RU2623902C2 - Device for identification of preferences of information protection - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: if it is impossible or difficult to establish the importance of the properties (characteristics) of the compared objects or their variants and the lack of a training sample for forming the evaluation scale, it is suggested to use the original data of the estimated sample when determining the preferred object, the means, the technical solution option. The invention can be used in the military field - for planning, developing, creating and accepting more advanced weapons systems and weapons, and in civil - for better and competitive goods and products.

EFFECT: simplification of the calculation of the complex quality indicator of the assessment of technical solutions and tools and the improvement of the accuracy of the evaluation in comparison with the methods of the worst and best standards.

3 dwg, 5 tbl

Description

The invention relates to digital computing, and in particular to digital computing systems for determining the quality of compared variants of complex systems, tools, products and various objects described by a significant number of heterogeneous unit indicators. If it is impossible or very difficult to establish the importance of the properties (characteristics) of the compared objects or their variants and the lack of a training sample, it is proposed to use the initial data of the most evaluated sample to determine the preferred object, means, and technical solution variant to form an assessment scale.

The invention can be used in the military industry - for planning, development, creation and adoption of more advanced weapons systems and weapons, and in the civilian - better and more competitive products and products.

The technical result of the invention is to simplify the calculation of a comprehensive quality indicator (CPC) for evaluating technical means and solutions, as well as improving the accuracy of the assessment in comparison with the methods of best and worst standards.

The choice of a preferred tool is associated with solving a multicriteria problem. Consider several well-known ways to solve it.

1) A known method based on the use of hierarchy analysis method [1] and its modifications. The application of these methods provides the possibility of establishing the ratio of the characteristics of the compared means:

- the properties or characteristics of the evaluated assets are equally significant and important;

- one property or characteristic is somewhat more important than another;

- one property or characteristic is more important than another;

- one property or characteristic is clearly more important than others;

- one property or characteristic is absolutely more important than others.

If it is possible to establish such relations, the problem reduces to finding the weighting coefficients W _j and combining the normalized unit indices q _j into the complex Q = ƒ (q _j , W _j ). However, if the means under consideration have clearly heterogeneous properties (characteristics), it becomes impossible to use the hierarchy method and its modifications.

2) Known methods for finding the weighting coefficients using ranking methods [2, 3] characteristics of the compared means, making it difficult for experts in the case of heterogeneity of properties.

           3) Known expert methods for assessing the quality of the considered tools [3-5], based on the simplification of the first two methods. These are methods associated with the lexicographic method, with the method of averaging single indicators, the maximin method of assessing quality, the ideal point method, and others.

All the considered methods of choosing the preferred tool have inherent errors in expert systems associated with the presence of “subjectivity”.

4) A known method [6], based on the formation of a training sample to create a scale for assessing quality, which allows you to compare products of the evaluated sample. Its disadvantage is the requirement for a training sample.

5) Other inventions are known [7-11] for assessing the quality of objects that do not solve the problem formulated by the authors.

In order to increase the objectivity and reliability of assessing the quality of the funds under consideration, the authors propose to use the estimated sample, consisting of normalized individual quality indicators that allow you to create two hypothetical standards - the worst and best quality, respectively. And based on their use, from the totality of the means under consideration, choose an object that has the best generalized (comprehensive) quality indicator (CPC).

Closest to this technical solution is the "Method and device for choosing the preferred means of information protection", patent No. 2495482 [12]. Its disadvantages are:

1. The cumbersome mathematical apparatus associated with the multidimensionality of the space of unit indicators (EP), with the operation of multidimensional lines and planes, their transformation, which includes 6 stages of assessing the CCP.

2. A difficult to perceive geometric illustration of the method of choosing the preferred option, means of information protection (SZI).

3. There is no comparison of the method of two standards with the often used methods of better and worse standards.

1. The method and device for assessing the quality of technical solutions and means when creating weapons, which consists in the fact that they commute information about the individual indicators of the compared means, the options being evaluated, write it to the first memory block, send it to the block for generating standards of the worst and best quality, forming respectively beginning and end of a straight line defining a quality assessment scale, characterized in that the complex quality indicators of technical solutions options or compared means are determined by the fraction values , in the numerator of which is the sum of the squared lengths of the rating scale and the distance from the worst standard to the point of the compared option, from which the square of the distance from the point of the compared tool to the best standard is subtracted, and the denominator is the doubled square of the length of the rating scale, the maximum fraction corresponds to the preferred technical solution or valued product.

The essence of the method is best illustrated by comparing objects with heterogeneous characteristics. These include means of information security SZI, individual indicators of which are shown in table 1.

As a training sample (OBV), we use the estimated sample itself (OTSV). To conduct a quality assessment, the heterogeneous SZI indicators in table 1 are reduced to a dimensionless form. After rationing

they take the values indicated in table 2. A prerequisite for normalization is the growth of the CPC with the growth of its individual indicators EP.

For the convenience of determining the preferred SIS from a number of alternative, we divide the evaluation process into two stages:

(см. фиг. 1 для k=2).1. The geometric meaning of assessing the quality of options, products, and technical solutions will be demonstrated in the evaluation of information security tools. The CPC is found by forming a quality assessment scale by the method of two standards (MDE) in the form of a straight line E _x E _l connecting the points of the worst E _x and the best E _l quality standards in the space of estimated EP q _j ,

(see Fig. 1 for k = 2).

Connect the point C, in the two-dimensional space EP with the ends of the estimated scale E _x E _l . Knowing the coordinates of the points Ex, El (Table 3), it is easy to determine the length of the scale L ₀ and the lengths of the segments R _x and R _l , calculated by the formulas

If we drop the perpendicular from the point C _i onto the straight line E _x E _l , then at point K _{1 it will} divide the quality assessment scale into two segments L _x , L _l , the squares of which are determined from the formulas

To eliminate the distance h from point C _i to the rating scale (point K _i ), subtracting the left and right sides of the dependencies from formula (2b) and formula (2a), we have

Considering that L _xi (C _i ) = L ₀ -L _li and Q _i = L _xi / L ₀ , then substituting the value of L _xi in formula (3a) and going to Q _i , we obtain the value of the CPC of the estimated SPI _i in the MDE scale

Recall that the length L _xi (C _i ), referred to the length L _{0 of the} whole scale, gives the CPC Q _i = L _xi (C _i ) / L _{0 the} interval of values 0≤Q _i ≤1 in the MDE.

2. Formula (4) and the expression for Q _i , as well as the presence of EP SZI (Table 2) make it possible to proceed to a specific assessment of the CCP by the method of 2 standards. In the table. 3 shows the values of hypothetical standards E _x , E _l and their differences Δq _j (E _x E _l ), substituting which in formula (1a), we obtain the scale length L ₀ (E _x E _l ) = 2.0283 between points Eh, El.

Other values of R _x (E _x ), R _l (E _l ), L _x (E _x , C _i ), calculated by the above formulas, allow us to obtain the QPC Q _i (C _i ) of the estimated SZI _i (see table 3a) taking into account L ₀ = 2.028 and Q _i (SZI _i ) = (L ₀ ² + R _xi ² -R _li ² ) / 2L ₀ ² . For example, for PDA SZI ₁

Q ₁ (DSS ^{_{1) = (2,028 2 +1.4265 2}} -1,398 2) /2*2.028 2 = 0.51 .

3. The identification of the preferred SZI is carried out by placing the values of the CCP Q _i in a row

Q ₃ = 0.748> Q ₁ = 0.51> Q ₂ = 0.468,

from which the ranking of SZI follows

and, ultimately, the preference for SZI ₃ .

As an alternative to the invention in table. 4 shows the results of the method for assessing quality according to MBE Q _li = 1-R _li / L ₀ based on the method of the best standard (ideal point), which provides closeness to the best standard Q _li = 1-R _li / L ₀ , and the method for assessing the quality of SZI Q _xi ( _{х x} ) = R _xi / L ₀ , which estimates the distance from the worst standard _{х x} (MCE).

From the table. 4 and from FIG. 2 shows:

- the ranking order (5) of the estimated SIS _i for all 3 methods is preserved;

- CPC values calculated by the method of remoteness from the worst standard are overestimated in comparison with the method of 2 standards;

- CPC values calculated by the method of proximity to the best standard are underestimated in comparison with the method of 2 standards.

The device for implementing the method according to claim 1, shown in FIG. 1, comprising a switch 1 and a first memory unit 2, characterized in that it includes a unit 3 for generating normalized EPs, a unit 4 for determining the values of the numerator and denominator of the fraction (4a), a second memory unit 5, a visualization unit 6, and a unit 7 control (CU), while the output of the switch 1 is connected to the information input of the first memory unit 2, the output of which 8 is connected to the unit of normalization EP (according to table 2), the output 9 of which is connected to the unit 4 for determining the values of the numerator and denominator of the fraction connected its output 10 with the input of the second block Memory 6 is connected to the output 11 imaging unit 7, the control outputs of the control unit 8 is connected to the control inputs of all the blocks 1-7.

, сравниваемых средств СЗИ_i,

поступают на вход коммутатора 1 и по управляющей команде с блока 7 управления записываются в 1-й блок 2 памяти. По команде с блока 7 управления значения q_ij поступают в блок 3 нормирования ЕП. Нормированные ЕП по команде с БУ 7 переносятся в блок 4, в котором реализуется формула (4а) путем нахождения всех ее составляющих L₀,

для вычисления Q_i. Величины КПК Q_i, Q₂, Q₃ оцениваемых средств или технических решений записываются во втором блоке 2 памяти. По управляющей команде с блока 7 управления величины КПК Q₁, Q₂, Q₃ сравниваемых средств поступают на блок 6 визуализации. На блоке визуализации высвечивается информация: «Значения КПК. Q₁=0.51; Q₂=0.468; Q₃=0.748».The operation of the device is as follows. In accordance with FIG. 1 values of the initial unit indicators (EP) q _ij ,

compared means SZI _i ,

arrive at the input of the switch 1 and by the control command from the control unit 7 are recorded in the 1st memory unit 2. On command from the control unit 7, the values of q _ij enter block 3 of the normalization of the EP. Normalized EPs on command with BU 7 are transferred to block 4, in which formula (4a) is implemented by finding all its components L ₀ ,

to calculate Q _i . The values of the PDA Q _i , Q ₂ , Q _{3 of the} evaluated means or technical solutions are recorded in the second memory block 2. By the control command from the control unit 7, the CPC values Q ₁ , Q ₂ , Q _{3 of the} compared means are supplied to the visualization unit 6. The information is displayed on the visualization block: “PDA values. Q ₁ = 0.51; Q ₂ = 0.468; Q ₃ = 0.748 ".

So, if it is impossible or very difficult to establish the importance of the properties (characteristics) of the compared tools or their variants and the absence of a training sample, the initial data of the most evaluated sample can be used to select the preferred product or compared technical solutions to form a rating scale.

The advantages of the proposed method are:

- the possibility of forming a rating scale in the presence of heterogeneous properties (characteristics) of the evaluated funds,

- the absence of restrictions both on the number of individual indicators and on the number of compared means;

- the ability to determine weighting factors in identifying the importance of the properties of technical solutions without the involvement of highly qualified experts;

- the presence of easily programmable formulas, which leads to the complete automation of the process of determining the preferred tool;

- the ability to perform a hierarchical assessment of complex assets with a breakdown into groups of properties (indicators);

- the feasibility of a visual comparison of quality assessment by the method of two standards with methods of proximity to the best standard and the method of remoteness from the worst standard;

- the ability to identify the following: in comparison with the method of two standards, the method of proximity to the best standard leads to an underestimation of the CCP of the estimated technical solutions and means, and the method of assessing the distance from the worst standard leads to an overestimation of the values of the CCP.

The implementation of the proposed method and device is expected at the first stage - in the form of a software package on a PC, at the second - as a separate device.

Information sources

1. Saati T. Decision making with dependencies and feedbacks: Analytical networks. Per. from English - M.: Book House "Librocom". 2009. - 360 p.

2. Chernorutsky I.G. Decision Making Methods. - SPb. BHV-Petersburg, 2005 - 416 p.

3. Makarov I.D. et al. Theory of choice and decision making. - M .: SCIENCE. Ch. ed. FML. 1992 .-- 328 p.

4. Podinovsky V.V. The use of qualitative information on the importance of criteria for solving multicriteria optimization problems. - M.: VIOLSA, 1977 .-- 36 p.

5. Denisov A.A., Kolesnikov D.N. The theory of large control systems. - L .: Energoizdat, 1982. - 287 p.

6. Chernoscutov A.I. (RU) and other device for assessing the quality of products. Copyright certificate for the invention No. 1597883, IPC G06F 15/46, 1982.

7. Chernoscutov A.I. (RU) and other device for assessing the quality of products. Copyright certificate for the invention No. 1485274, IPC G06F 15/46, 1987.

8. Kazakov I.V. (RU) and others. A device for solving the problems of assessing the quality of arms and military equipment. Application No. 95120833/09, IPC G06F 17/00, 12/07/1995.

9. Makhutov N.A. (RU) and others. A method for determining the quality of a product from the reliable and probabilistic parts of residual defectiveness. Patent No. 2243586, IPC 7 G05B 23/02, G06F 17/00, published: 12/27/2004.

10. Torovin A.H. (RU) A method for ranking search results. Application No. 2008152920/08, G06F 17/30, 12/31/2008.

11. Burba A.A. (RU) and others. A device for evaluating and comparing the functioning efficiency of similar organizations. Patent No. 2363042 (13) C1, IPC G06F 17 /, G06F 7/02, G06N 1/00. Published: July 27, 2009.

12. Chernoscutov A.I. and others. A method and apparatus for selecting a preferred means of information protection. Patent No. 2495482, IPC G06F 17/00.

Claims (1)

A device for identifying a preferred means of information protection, comprising a switch and a first memory unit, characterized in that it includes a unit for normalizing individual indicators, a unit for calculating the values of the numerator and denominator of the fraction, a second memory unit, a visualization unit, and a control unit, while the output of the switch is connected to the information input of the first memory block, connected in series through the unit normalization unit indicators with the unit for calculating the values of the numerator and denominator of the fraction, and in fractionator the sum of the squared lengths of the rating scale and the distance from the point of the worst standard to the point of the compared information protection option in the space of unit indicators, from which the square of the distance from the point of the compared information protection means to the point of the best standard in the space of unit indicators is subtracted, and the fraction denominator is double square of the length of the rating scale, while the maximum fraction corresponds to the preferred means of information protection, the output of the value calculation unit the numerator and denominator of the fraction through the second memory block is connected to the input of the visualization unit, the control outputs of the control unit are connected to the control inputs of all units of the device.

Images