KR102592997B1 - Method of detecting software malfunction using Binary Image - Google Patents

Abstract

A software malfunction detection method according to an embodiment of the present invention includes the steps of (A) starting a software malfunction detection system to start detecting software malfunction; (B) A control unit of the malfunction detection system receiving a plurality of sensing information from a sensing unit; (C) the control unit binarizing the plurality of sensing information; (D) the control unit compares and learns using a machine learning neural network; (E) the control unit determining whether a software malfunction has occurred; and (F) the control unit outputting text or sound to indicate whether a malfunction of the software has occurred through an output unit.

Description

Method of detecting software malfunction using Binary Image}

The present invention relates to a method of detecting malfunction of software using a binarized image, and in particular, to a method of detecting malfunction of software that shapes sensing information according to the runtime execution flow of software into a binarized image and detects malfunction using a machine learning neural network. .

As smart electronic systems are being applied to self-driving cars, artificial intelligence robots, unmanned devices, etc., they are being developed to operate through sequential measurement linkage between detailed blocks of complex software.

In these smart electronic systems, when a specific part of the software malfunctions, it is very difficult to analyze the cause. Code errors in the block may also be the cause, but when errors occur in interactions between other connected modules, it is very difficult to analyze the cause. Patch work takes a considerable amount of time, so there are limitations to doing it manually.

In particular, while conventional smart electronic systems are controlled by complex software, there are various types of code error patterns caused by dynamic code execution rather than static code execution. At this time, defining the detection code as the code itself for each of these cases increases development complexity and has the problem of increasing the capacity of the code memory when inserting the inspection code.

Patent Document: Registered Patent Publication No. 10-2088164

The present invention was created to solve the above problems, and the purpose of the present invention is to provide a software malfunction detection method that shapes sensing information according to the runtime execution flow of software into a binary image and detects malfunction using a machine learning neural network. There is something to do.

A software malfunction detection method according to an embodiment of the present invention includes the steps of (A) starting a software malfunction detection system to start detecting software malfunctions; (B) a control unit of the malfunction detection system receiving a plurality of sensing information from a sensing unit; (C) the control unit binarizing the plurality of sensing information; (D) the control unit compares and learns using a machine learning neural network; (E) the control unit determining whether a software malfunction has occurred; and (F) the control unit outputting text or sound to indicate whether a malfunction of the software has occurred through an output unit.

In the software malfunction detection method according to an embodiment of the present invention, the plurality of sensing information includes any one of program counter, code address, memory access location, CPU registers, memory usage, and variable call. It is characterized by

In the method for detecting malfunction of software according to an embodiment of the present invention, step (C) indicates (C-1) a portion where each of the plurality of sensing information is high based on a critical plane regarding the threshold value (TH) as 1, marking the low part as 0; (C-2) forming a binarized image matrix in which the portion marked as 1 is displayed with black pixels and the portion marked with 0 is displayed with white pixels; and (C-3) converting the binarized image matrix into a one-dimensional array image.

In the software malfunction detection method according to an embodiment of the present invention, step (D) involves inputting the one-dimensional array image and the corresponding one-dimensional array reference image representing normal operation into the machine learning neural network for comparative learning. Do it as

In the software malfunction detection method according to an embodiment of the present invention, step (E) derives an error value, determines whether the error value is higher than a preset malfunction reference value, and determines whether the malfunction has occurred. It is characterized by

The features and advantages of the present invention will become more apparent from the following detailed description based on the accompanying drawings.

Prior to this, terms or words used in this specification and claims should not be interpreted in their usual, dictionary meaning, and the inventor should appropriately define the concept of the term in order to explain his or her invention in the best way. It must be interpreted with meaning and concept consistent with the technical idea of the present invention based on the principle that it can be done.

The software malfunction detection method according to an embodiment of the present invention converts sensing information into a binarized image using a critical plane, thereby minimizing data loss and reducing data size, and easily detecting software malfunction. It works.

1 is a configuration diagram of a software malfunction detection system to which a software malfunction detection method according to an embodiment of the present invention is applied.
Figure 2 is a flow chart of a method for detecting malfunction of software according to an embodiment of the present invention.
Figure 3 is an example of sensing information input according to a method for detecting malfunction of software according to an embodiment of the present invention.
Figure 4 is an example diagram showing the process of binarizing input sensing information according to a method for detecting malfunction of software according to an embodiment of the present invention.
Figure 5 is an example diagram showing the result of binarizing input sensing information according to a method for detecting malfunction of software according to an embodiment of the present invention.
Figure 6 is an example diagram showing a software malfunction detection process according to a software malfunction detection method according to an embodiment of the present invention.

The objectives, specific advantages and novel features of the present invention will become more apparent from the following detailed description and preferred embodiments taken in conjunction with the accompanying drawings. In this specification, when adding reference numbers to components in each drawing, it should be noted that identical components are given the same number as much as possible even if they are shown in different drawings. Additionally, terms such as first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. Additionally, in describing the present invention, if it is determined that a detailed description of related known technologies may unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. 1 is a configuration diagram of a software malfunction detection system to which a software malfunction detection method according to an embodiment of the present invention is applied.

The software malfunction detection system 10 according to an embodiment of the present invention includes a control unit 11, a memory unit 12, a sensing unit 13, and an output unit 14, as shown in FIG. 1.

The control unit 11 has at least one machine learning neural network inside, controls the overall operation of the software malfunction detection system, and stores the sensing information received through the sensing unit 13 when the software is executed. The process of detecting software malfunction is controlled by comparing it with the reference information stored in the unit 12.

The memory unit 12 is connected to the control unit 11 and stores a number of software related to the operation of the smart electronic system, reference information for detecting malfunctions, and sensing information received through the sensing unit 13.

The sensing unit 13 includes a plurality of sensors each connected to the control unit 11, the memory unit 12, the output unit 14, and other components (not shown) constituting the system, and software is activated through the plurality of sensors. Sensing information during execution is received and transmitted to the control unit 11. Here, the sensing information is an operation signal detected in the system configuration when the software is executed, for example, a change in the stack point related to the memory access position of the memory unit 12, a change in the code address, a change in memory usage, and a variable call. Including changes, etc.

The output unit 14 is connected to the control unit 11 and, under the control of the control unit 11, can display text or sound to indicate whether the software is malfunctioning.

In the software malfunction detection system 10 to which the software malfunction detection method according to the embodiment of the present invention configured as described above is applied, the control unit 11 binarizes the sensing information received through the sensing unit 13 when the software is executed. , malfunctions in the software can be detected by inputting the binarized sensing image, along with the standard binarized sensing image stored in the memory unit 12, into a machine learning neural network provided inside, such as LSVM (Linear Support Vector Machine), and comparing and learning. there is.

Hereinafter, a method for detecting malfunction of software according to an embodiment of the present invention will be described with reference to FIGS. 2 to 7. Figure 2 is a flowchart of a method for detecting a software malfunction according to an embodiment of the present invention, Figure 3 is an example diagram of sensing information input according to a method for detecting a software malfunction according to an embodiment of the present invention, and Figure 4 is a flowchart of a method for detecting a software malfunction according to an embodiment of the present invention. It is an example diagram showing the process of binarizing the input sensing information according to the software malfunction detection method according to the embodiment of the present invention, and Figure 5 shows the binarization of the input sensing information according to the software malfunction detection method according to the embodiment of the present invention. This is an exemplary diagram showing the results, and Figure 6 is an exemplary diagram showing the software malfunction detection process according to the software malfunction detection method according to an embodiment of the present invention.

The software malfunction detection method according to an embodiment of the present invention first starts the software malfunction detection system 10 shown in FIG. 1 to detect software malfunction (S210).

Accordingly, the control unit 11 receives a plurality of sensing information from the sensing unit 13 (S220).

Specifically, as the smart electronic system runs and executes software, the control unit 11 starts the malfunction detection system 10 of the software shown in FIG. 1 and detects FIG. 1 through a plurality of sensors included in the sensing unit 13. As shown in Figure 3, sensing information, that is, sensing information expressed as a graph of changes over time, can be received by detecting various operation signals that appear when the software is executed.

At this time, the operation signal includes signals of various variables such as program counter, code address, memory access location, CPU registers, memory usage, and variable call.

After receiving a plurality of sensing information, the control unit 11 binarizes the received sensing information (S230).

Specifically, the control unit 11 receives the sensing information shown in FIG. 4 and converts the received sensing information based on the critical plane 101 regarding the threshold TH shown in FIG. 4. That is, the control unit 11 uses [Equation 1] below to select portions (S11, S12) that are higher than the critical plane 101 regarding the threshold (TH) shown in FIG. 4 in the first sensing information (S1). It is set to have a value of 1, and in the second sensing information (S2), the part (S21) higher than the critical plane 101 with respect to the threshold (TH) shown in FIG. 4 is set to have a value of 1, and the third sensing information (S2) is set to have a value of 1. In S3), the parts (S31, S32) higher than the critical plane 101 with respect to the threshold TH shown in FIG. 4 are set to have a value of 1, and the other parts lower than the critical plane 101 are set to have a value of 0. Let's have it.

Here, R(x,y) represents the binarized image matrix, r(x,y) represents the coordinates of the sensing information, and TH represents the threshold.

At this time, the threshold (TH) may be set in advance according to the sensitivity of the smart electronic system, the number of sensing information, and the stability of the sensing information.

As shown in FIG. 5, the binarized sensing information displays the sensing information (s1, s2, s3) binarized in the column direction for the row at time (t), and the value of 1, such as "Ⅰ", is A binarized image matrix 200 may be formed in which pixels are displayed in black and pixels with a value of 0, such as "II", are displayed in white.

After obtaining the binarized image matrix 200, the control unit 11 converts the binarized image matrix 200 into a one-dimensional array image 200' using [Equation 2] below.

Here, RP represents a one-dimensional array image, and R(x,y) represents a binarized image matrix.

After converting the one-dimensional array image 200', the control unit 11 inputs the one-dimensional array image 200' into a machine learning neural network and compares and learns it (S240).

Specifically, the control unit 11 acquires the one-dimensional array image 200' shown in FIG. 6, and applies the acquired one-dimensional array image 200' to a machine learning neural network, such as a Linear Support Vector Machine (LSVM) neural network. You can enter it. Here, the LSVM neural network is a simple neural network with a relatively small capacity compared to other neural networks, and can improve the performance efficiency of malfunction judgment. Of course, other machine learning neural networks can be used in addition to the LSVM neural network.

Accordingly, the control unit 11 can learn by comparing the one-dimensional array image 200' with the one-dimensional array reference image 300 representing the corresponding normal operation, as shown in FIG. 6. Here, the one-dimensional array reference image 300 is included in the reference information for detecting malfunctions in the memory unit 12 and may be selected by the control unit 11 and input into the LSVM neural network.

By comparing and learning the one-dimensional array image 200' and the one-dimensional array reference image 300, the control unit 11 determines whether a software malfunction has occurred (S250).

At this time, determination of whether a malfunction has occurred can be performed using the LSVM neural network using [Equation 3] below.

Here, RP represents the one-dimensional array image 200', and td represents the one-dimensional array reference image 300 indicating normal operation.

By determining whether a malfunction has occurred using [Equation 3], an error value of 3 is derived, and the control unit 11 determines whether a malfunction has occurred by determining whether the derived error value is higher than a preset malfunction standard value. can do.

For example, the control unit 11 may determine that a software malfunction has occurred because the error value of 3 derived by determining whether a malfunction has occurred is higher than the preset malfunction reference value of 1.

As it is determined that a malfunction of the software has occurred, the control unit 11 can display the malfunction of the software in text or notify it with a sound through the output unit 14 (S260).

On the other hand, if the error value derived from the determination of whether a malfunction has occurred is lower than the preset malfunction reference value, the control unit 11 repeats the process again starting from the step (S220) of receiving a plurality of sensing information from the sensing unit 13. .

The software malfunction detection method according to the embodiment of the present invention in this process converts the sensing information into a binarized image using the critical plane 101, thereby minimizing data loss and reducing the data size, and detecting malfunction of the software. can be easily detected.

Although the technical idea of the present invention has been described in detail according to the above preferred embodiments, it should be noted that the above-described embodiments are for illustrative purposes only and are not intended for limitation.

Additionally, an expert in the technical field of the present invention will understand that various implementations are possible within the scope of the technical idea of the present invention.

10: Malfunction detection system 11: Control unit
12: memory unit 13: sensing unit
14: output unit 101: critical plane
200: Binarized image matrix 200': One-dimensional array image
300: One-dimensional array reference image

Claims (5)

(A) starting the software malfunction detection system to detect software malfunction;
(B) a control unit of the malfunction detection system receiving a plurality of sensing information from a sensing unit;
(C) the control unit binarizing the plurality of sensing information;
(D) the control unit compares and learns using a machine learning neural network;
(E) the control unit determining whether a software malfunction has occurred; and
(F) the control unit outputting text or sound as to whether a malfunction of the software has occurred through an output unit;
Including,
A method of detecting malfunction of software, characterized in that the plurality of sensing information includes any one of program counter, code address, memory access location, CPU registers, memory usage, and variable call.
delete According to claim 1,
The step (C) is
(C-1) displaying a high part of each of the plurality of sensing information as 1 and a low part of each of the plurality of sensing information as 0 based on a critical plane regarding the threshold value (TH);
(C-2) forming a binarized image matrix in which the portion marked as 1 is displayed with black pixels and the portion marked with 0 is displayed with white pixels; and
(C-3) converting the binarized image matrix into a one-dimensional array image;
A software malfunction detection method further comprising:
According to claim 3,
The step (D) is a method of detecting malfunction of software, characterized in that the one-dimensional array image and the corresponding one-dimensional array reference image representing the normal operation are input together into the machine learning neural network for comparative learning.
According to claim 1,
The step (E) is a method of detecting malfunction of software, characterized in that it derives an error value, determines whether the error value is higher than a preset malfunction reference value, and determines whether the malfunction has occurred.

Images