CN109815705B - Hardware Trojan horse detection method based on scan chain feature analysis - Google Patents
Hardware Trojan horse detection method based on scan chain feature analysis Download PDFInfo
- Publication number
- CN109815705B CN109815705B CN201910097553.6A CN201910097553A CN109815705B CN 109815705 B CN109815705 B CN 109815705B CN 201910097553 A CN201910097553 A CN 201910097553A CN 109815705 B CN109815705 B CN 109815705B
- Authority
- CN
- China
- Prior art keywords
- netlist
- scan chain
- trojan
- circuit
- hardware trojan
- 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.)
- Expired - Fee Related
Links
- ZXQYGBMAQZUVMI-GCMPRSNUSA-N gamma-cyhalothrin Chemical compound CC1(C)[C@@H](\C=C(/Cl)C(F)(F)F)[C@H]1C(=O)O[C@H](C#N)C1=CC=CC(OC=2C=CC=CC=2)=C1 ZXQYGBMAQZUVMI-GCMPRSNUSA-N 0.000 title claims abstract description 97
- 238000001514 detection method Methods 0.000 title claims abstract description 43
- 238000004458 analytical method Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 claims abstract description 31
- 238000013461 design Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 150000001875 compounds Chemical class 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 9
- 238000010801 machine learning Methods 0.000 abstract description 8
- 230000007547 defect Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 22
- 238000010586 diagram Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000012795 verification Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000002513 implantation Methods 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000007943 implant Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
Images
Landscapes
- Tests Of Electronic Circuits (AREA)
Abstract
The invention relates to a hardware Trojan detection method based on scan chain feature analysis, which is provided in a chip design stage starting from a gate-level netlist of an integrated circuit, converts the gate-level netlist of the integrated circuit into the scan chain netlist, combines the advantages of a traditional circuit feature analysis method, provides two new Trojan features, and classifies a Trojan net and a normal net by using a LightGBM algorithm. Meanwhile, the invention combines the machine learning algorithm and the circuit characteristic analysis method to be applied to the hardware Trojan horse detection method and has the characteristics of real time, accuracy, low cost and the like. Aiming at the defects that the time complexity is high and the individual Trojan detection efficiency is low in the traditional hardware Trojan circuit characteristic analysis method, the scan chain netlist characteristic analysis method provided by the invention can improve the detection effect of the hardware Trojan and the detection efficiency of the method can be improved by machine learning of the LightGBM algorithm.
Description
Technical Field
The invention relates to the field of integrated circuit Trojan horse detection, in particular to a hardware Trojan horse detection method based on scan chain feature analysis.
Background
With the development of wireless communication, sensor technology, embedded systems and microelectronic technology, the security problem of integrated circuits is widely regarded by academic and industrial circles.
At present, hardware Trojan horse detection technologies can be roughly divided into two types: the two modes of formal verification and circuit characteristic analysis have respective advantages and disadvantages.
Formal verification is an algorithm-based logic verification method that can exhaustively demonstrate a set of predefined security attribute rules that an IC design should satisfy. As with the untrusted circuit identification technique proposed by Hicks et al, the suspected net of circuits is determined by looking for RTL code lines that have not been executed during the test; rajendran et al propose various constrained designs for third party IP cores to prevent and isolate portions of circuits infected with hardware trojans. These techniques, however, result in higher area and power costs and do not prevent gate level hardware trojan implants.
And the circuit characteristic analysis technology utilizes the characteristic information of the circuit to reflect the health state of the chip. For example, Hasegawa et al propose using a machine learning algorithm to classify a Trojan netlist from a normal netlist. The method has high detection efficiency and low cost, but the detection precision and efficiency of the method depend on the selection of Trojan horse characteristics, and the stability is lacked; in addition, a hardware Trojan detection method based on side channel analysis is also provided, the method searches for the hardware Trojan by comparing side channel information of a standard circuit and a Trojan circuit, unfortunately, a 'standard' chip sample is difficult to obtain in reality, and the detection effect of a side channel depends on the accuracy of an instrument seriously, so that the method is not suitable for a super-large-scale integrated circuit. And the side channel method has weak interference processing capability on process noise, and the denoising processing of the bypass noise may affect the original signal again.
Disclosure of Invention
In view of this, the present invention provides a hardware Trojan horse detection method based on scan chain feature analysis, which can improve the detection effect of the hardware Trojan horse.
The invention is realized by adopting the following scheme: a hardware Trojan horse detection method based on scan chain feature analysis specifically comprises the following steps:
step S1: analyzing the logic structure of the gate-level netlist, and extracting the characteristics of the standard netlist; the main characteristics include: the number of circuit fan-in pins, the number of circuit fan-out pins, the number of multiplexers, the number of circuit loops and circuit constant information (such as grounding or constant signals);
step S2: converting the logically synthesized gate-level circuit netlist into a scan chain gate-level netlist through testability design;
step S3: extracting circuit characteristics according to the scan chain gate-level netlist circuit information, wherein the extracted circuit characteristics comprise: a suspicious Trojan trigger module, suspicious observation nodes;
step S4: dividing the circuit characteristic sample extracted in the step S3 into a training data set and a testing data set, and training a two-classification LightGBM algorithm by adopting the training data set to obtain a LightGBM-based hardware Trojan classifier;
step S5: and inputting the test data set into a trained LightGBM hardware Trojan classifier through a cross-validation method to obtain a hardware Trojan prediction result based on the model. And comparing the prediction result of each group of samples with the real result, and calculating the accuracy, precision and recall of the LightGBM hardware Trojan classifier.
Starting from a gate-level netlist of an integrated circuit, the invention provides a hardware Trojan detection method based on scan chain feature analysis in a chip design stage, the gate-level netlist of the integrated circuit is firstly converted into the scan chain netlist, and then two new Trojan features are provided by combining the advantages of the traditional circuit feature analysis method, and a LightGBM algorithm is used for classifying a Trojan net and a normal net. Meanwhile, the invention combines the machine learning algorithm and the circuit characteristic analysis method to be applied to the hardware Trojan horse detection method and has the characteristics of real time, accuracy, low cost and the like. Aiming at the defects that the time complexity is high and the individual Trojan detection efficiency is low in the traditional hardware Trojan circuit characteristic analysis method, the scan chain netlist characteristic analysis method provided by the invention can improve the detection effect of the hardware Trojan and the detection efficiency of the method can be improved by machine learning of the LightGBM algorithm.
Further, step S3 specifically includes the following steps:
step S31: initializing a Trojan trigger module STRiModle and a Trojan observation point SObsPoint;
step S32: analyzing the combination elements in the scan chain netlist, if a module which is not converted exists in the scan chain netlist, the module is regarded as a Trojan horse trigger module, and making STRiModule be STRiModule + 1;
step S33: if some special nets exist in the scan chain netlist, the special nets are nets which destroy the chain structure of the scan chain net, the special nets are regarded as suspicious observation nodes, and sobsPoint is equal to sobsPoint + 1.
Preferably, the present invention provides the above two new Trojan horse characteristics, and adds the two characteristics into the conventional circuit analysis technology, so as to improve the detection accuracy of the hardware Trojan horse.
With respect to the suspicious trigger module: the trigger part of the hardware trojan can be characterized, since in the scan chain technique, every normal sequence element (D flip-flop) will be replaced by a scan sequence element (scan D flip-flop). And in order to ensure the concealment of the hardware Trojan horse, a Trojan horse manufacturer can prevent a trigger module of the Trojan horse from being converted into a scan chain element. Therefore, by comparing the standard netlist with the scan chain netlist, the suspected Trojan trigger module can be found.
Regarding suspicious observation points: to ensure that Trojan nets are not missed, it is also necessary to identify Trojan nets that are hidden in the normal netlist, and if an attacker wants to steal important private information, the signal must pass through the final output. The scan chain may convert sequential circuits into combinational circuits that will maintain a shifted chain structure. Thus, the injection of a hardware trojan will in part affect the linear relationship of the chain structure. Thus breaking a wire mesh of a chain structure can be defined as a suspicious observation point.
Further, in step S4, the training process specifically includes the following steps:
step S41: pre-sorting all the characteristics according to numerical values;
step S42: calculating the splitting gain of each characteristic, and selecting the optimal dividing point of the sample characteristic;
step S43: and according to the characteristics and the optimal segmentation points, adding the submodels one by one, optimizing the objective function and ensuring that the loss function is continuously reduced.
Further, step S42 specifically includes the following steps:
step S421: let the total objective function l of the t-th iteration(t)The following were used:
in the formula (I), the compound is shown in the specification,to representA predicted value of the model;is a squared error function; f. oft(xi) A value representing a prediction of the newly added base tree, corresponding to an argument increment of the loss function; omega (f)t) The regular penalty term representing the target function is mainly used for limiting the number of leaf nodes and preventing overfitting;
step S422: the splitting Gain is obtained by performing taylor expansion on the loss function, and the formula is as follows:
wherein i represents a sample index, giRepresents the first derivative of the loss function at t-1, hiRepresenting the second derivative of the loss function, λ represents the expanded redundancy term, γ represents the complexity cost of the new leaf node added, I represents the samples obtained without splitting the tree, I representsRRepresenting the right subtree sample, ILRepresenting a left subtree sample;
step S423: let the right subtree Gain value be:
let the left subtree Gain value be:
let the Gain value of the undivided tree be:
further, the gate-level circuit netlist is a circuit netlist before tape-out. Aiming at the problem that the side channel analysis method is susceptible to process noise, the invention adopts the circuit netlist before the tape-out to analyze possible Trojan threats, the detection method has lower cost, and the implantation position of the Trojan can be further positioned by analyzing the hardware Trojan from the circuit logic structure so as to improve the safety design of the chip.
Compared with the prior art, the invention has the following beneficial effects:
1. the method can effectively realize the positioning of the hardware trojan in the netlist, has more flexible operation, does not change the original architecture of the circuit, and is suitable for the safety detection of large-scale integrated circuits.
2. The invention adopts the scan chain characteristic analysis technology, can more obviously reflect the structural characteristics of the hardware Trojan horse, is suitable for the protection of various types of malicious circuits, and is beneficial to improving the detection precision of the traditional hardware Trojan horse defense technology.
3. The invention also adopts a LightGBM gradient lifting algorithm, applies a machine learning algorithm to the detection and analysis of the hardware Trojan horse, has the advantages of strong model expression capability, capability of processing high-dimensional sparse characteristic data, capability of realizing an end-to-end learning mode by an intelligent technology frame, application of the priori knowledge of the hardware Trojan horse method developed to date and the existing Trojan horse characteristic library, self-adaptive updating of the detection method, contribution to improving the Trojan horse detection efficiency, reduction of the expenditure of time and cost, and suitability for the ultra-large scale integrated circuit.
Drawings
FIG. 1 is a schematic flow chart of a method according to an embodiment of the present invention.
Fig. 2 shows the detection effect of the LightGBM algorithm in 15 gate-level hardware trojan samples according to the embodiment of the present invention.
Fig. 3 shows the detection effect of the method according to the embodiment of the present invention for different types of hardware trojans.
FIG. 4 is a structural diagram of a scan chain gate-level netlist according to an embodiment of the present invention.
Detailed Description
The invention is further explained by the following embodiments in conjunction with the drawings.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
As shown in fig. 1, the present embodiment provides a hardware trojan detection method based on scan chain feature analysis, which specifically includes the following steps:
step S1: analyzing the logic structure of the gate-level netlist, and extracting the characteristics of the standard netlist; the main characteristics include: the number of circuit fan-in pins, the number of circuit fan-out pins, the number of multiplexers, the number of circuit loops and circuit constant information (such as grounding or constant signals);
step S2: converting the logically synthesized gate-level circuit netlist into a scan chain gate-level netlist through testability design;
step S3: extracting circuit characteristics according to the scan chain gate-level netlist circuit information, wherein the extracted circuit characteristics comprise: a suspicious Trojan trigger module, suspicious observation nodes;
step S4: dividing the circuit characteristic sample extracted in the step S3 into a training data set and a testing data set, and training a two-classification LightGBM algorithm by adopting the training data set to obtain a LightGBM-based hardware Trojan classifier;
step S5: and inputting the test data set into a trained LightGBM hardware Trojan classifier through a cross-validation method to obtain a hardware Trojan prediction result based on the model. And comparing the prediction result of each group of samples with the real result, and calculating the accuracy, precision and recall of the LightGBM hardware Trojan classifier.
Starting from the gate-level netlist of the integrated circuit, the embodiment provides a hardware Trojan detection method based on scan chain feature analysis in the chip design stage, the gate-level netlist of the integrated circuit is firstly converted into the scan chain netlist, then two new Trojan features are provided by combining the advantages of the traditional circuit feature analysis method, and a LightGBM algorithm is used for classifying a Trojan net and a normal net. Meanwhile, the invention combines the machine learning algorithm and the circuit characteristic analysis method to be applied to the hardware Trojan horse detection method and has the characteristics of real time, accuracy, low cost and the like. Aiming at the defects that the time complexity is high and the individual Trojan detection efficiency is low in the traditional hardware Trojan circuit characteristic analysis method, the scan chain netlist characteristic analysis method provided by the invention can improve the detection effect of the hardware Trojan and the detection efficiency of the method can be improved by machine learning of the LightGBM algorithm.
In this embodiment, step S3 specifically includes the following steps:
step S31: initializing a Trojan trigger module STRiModle and a Trojan observation point SObsPoint;
step S32: analyzing the combination elements in the scan chain netlist, if a module which is not converted exists in the scan chain netlist, the module is regarded as a Trojan horse trigger module, and making STRiModule be STRiModule + 1;
step S33: if some special nets exist in the scan chain netlist, the special nets are nets which destroy the chain structure of the scan chain net, the special nets are regarded as suspicious observation nodes, and sobsPoint is equal to sobsPoint + 1.
Preferably, the present embodiment provides the above two new trojan features, and adds the two features into the conventional circuit analysis technology, so as to improve the detection accuracy of the hardware trojan.
With respect to the suspicious trigger module: the trigger part of the hardware trojan can be characterized, since in the scan chain technique, every normal sequence element (D flip-flop) will be replaced by a scan sequence element (scan D flip-flop). And in order to ensure the concealment of the hardware Trojan horse, a Trojan horse manufacturer can prevent a trigger module of the Trojan horse from being converted into a scan chain element. Therefore, by comparing the standard netlist with the scan chain netlist, the suspected Trojan trigger module can be found.
Regarding suspicious observation points: to ensure that Trojan nets are not missed, it is also necessary to identify Trojan nets that are hidden in the normal netlist, and if an attacker wants to steal important private information, the signal must pass through the final output. The scan chain may convert sequential circuits into combinational circuits that will maintain a shifted chain structure. Thus, the injection of a hardware trojan will in part affect the linear relationship of the chain structure. Thus breaking a wire mesh of a chain structure can be defined as a suspicious observation point.
In this embodiment, in step S4, the training process specifically includes the following steps:
step S41: pre-sorting all the characteristics according to numerical values;
step S42: calculating the splitting gain of each characteristic, and selecting the optimal dividing point of the sample characteristic;
step S43: and according to the characteristics and the optimal segmentation points, the submodels are added one by one, the objective function is optimized, and the loss function is ensured to be continuously reduced.
In this embodiment, step S42 specifically includes the following steps:
step S421: let the total objective function l of the t-th iteration(t)The following were used:
in the formula (I), the compound is shown in the specification,representing a predicted value of the model;is a squared error function; f. oft(xi) A value representing a prediction of the newly added base tree, corresponding to an argument increment of the loss function; omega (f)t) The regular penalty term representing the target function is mainly used for limiting the number of leaf nodes and preventing overfitting;
step S422: the splitting Gain is obtained by performing taylor expansion on the loss function, and the formula is as follows:
wherein i represents a sample index, giRepresents the first derivative of the loss function at t-1, hiRepresenting the second derivative of the loss function, λ represents the expanded redundancy term, γ represents the complexity cost of the new leaf node added, I represents the samples obtained without splitting the tree, I representsRRepresenting the right subtree sample, ILRepresenting a left subtree sample;
step S423: let the right subtree Gain value be:
let the left subtree Gain value be:
let the Gain value of the undivided tree be:
in this embodiment, the gate-level circuit netlist is a circuit netlist before tape-out. Aiming at the problem that the side channel analysis method is susceptible to process noise, the invention adopts the circuit netlist before the tape-out to analyze possible Trojan threats, the detection method has lower cost, and the implantation position of the Trojan can be further positioned by analyzing the hardware Trojan from the circuit logic structure so as to improve the safety design of the chip.
Specifically, fig. 2 shows the detection effect of the LightGBM algorithm according to the embodiment of the present invention on 15 gate-level hardware trojan samples. Fig. 3 shows the detection effect of the method according to the embodiment of the present invention for different types of hardware trojans. Fig. 4 is a schematic structural diagram of a scan chain gate-level netlist according to an embodiment of the present invention.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is directed to preferred embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow. However, any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present invention are within the protection scope of the technical solution of the present invention.
Claims (2)
1. A hardware Trojan horse detection method based on scan chain feature analysis is characterized in that: the method comprises the following steps:
step S1: analyzing the logic structure of the gate-level netlist, and extracting the characteristics of the standard netlist;
step S2: converting the logically synthesized gate-level circuit netlist into a scan chain gate-level netlist through testability design;
step S3: extracting circuit characteristics according to the scan chain gate-level netlist circuit information, wherein the extracted circuit characteristics comprise: a suspicious Trojan trigger module, suspicious observation nodes;
step S4: dividing the circuit characteristic sample extracted in the step S3 into a training data set and a testing data set, and training a two-classification LightGBM algorithm by adopting the training data set to obtain a LightGBM-based hardware Trojan classifier;
step S5: inputting the test data set into a trained LightGBM hardware Trojan horse classifier through a cross-validation method to obtain a LightGBM-based hardware Trojan horse prediction result of the LightGBM hardware Trojan horse classifier;
step S3 specifically includes the following steps:
step S31: initializing a Trojan trigger module STRiModule and a Trojan observation point SObsPoint;
step S32: analyzing the combination elements in the scan chain netlist, if a module which is not converted exists in the scan chain netlist, the module is regarded as a Trojan horse trigger module, and making STRiModule be STRiModule + 1;
step S33: if some special nets exist in the scan chain netlist, wherein the special nets are nets with broken chain structures of the scan chain net, the special nets are regarded as suspicious observation nodes, and sobsPoint is equal to sobsPoint + 1;
in step S4, the training process specifically includes the following steps:
step S41: pre-sorting all the characteristics according to numerical values;
step S42: calculating the splitting gain of each feature, and selecting the optimal splitting point of the sample feature;
step S43: according to the characteristics and the optimal segmentation points, the submodels are added one by one, the objective function is optimized, and the loss function is ensured to be continuously reduced;
step S42 specifically includes the following steps:
in the formula (I), the compound is shown in the specification,representing a predicted value of the model;is a squared error function; f. oft(xi) A value representing a prediction of the newly added base tree, corresponding to an argument increment of the loss function; omega (f)t) A regular penalty term representing an objective function;
step S422: the splitting Gain is obtained by performing taylor expansion on the loss function, and the formula is as follows:
wherein i represents a sample index, giRepresents the first derivative of the loss function at t-1, hiRepresenting the second derivative of the loss function, λ represents the expanded redundancy term, γ represents the complexity cost of the new leaf node added, I represents the samples obtained without splitting the tree, I representsRRepresenting the right subtree sample, ILRepresenting a left subtree sample;
step S423: let the right subtree Gain value be:
let the left subtree Gain value be:
let the Gain value of the undivided tree be:
2. the hardware Trojan horse detection method based on scan chain feature analysis according to claim 1, characterized in that: and the gate-level circuit netlist is a circuit netlist before tape-out.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910097553.6A CN109815705B (en) | 2019-01-31 | 2019-01-31 | Hardware Trojan horse detection method based on scan chain feature analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910097553.6A CN109815705B (en) | 2019-01-31 | 2019-01-31 | Hardware Trojan horse detection method based on scan chain feature analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109815705A CN109815705A (en) | 2019-05-28 |
CN109815705B true CN109815705B (en) | 2022-07-08 |
Family
ID=66606170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910097553.6A Expired - Fee Related CN109815705B (en) | 2019-01-31 | 2019-01-31 | Hardware Trojan horse detection method based on scan chain feature analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109815705B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110287735B (en) * | 2019-07-04 | 2021-05-04 | 电子科技大学 | Trojan horse infected circuit identification method based on chip netlist characteristics |
CN112231775B (en) * | 2019-07-15 | 2022-10-21 | 天津大学 | Hardware Trojan horse detection method based on Adaboost algorithm |
CN110458111B (en) * | 2019-08-14 | 2023-02-21 | 福州大学 | LightGBM-based rapid extraction method for vehicle-mounted laser point cloud power line |
CN111523116B (en) * | 2020-03-23 | 2023-05-16 | 南京航空航天大学 | Mixed mode multi-level gate-level hardware Trojan horse detection method based on machine learning |
CN112528284A (en) * | 2020-12-18 | 2021-03-19 | 北京明略软件系统有限公司 | Malicious program detection method and device, storage medium and electronic equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104239616A (en) * | 2014-09-02 | 2014-12-24 | 工业和信息化部电子第五研究所 | Design method of integrated circuit and hardware trojan detection method |
CN106649959A (en) * | 2016-10-10 | 2017-05-10 | 中国电子科技集团公司第五十八研究所 | Scan chain-based circuit design method and hardware Trojan detection method |
CN107480561A (en) * | 2017-07-21 | 2017-12-15 | 天津大学 | Hardware Trojan horse detection method based on few state node traverses |
CN107886012A (en) * | 2017-10-28 | 2018-04-06 | 天津大学 | One shot hardware Trojan horse detection method based on gate leve architectural feature |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9536031B2 (en) * | 2014-07-14 | 2017-01-03 | Mediatek Inc. | Replacement method for scan cell of integrated circuit, skewable scan cell and integrated circuit |
-
2019
- 2019-01-31 CN CN201910097553.6A patent/CN109815705B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104239616A (en) * | 2014-09-02 | 2014-12-24 | 工业和信息化部电子第五研究所 | Design method of integrated circuit and hardware trojan detection method |
CN106649959A (en) * | 2016-10-10 | 2017-05-10 | 中国电子科技集团公司第五十八研究所 | Scan chain-based circuit design method and hardware Trojan detection method |
CN107480561A (en) * | 2017-07-21 | 2017-12-15 | 天津大学 | Hardware Trojan horse detection method based on few state node traverses |
CN107886012A (en) * | 2017-10-28 | 2018-04-06 | 天津大学 | One shot hardware Trojan horse detection method based on gate leve architectural feature |
Non-Patent Citations (1)
Title |
---|
集成电路生产外包硬件木马电路知识产权窃取I/O混淆技术;崔晓通;《万方数据学位论文库》;20181218;第1-106页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109815705A (en) | 2019-05-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109815705B (en) | Hardware Trojan horse detection method based on scan chain feature analysis | |
Subramanyan et al. | Reverse engineering digital circuits using functional analysis | |
Li et al. | Wordrev: Finding word-level structures in a sea of bit-level gates | |
CN108268777B (en) | Similarity detection method for carrying out unknown vulnerability discovery by using patch information | |
CN110414277B (en) | Gate-level hardware Trojan horse detection method based on multi-feature parameters | |
Han et al. | Hardware trojans detection at register transfer level based on machine learning | |
CN109657461B (en) | RTL hardware Trojan horse detection method based on gradient lifting algorithm | |
CN108491228A (en) | A kind of binary vulnerability Code Clones detection method and system | |
US11321512B2 (en) | Method for automatic detection of a functional primitive in a model of a hardware system | |
CN104933364B (en) | A kind of malicious code based on the behavior of calling automates homologous determination method and system | |
CN107590313A (en) | Optimal inspection vector generation method based on genetic algorithm and analysis of variance | |
CN109067800A (en) | A kind of cross-platform association detection method of firmware loophole | |
CN110287735A (en) | Wooden horse based on chip netlist feature infects circuit identification method | |
Awais et al. | An MCTS-based framework for synthesis of approximate circuits | |
CN113868650B (en) | Vulnerability detection method and device based on code heterogeneous middle graph representation | |
CN108595986B (en) | Bounded model-based micro Trojan horse detection method | |
Huang et al. | A hardware trojan detection and diagnosis method for gate-level netlists based on different machine learning algorithms | |
CN110955892B (en) | Hardware Trojan horse detection method based on machine learning and circuit behavior level characteristics | |
CN116702157B (en) | Intelligent contract vulnerability detection method based on neural network | |
CN117134958A (en) | Information processing method and system for network technology service | |
CN110390215A (en) | A kind of hardware Trojan horse detection method and system based on raising activation probability | |
Darem | A Novel Framework for Windows Malware Detection Using a Deep Learning Approach. | |
US20220222341A1 (en) | Embedded malware detection using spatial voting and machine learning | |
Alrahis et al. | Graph neural networks for hardware vulnerability analysis—can you trust your GNN? | |
CN114021136A (en) | Back door attack defense system for artificial intelligence model |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220708 |