AUTOMATIC MONITORING SYSTEMS FOR RADIO-ELECTRONIC EQUIPMENT

N. P. Baida, V. M. Razin, Ya. I. Kapitskii,
G. P. Pavlenko, V. M. Tanaseichuk

(Presented by the scientific seminar
of the Department of Computer Engineering)

Automatic monitoring systems (AMS) for radio-electronic equipment (REE) appeared as a result of progress in the field of automation of radio measurements and at present constitute a new branch of technology. An AMS makes it possible to carry out practically complete automation of the process of checking operability and locating faults in REE, to reduce recovery time, and to reduce the number of service personnel. At the same time, the reliability and objectivity of monitoring are significantly increased, and the information needed for predicting gradual failures is accumulated [1—12].

Depending on the type of connection with the equipment being tested and on the intended purpose, AMS for REE can be divided into three large groups [1]: external, internal (built-in), and combined.

External AMS are used for testing REE of medium complexity; the connection of external AMS with the equipment under test is effected by radio links or by cable communication lines (Fig. 1). External AMS are in most cases universal, i.e., suitable for testing various types of equipment.

Internal (built-in) AMS are functionally connected with the main equipment. Internal AMS (Fig. 2) include built-in instruments and sensors, generators of stimulating signals, switches, data-processing units, and devices for recording and indicating monitoring results. Built-in AMS are in most cases specialized, i.e., suitable for testing one type of equipment. In such systems the sequence for checking parameters is established during the development of the equipment and is not subsequently changed. In universal built-in AMS there is the possibility of changing the monitoring program.

Combined AMS (Fig. 3) are a combination of external and internal monitoring systems. The internal monitoring system ensures checking of the overall operability of REE during the process of func—

![Fig. 1 diagram: AMS — communications equipment — REE under test]

Fig. 1. Simplified block diagram of an external AMS.

[Figure diagram labels: Built-in instruments; Built-in sensors; Generators of stimulating signals; Built-in switches; Data-processing device; Device for recording and indicating monitoring results.]

Fig. 2. Simplified block diagram of an internal (built-in) automatic monitoring system

[Figure diagram labels: Built-in instruments; Built-in sensors; Generators of stimulating signals; Built-in switches; Data-processing device; Automatic test device of the external monitoring system.]

Fig. 3. Simplified block diagram of a combined monitoring system.

…of operation and issues information on the basic parameters of individual equipment units. Fault finding in radio-electronic equipment and a comprehensive all-round check are performed with the aid of an external monitoring system.

Monitoring Methods

The operability of radio-electronic equipment can be monitored by several methods that determine its readiness for action with varying degrees of reliability. The use of one method or another is determined by the purpose of the radio-electronic equipment being monitored, its complexity and reliability, and sometimes

and technical capabilities. A detailed classification of control methods is given in [3].

The main methods for checking the operability of radio-electronic equipment (REA) are functional, tolerance, diagnostic, and preventive control [2].

Functional control is the most widespread method of automatic assessment of the operability of REA. In this case, a disturbing signal is usually applied to the input of the device being tested, and the responses to it are evaluated by the actuation of actuating devices or by the signals on the operating indicators of the equipment being tested. In functional control, the quantitative characteristics of the output parameters of REA are not measured; therefore its reliability is limited, but because of its simplicity of implementation it is widely used in practice.

Tolerance control is carried out by quantitative measurements of the output parameters of the equipment according to the “go/no-go” or “below—go—above” principle. This method makes it possible to obtain, with greater reliability, information about the operability of REA and its output parameters, which determine the technical characteristics of the equipment.

In diagnostic control, faulty units and blocks of the equipment are detected, and sometimes also individual elements that are the cause of failures. The application of this method increases the availability of REA and reduces the time required for its restoration.

As is known [2], the availability coefficient of the equipment is determined by the expression:

\[ K_{\mathrm{r}}=\frac{T_{0}}{T_{0}+T_{\mathrm{v}}}, \]

where

\(T_{0}\) is the mean operating time of the equipment per failure;
\(T_{\mathrm{v}}=t_{\mathrm{un}}+t_{\mathrm{pn}}\) is the mean time of restoration of the equipment;
\(t_{\mathrm{pn}}\) is the mean time spent searching for one fault;
\(t_{\mathrm{un}}\) is the mean time spent eliminating one fault.

To increase the availability coefficient it is necessary to reduce the restoration time of the equipment, which is determined mainly by the quantity \(t_{\mathrm{pn}}\). Several methods have been developed for optimizing the process of fault search [4, 6]: the “half-splitting” (midpoint) method; the “time-probability” method; information search methods; the syndrome method; the pattern-recognition method; and others.

In the “half-splitting” method, unbranched sections of the circuit are divided in half, with each subsequent check being made in the middle of the remaining unchecked part of the circuit.

The “time-probability” method is based on knowledge of the failure rates and of the mean time spent checking different sections of the circuit. The method gives a considerable gain in time when searching for failures with a high rate and is ineffective when searching for failures with a low rate.

Information search methods are based on applying information theory to the process of finding faults in complex radio-electronic systems. The fault-search process is regarded as a process of removing the uncertainty of the state of the REA being monitored. The amount of information obtained in carrying out one checking step is equal to:

\[ H_{k}=-\left[P_{k}\log_{2}P_{k}+(1-P_{k})\log_{2}(1-P_{k})\right], \]

where: $P_k$ is the probability of failure-free operation of the REA with respect to the $K$-th parameter.

The sequence of the fault-search program is compiled from the condition:

\[ H_{k-1} > H_k > H_{k+1}. \]

In the syndrome method, the inspection covers all of the monitored equipment as a whole and its individual elements, down to the level to which faults are traced. In the course of analysis, each inspection result “fit”—“unfit” is assigned to a definite syndrome (feature). A library of such syndromes is compiled in advance, programmed, and entered into the memory device of the monitoring system. The advantage of the method is that, in order to test the REA, it is not necessary to apply stimulating signals.

The essence of the pattern-recognition method consists in dividing the $m$-dimensional space of values of the REA output parameters into $P$ subspaces of fault features. This method, like the syndrome method, is implemented with the aid of a digital computer.

In preventive or predictive monitoring, the output parameters and the main units of the REA are periodically checked so that, on the basis of the accumulated information about the state of the equipment, it can be predicted which assembly or unit will be the cause of failure within a definite time interval. Timely prevention of failures makes it possible to substantially increase the reliability of REA operation. The basis of preventive monitoring is formed by the method of boundary tests and statistical methods for predicting failures.

In practice, all methods of automatic monitoring of equipment operability are used; however, the most widespread is the method of tolerance monitoring in combination with diagnostic monitoring and prediction of equipment failures.

Characteristics of AMS

The properties of any AMS are determined by a number of characteristics. The most general of them are [1, 4, 6]:

— reliability of monitoring results;
— effectiveness of the AMS;
— reliability;
— weight and overall dimensions;
— total number of monitored parameters;
— completeness of monitoring;
— accuracy of parameter measurement;
— possibility and accuracy of predicting the state of the REA being checked;
— availability and type of self-monitoring;
— possibility of automatic switching-on of reserve units.

The reliability of monitoring results characterizes the ability of the monitoring system to correctly reflect the real state of the REA being checked and is the most important technical characteristic of any AMS.

At present, the concept of monitoring effectiveness is not unambiguous. Some authors [8] understand the effectiveness of a system as the ratio of the effect it produces to the costs. Others, in determining effectiveness, take into account the accuracy characteristics of the system and the total costs of monitoring. Some authors [9—11] relate effectiveness to the probability that the system will accomplish the assigned task—

sensors. In the general case, the effectiveness of an ACS is understood as a qualitative assessment of the solution of the overall monitoring problem with a reliability no lower than the specified one within the specified time.

The completeness of monitoring characterizes what percentage of the total number of parameters determining the operability of the equipment is covered by monitoring:

\[ \Pi_k=\frac{N_k}{N_\Sigma}, \]

where:

\(N_k\) is the number of parameters covered by monitoring;
\(N_\Sigma\) is the total number of parameters characterizing the operability of REE.

Principles for Constructing Automatic Monitoring Systems for Radio-Electronic Equipment

A generalized structural diagram of an ACS for REE is shown in Fig. 4. A real monitoring system must contain only those units that make it possible to ensure that its characteristics meet the requirements of operation. In the general case, an ACS for REE includes the following devices [4, 6, 13, 14]:

Labels in the diagram:
Tested REE; device for communication with the tested REE; generators of stimulating and reference signals; conversion devices; computing device; control device; input device; output device.

Fig. 4. Generalized structural diagram of an ACS

device for communication with the tested REE;
generators of stimulating and reference signals;
conversion devices;
computing device (CD);
control device;
input device;

output device.

Communication devices contain equipment for transmitting the monitored and stimulating signals over radio links or over cable communication lines, as well as signal normalizers and switches. The latter are implemented either with contact elements (relays, stepping switches, magnetically controlled contacts) or with contactless elements (diodes, transistors, gas-discharge tubes).

Generators of stimulating and reference signals produce the input stimulating (exciting) signals required for monitoring, as well as reference signal values with which the measurement results are compared. They are subdivided into generators of electrical voltages and currents, time intervals, frequency generators, pulse generators, etc.

Converting devices (analog-to-digital converters) convert the monitored signals into a form convenient for input into a digital computer.

The computing device processes information on the state of the radio-electronic equipment under test according to a specified program.

The input device stores monitoring programs and determines the operating mode of the other units of the AMS. For this purpose, punched cards, punched tapes, magnetic drums, storage devices on ferrite cores, magnetic tapes, and magnetic disks are used.

The output device prints and indicates the monitoring results.

To increase the reliability of monitoring results, self-test circuits are introduced into AMS devices. Partial self-checking of the AMS is carried out during operation, while a complete self-check is performed before the start of operation, in the preventive-maintenance mode before the start of operation, in the preventive-maintenance mode, or after a negative test result has been obtained.

At present, chiefly combined AMS are being developed for monitoring individual types of radio-electronic equipment. Attempts are being made to create universal combined AMS which, after a minor change in the structural diagram by replacing or adding certain standard units (signal generators, switches, sensors, etc.), can be used to test various radio-electronic complexes.

Present State and Prospects for the Development of Automatic Monitoring Systems for Radio-Electronic Equipment

In recent years, extensive work has been carried out in our country and abroad on the investigation, development, production, and operation of AMS for radio-electronic equipment. These systems differ in purpose, complexity, cost, programming principles, control and information-processing principles, and also in the technical level of their implementation [1,15].

Modern AMS for radio-electronic equipment are built into the equipment or attached to it. They are implemented using integrated circuits (ICs) with the use of digital computers. A promising method for developing AMS is considered to be the one called “automatic system self-monitoring” (ASS) [16].

According to the basic principle of ASS, monitoring equipment must be designed simultaneously with the operating elements of the equipment at every stage in the development of a radio-electronic system.

Using this method, a system for monitoring the equipment of spacecraft that can operate for many years without human intervention is being developed in the United States [17]. The aim of the work is to create an on-board computer for spacecraft that requires no maintenance for 10 years. All basic operations for error detection are performed by the computer. On the basis of data from the controlling diagnostic unit, the necessary reserve is switched in. All units are implemented on integrated circuits.

However, the existing schemes for constructing monitoring systems have a number of shortcomings which sometimes make it impossible to ensure high reliability and effectiveness of monitoring.

The principal shortcomings of modern automatic monitoring systems (AMS) are [1]:

great complexity;
limited logical operations;
low reliability;
insufficient noise immunity;
lack of self-learning capability.

In this connection, extensive work is being carried out to investigate the possibility, on the one hand, of simplifying monitoring systems and increasing their reliability, and, on the other hand, of expanding the logical operations of AMS and improving measurement accuracy.

Simplification of the monitoring system can be achieved as a result of a systems approach to the development of the main and monitoring equipment, for which the ACC method described above is used.

In order to increase reliability, modern AMS are built from highly reliable elements (integrated circuits, transfuxors, parametrons, etc.).

The possibility is being studied of using nonelectrical signals for transmitting information in order to increase the resistance of systems to external electromagnetic influences. Thus, for example, monitoring systems have recently been created in which the transmission and logical processing of information were carried out with the aid of light. In addition, the possibility of constructing monitoring systems on the basis of pneumatics is being investigated.

The introduction of monitoring methods based on the level of infrared radiation (IR) is proposed. Measurement of the IR of electronic circuits makes it possible to detect defects that are not revealed by ordinary monitoring methods [18].

A promising monitoring method is considered to be one that uses changes in the parameters of its mathematical model, carried out until its characteristics coincide with the output characteristics of the real apparatus. The principal advantage of this method is the absence of any need for additional monitoring points [19].

In recent years, research has begun to determine the possibilities of creating monitoring systems on the principles of neurocybernetics, using electrical models of neurons. The results obtained show that this will make it possible to expand significantly the logical capabilities of systems and to create self-learning monitoring systems.

At present, universal AMS are being developed on the basis of standardized functional units that can be adapted for checking various classes of equipment.

New areas of application for AMS are being outlined. For example, it is proposed to create equipment for measuring and indicating the parameters of the psychological state of the pilot on board a supersonic passenger aircraft [20]. Automatic monitoring systems will become an integral part of self-organizing systems, in which they will be...

perform the functions of continuous diagnostics and self-repair, and, in the presence of an irreparable failure, select the optimal operating mode of the equipment with degraded characteristics.

REFERENCES

1. M. D. Lebedev. The State and Development of Automatic Monitoring Systems. Energiya, Moscow–Leningrad, 1967.

2. V. A. Dolgov. Built-in Automated Monitoring Systems. Energiya, Moscow–Leningrad, 1968.

3. Fundamentals of Operation of Radio-Electronic Equipment. Edited by V. Yu. Lavrinenko. Vysshaya shkola, Moscow, 1968.

4. Automation of Radio Measurements. Edited by V. P. Balashov. Sovetskoe radio, 1966.

5. Introduction to Technical Diagnostics. Edited by K. B. Karandeev. Energiya, Moscow, 1968.

6. V. V. Somik, N. G. Konkov. An Automatic Device Checks an Aircraft and a Rocket. Moscow, Voenizdat, 1967.

7. Electrical Methods of Automatic Monitoring. Edited by K. B. Karandeev. Energiya, Moscow–Leningrad, 1965.

8. V. K. Burtsev, D. V. Svicharnik. On the reliability and effectiveness of automatic monitoring systems. Priborostroenie, No. 6, 1963.

9. B. V. Vasil'ev, B. A. Kozlov, L. G. Tkachenko. Reliability and Effectiveness of Radio-Electronic Devices. Sovetskoe radio, 1965.

10. A. S. Kasatkin, I. V. Kuz'min. Evaluation of the Effectiveness of Automated Monitoring Systems. Energiya, 1967.

11. V. A. Lutskii. Calculation of the Reliability and Effectiveness of Radio-Electronic Equipment. Publishing House of the Academy of Sciences of the Ukrainian SSR, Kiev, 1963.

12. Maintainability of Radio-Electronic Equipment. Collection of translations edited by O. F. Poslavskii. Sovetskoe radio, Moscow, 1964.

13. Elements of Automatic Monitoring Systems. Edited by P. I. Kuznetsov. Energiya, Moscow–Leningrad, 1967.

14. Automatic Testing of Aircraft and Rocket Equipment. Collection of articles edited by V. A. Bodner. Moscow, Izd-vo inostrannoi literatury, 1962.

15. Elektronika (translated from English), 1968, vol. 41, No. 10, pp. 11–19.

16. Elektronika (translated from English), 1967, vol. 40, No. 18, p. 49.

17. Elektronika (translated from English), 1967, vol. 40, No. 7, pp. 16–28.

18. Elektronika (translated from English), 1967, vol. 40, No. 1, pp. 83–86.

19. Elektronika (translated from English), 1968, vol. 41, No. 9, pp. 23–32.