters of the security chip acquired by the multi-sensor acquisition module.

[0045] Optionally, the multidimensional signal parameters include at least two of the following parameters: the external power supply voltage of the security chip, the internal core voltage domain voltage of the security chip, the temperature of the package surface or core die of the security chip, the light intensity received by the security chip, the external input clock frequency of the security chip, the external power supply glitches of the security chip, the internal power supply glitches of the security chip, the current consumption of the entire or part of the functional modules of the security chip, and the package integrity of the security chip.

[0046] Specifically, the multidimensional signal parameters include four types of characteristics: the detected values ​​of each sensor, the rate of change, the duration, and the number of abnormal triggers. In addition, the processor will also coordinate with the interrupt controller and timer of the APB bus to timestamp each frame of acquired data, ensuring the time-series traceability of the data.

[0047] S302, the processor preprocesses the multidimensional signal parameters to generate a fused feature vector.

[0048] Specifically, by preprocessing the multidimensional signal parameters to generate a fused feature vector, the temporal deviation and dimensional difference between the multidimensional signal parameters can be eliminated, and the original physical signal can be converted into a feature vector that can be directly processed by the attack recognition model.

[0049] In addition, after generating the fused feature vector, the multi-dimensional signal parameters and the generated fused feature vector are also transmitted in batches to the RAM0 secure storage area inside the security chip, so that when a side-channel attack is determined to exist, the data can be used for verification or tracing.

[0050] The S303 processor will integrate the feature vector input attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level.

[0051] Specifically, the attack detection model is a neural network model built based on a lightweight CNN-LSTM hybrid deep learning algorithm. The structural design of the attack detection model is as follows: First, the CNN layer extracts spatial features from the input multi-dimensional fused feature vector, capturing the correlation patterns between multi-dimensional signal parameters through convolutional kernels. For example, a laser fault injection attack can simultaneously cause synchronous abnormal jumps in multiple LD signals, IVD voltage fluctuations, and ID current spikes. The CNN layer can effectively extract the cooperative abnormal patterns between these multi-dimensional signal parameters, and its output is a high-dimensional spatial feature map. Then, the LSTM layer performs temporal modeling on the spatial features output by the CNN, using its gating mechanism to capture the dynamic changing trends of sensor data within a continuous time window. For example, the gradual enhancement or sudden jump features of the attack signal. Through the cascaded structure of CNN and LSTM, the attack detection model simultaneously considers the spatial correlation and temporal evolution of the attack signal, improving the ability to identify complex attacks.

[0052] Optionally, when training the attack identification model, the training samples cover the fused feature vectors and corresponding attack type labels under the above-mentioned multi-dimensional signal parameter side-channel attack scenarios, with a sample size of no less than 100,000. These samples include normal operation data, environmental interference data, and labeled data for various attack types such as power supply glitches, laser fault injection attacks, voltage stress attacks, temperature stress attacks, electromagnetic radiation attacks, physical attacks, and combined attacks. After model training, the hyperparameters need to be optimized using a validation set to ensure that the attack identification accuracy is no less than 99% and the false positive rate is less than 0.5%.

[0053] Optionally, the trained attack identification model is stored in the non-volatile secure storage area of ​​the security chip, and the model parameters are securely loaded into the private RAM through the MMU memory management unit of the AMBA bus to prevent the attack identification model from being maliciously tampered with or stolen.

[0054] Specifically, after receiving the fused feature vector, the processor calls the attack identification model to perform inference, and finally the fully connected layer of the attack identification model outputs the judgment result, including whether there is a side-channel attack, the attack type and the attack intensity level. The attack intensity level is 0-10, where level 0 means no attack and the higher the level, the more severe the attack.

[0055] S304 In the event of a side-channel attack, the processor causes the security chip to execute protective response measures corresponding to the attack type and attack strength level.

[0056] Specifically, in the event of a side-channel attack, by implementing protective response measures corresponding to the attack type and intensity level, the normal service availability of the security chip can be maximized while ensuring its security.

[0057] In this embodiment, during the operation of the security chip, it first receives multi-dimensional signal parameters collected by a multi-sensor acquisition module and preprocesses these parameters into a fused feature vector. This fused feature vector is then input into an attack identification model to determine the presence of a side-channel attack, as well as the attack type and intensity level. Compared to related technologies that only compare a single signal parameter, this approach allows for the detection of attack traces in other dimensions even when a single signal parameter is spoofed, improving the accuracy of side-channel attack identification. After confirming the existence of a side-channel attack, the security chip is controlled to execute protective response measures corresponding to the attack type and intensity level, ensuring uninterrupted operation even under minor side-channel attacks. Therefore, this embodiment improves the accuracy of side-channel attack identification in complex attack environments while preventing overreaction by the security chip under minor attacks.

[0058] This disclosure provides another method for detecting side-channel attacks on security chips, combined with... Figure 4 As shown, the detection method includes: S401, the processor receives multi-dimensional signal parameters from the security chip acquired by the multi-sensor acquisition module.

[0059] S402, the processor performs time synchronization processing on multi-dimensional signal parameters.

[0060] Specifically, there may be data timing misalignment issues between multidimensional signal parameters due to acquisition. By performing time synchronization processing on the multidimensional signal parameters, it can be ensured that different signal parameters can be accurately aligned to the same security chip operating time.

[0061] The S403 processor uses the time-synchronized multidimensional signal parameters for min-max normalization to generate a fused feature vector.

[0062] Specifically, after time synchronization, the multidimensional signal parameters still retain their independent physical dimensions and numerical ranges. If these data are directly input into the attack identification model, the difference in dimensions will lead to an imbalance in the model's weights, making it impossible for the model to clearly define the true mapping relationship between various signal parameters and attack types, thus affecting the accuracy of attack identification. Therefore, it is necessary to perform min-max normalization on the time-synchronized multidimensional signal parameters to generate a fused feature vector.

[0063] Alternatively, the expression for min-max normalization is as follows:

[0064] in These are the original signal parameters. , These are the historical extreme values ​​of this signal parameter. These are the normalized eigenvalues.

[0065] The S404 processor will integrate the feature vector input attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level.

[0066] S405, in the event of a side-channel attack, the processor causes the security chip to execute protective response measures corresponding to the attack type and attack strength level.

[0067] In this embodiment, a fused feature vector is generated by performing time synchronization processing on the multi-dimensional signal parameters and then applying min-max normalization to the time-synchronized multi-dimensional signal parameters. This unifies the physical dimensions and numerical ranges of the signal parameters in the fused feature vector, which helps improve the accuracy of side-channel attack identification.

[0068] This disclosure provides another method for detecting side-channel attacks on security chips, combined with... Figure 5 As shown, the detection method includes: S501, the processor receives multi-dimensional signal parameters from the security chip acquired by the multi-sensor acquisition module.

[0069] The S502 processor preprocesses multidimensional signal parameters to generate a fused feature vector.

[0070] The S503 processor will integrate the feature vector input attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level.

[0071] In the case of a side-channel attack, the processor calculates a strength score based on the attack type and attack strength level.

[0072] Specifically, attack types include power supply glitches, laser fault injection attacks, voltage coercion attacks, temperature coercion attacks, electromagnetic radiation attacks, physical attacks (such as chip decapsulation and FIB modification), and composite attacks involving multiple concurrent attacks. The attack strength level is an initial score derived by the attack identification model based on a comprehensive evaluation of the abnormal amplitude, rate of change, duration, and number of abnormal triggers for each signal parameter in the fused feature vector. However, different types of attacks pose significantly different threats to the core assets of the security chip; even with the same attack strength level, their actual harm may differ. For example, in attack scenarios with the same strength level of 6, a laser fault injection attack targeting the cryptographic engine might directly cause bit flipping in the key register, with a harm far exceeding that of environmental temperature fluctuations of the same strength. Therefore, selecting response measures solely based on the attack strength level cannot distinguish the differences in harm between attack types, potentially leading to insufficient mild response measures to defend against high-harm attacks, or unnecessary business interruptions caused by severe response measures under low-harm attacks. Therefore, it is necessary to first calculate the strength score based on the attack type and attack strength level.

[0073] Specifically, a corresponding risk coefficient is pre-configured for each attack type. The coefficient for each attack type is determined comprehensively based on attack principle analysis, historical attack case statistics, and chip security architecture threat model. The specific calculation method is as follows: strength score for each attack type = attack strength level × attack type risk coefficient. Wherein, the attack strength level is the original score level output by the model; the attack type risk coefficient has a preset weight value, and the final strength score is the sum of the strength scores of various attacks.

[0074] S505, the processor executes the target protection response measures corresponding to the strength score.

[0075] Optionally, target protection response measures corresponding to the strength score are executed, including: when the strength score is in the low score range, logging, internal alarms and dynamic noise injection are executed; when the strength score is in the medium score range, non-core functions are suspended, key fragmentation migration and access permission isolation are executed; when the strength score is in the high score range, sensitive data erasure and chip security locking are executed.

[0076] Optionally, the score range for the lower level is 0 to 3 points, the score range for the intermediate level is 4 to 7 points, and the score range for the higher level is 8 to 10 points.

[0077] In this embodiment, a strength score is first calculated based on the attack type and attack strength level, and then the target protection response measures corresponding to the strength score are executed. This avoids the problem that selecting response measures solely based on attack strength level cannot distinguish the differences in the harm caused by different attack types, which may result in mild response measures being insufficient to resist high-harm attacks, or severe response measures causing unnecessary business interruptions under low-harm attacks.

[0078] In some embodiments, the detection method further includes: when the strength score is greater than or equal to the lower limit of the intermediate score range, recording the multidimensional signal parameters of the side-channel attack, the attack type, the attack strength level, and the target protection response measures implemented, generating an attack tracing report; and storing the attack tracing report in a non-volatile secure storage area.

[0079] In this embodiment, if the strength score is greater than or equal to the lower limit of the intermediate score range, it indicates that the security chip has been subjected to a relatively strong attack. In this case, the multi-dimensional signal parameters of the side-channel attack, the attack type, the attack strength level, and the target protection response measures implemented are recorded to generate an attack tracing report. The attack tracing report is then stored in a non-volatile secure storage area, achieving secure retention of traceable evidence.

[0080] Combination Figure 6 As shown in the figure, this disclosure provides a detection system 600 for side-channel attacks on security chips, including: a signal acquisition module 601, a data preprocessing module 602, an attack identification module 603, and a graded response module 604. The signal acquisition module 601 is used to acquire multi-dimensional signal parameters of the security chip using a multi-sensor acquisition module. The data preprocessing module 602 is used to preprocess the multi-dimensional signal parameters to generate a fused feature vector. The attack identification module 603 is used to input the fused feature vector into an attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level. The graded response module 604 is used to execute protective response measures corresponding to the attack type and attack strength level in the event of a side-channel attack.

[0081] Optionally, the detection system 600 for side-channel attacks on security chips also includes an attack tracing module 605. The attack tracing module records the multi-dimensional signal parameters, attack type, attack strength level, and target protection response measures executed in the side-channel attack, generating an attack tracing report; the attack tracing report is stored in a non-volatile secure storage area. Combined with... Figure 7As shown, this disclosure provides a detection device 700 for side-channel attacks on security chips, including a processor 701 and a memory 702. Optionally, the device may further include a communication interface 703 and a bus 704. The processor 701, communication interface 703, and memory 702 can communicate with each other via the bus 704. The communication interface 703 can be used for information transmission. The processor 701 can call logical instructions in the memory 702 to execute the detection method for side-channel attacks on security chips described in the above embodiment.

[0082] Furthermore, the logic instructions in the aforementioned memory 702 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0083] The memory 702, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor 701 executes functional applications and data processing by running the program instructions / modules stored in the memory 702, thereby implementing the detection method for side-channel attacks on security chips in the above embodiments.

[0084] The memory 702 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 702 may include high-speed random access memory and may also include non-volatile memory.

[0085] This disclosure provides a computer-readable storage medium storing computer-executable instructions configured to execute the above-described detection method for side-channel attacks on security chips.

[0086] The technical solutions of this disclosure can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes one or more instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in this disclosure. The aforementioned storage medium can be a non-transitory storage medium, such as a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., and other media capable of storing program code.

[0087] The foregoing description and accompanying drawings fully illustrate embodiments of this disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, the singular forms “a,” “an,” and “the” are intended to equally include the plural forms unless the context clearly indicates otherwise. Similarly, the term “and / or” as used in this application means including one or more of the associated listed items and all possible combinations thereof. Additionally, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Without further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. In this document, each embodiment may focus on the differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, the relevant parts can be referred to the description of the method section.

[0088] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A method for detecting side-channel attacks on security chips, characterized in that, The security chip has a multi-sensor acquisition module inside, and the detection methods include: Receives multi-dimensional signal parameters from the security chip acquired by the multi-sensor acquisition module; Multidimensional signal parameters are preprocessed to generate fused feature vectors; The fused feature vectors are input into the attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level. In the event of a side-channel attack, the security chip will execute protective response measures corresponding to the attack type and attack strength level.

2. The detection method according to claim 1, characterized in that, Multidimensional signal parameters include: The external power supply voltage of the security chip, the internal core voltage domain voltage of the security chip, the temperature of the packaging surface or core die of the security chip, the light intensity received by the security chip, the external input clock frequency of the security chip, the external power supply glitches of the security chip, the internal power supply glitches of the security chip, the current consumption of the entire or part of the functional modules of the security chip, and the packaging integrity of the security chip.

3. The detection method according to claim 1, characterized in that, The multidimensional signal parameters are preprocessed to generate a fused feature vector, including: Time synchronization processing of multidimensional signal parameters; The multidimensional signal parameters after time synchronization are subjected to min-max normalization to generate a fused feature vector.

4. The detection method according to claim 3, characterized in that, The expression for min-max normalization is as follows: in These are the original signal parameters. , These are the historical extreme values ​​of this signal parameter. These are the normalized eigenvalues.

5. The detection method according to claim 1, characterized in that, To enable the security chip to perform protective response measures corresponding to the attack type and attack strength level, including: A strength score is calculated based on the attack type and attack intensity level. Implement the target protection response measures corresponding to the intensity score.

6. The detection method according to claim 5, characterized in that, Implement the target protection response measures corresponding to the intensity score, including: When the intensity score is in the low score range, log recording, internal alarms, and dynamic noise injection are performed. When the strength score is in the intermediate range, perform the following actions: suspend non-core functions, migrate key fragments, and isolate access permissions. When the strength score is in the high score range, perform sensitive data erasure and chip security lock.

7. The detection method according to claim 6, characterized in that, Also includes: If the strength score is greater than or equal to the lower limit of the intermediate score range, record the multi-dimensional signal parameters of the side-channel attack, the attack type, the attack strength level, and the target protection response measures implemented, and generate an attack tracing report. The attack attribution report is stored in a non-volatile secure storage area.

8. A detection system for side-channel attacks on security chips, characterized in that, The security chip has a multi-sensor acquisition module inside, and the detection device includes: The signal acquisition module is used to receive multi-dimensional signal parameters from the security chip acquired by the multi-sensor acquisition module. The data preprocessing module is used to preprocess multidimensional signal parameters and generate fused feature vectors; The attack identification module is used to input the fused feature vector into the attack identification model to determine whether a side-channel attack exists, as well as the attack type and attack strength level. The graded response module is used to enable the security chip to perform protective response measures corresponding to the attack type and attack strength level in the event of a side-channel attack.

9. A detection device for side-channel attacks on security chips, comprising a processor and a memory storing program instructions, characterized in that, The processor is configured to execute, when running the program instructions, a detection device for side-channel attacks on a security chip as described in any one of claims 1 to 7.

10. A security chip, characterized in that, include: Chip body; The detection system for side-channel attacks on security chips as described in claim 8, or the detection device for side-channel attacks on security chips as described in claim 9, is integrated inside the chip body.