JPS5932233A - Fading discriminating processing system - Google Patents

Abstract

PURPOSE:To discriminate a substantial failure from the effect due to fading, by applying Fourier transformation to an alarm signal in a certain time, and obtaining an approximate value of a frequency representing the peak value of a spectrum. CONSTITUTION:An alarm signal from radio devices RA-RC is detected at a sensor SEN and transferred to a computer CPU in a monitor station SV. The CPU summarizes the alarm signal for a prescribed time T from that point of time so as to discriminate a failure when the generation/restoring information of the alarm signal from the SEN is inputted. When the generation/restoration of the alarm signal continues, the next summarizing is performed from the time of generation/restoration time of the alarm signal just after the 1st summarization. The CPU counts the generation/restoration number of times (n) at each alarm after the end of each summarizing time so as to obtain n/2T. This value is compared with a prescribed value (f), and when the generation of alarm at the end of summarization, it is assumed as the fading in case of f<=n/2T and as the generation of a failure in case of f>n/2T.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field of the Invention The present invention is directed to a system for wireless communication equipment using a conbeater, and the key to generating/restoring an alarm signal emitted by the wireless communication equipment in a counterattack state. The present invention is based on a fading determination processing method that determines flJ based on the characteristics shown by the spectral density based on the theory of Fourier transform, and can distinguish between an original device failure and an alarm occurrence due to fading.

(2) Technological background Traditionally, for example, wireless equipment for unmanned stations located far from the monitoring station has been configured as shown in Figure 1. Monitoring is performed by monitoring station Sv.

For example, when the radio device RAK is detected, an alarm signal emitted by the radio devices rM and RA is detected by the seven-pin νSEN and sent to the monitoring station SV. Here, the possible causes for the generation of the alarm signal are a failure of the wireless device itself, a failure of the black line device facing the wireless device, or a failure due to fading. An alarm signal is generated from the black wire device on both * machines as a synchronized phenomenon. For this reason, unless the monitoring station VC distinguishes between h1 and the same swallow's womb 1 warm, equipment failures, and fading, maintenance personnel will be needlessly dispatched to the site.

(3) Conventional techniques and problems For this reason, conventional techniques have been used to determine whether or not an alarm signal is due to fading. Tsu Q, [,
All alarm-', signal 7c care'' + 10 years old are treated as failure occurrence/ir, i 11'', and fading-(
``If there is any difficulty, the operator is -i'lJ)ji L s
Dust, I
One of the To is! , +421♀Ha), Keisonki (C
PU), the occurrence of alarm signals is summarized in a certain period of time, and ψ44 is determined, and within that aggregation time, the occurrence of the alarm signal starts with no, and ends with uJl (for the alarm with 1). , fading and 111 constants.

Therefore, with the former method, the CPU has to carry out complex processing each time the alarm signal is generated/replaced, which increases the CPU load, and the human The drawbacks of inefficiency 1jt+ and uncertainty of the 1177-article production are serious, and in the latter method, as shown in Fig. If 49. is present, the state of the alarm at the end of aggregation r will be delayed to the occurrence fill, and due to fading it will be erroneously judged as the occurrence of 17.
The CPU performs unnecessary processing (4) Purpose of the invention The present invention takes the above points into consideration, considers the generation of the value code, and considers the old and new values to be one digital signal. The frequency at which the spectral density peaks when subjected to Fourier transform is determined, and this value is used to distinguish normal failures from fading.Faults can be accurately determined without human intervention. The purpose is to provide a judgment processing method.

(5) Structure of the Invention In the above objective 1, a system that captures alarm signals emitted from communication equipment as binary information and monitors failure states of communication and equipment using a computer, Fourier transforms the alarm signals within a certain period of time. This is achieved by a fading ``r1'' processing method that allows the user to find a value close to the frequency at which the peak value of the spectrum is determined, and then use that value to distinguish between an original failure and a cedar echo caused by fading.

((1) Examples of the invention The present invention will be described above based on examples.

Activation of alarm No. 16 that goes off when a death occurs, when a fading occurs, and when a fading occurs in Book 9. The journey of the interval is done by taking advantage of the fact that it can be grasped as a fixed plate by using the following method.

From communication equipment y13. Fluorization of alarm No. 16 /
It can be regarded as the ``Kami no Digital 1''. (Figure 3 (-+1ε11i・)t/ζ,
jd3 As shown in figure (b), this signal is 1t at a certain limited time.
Looking at J-ku, it can be considered as a rectangular wave with a constant frequency, having the same number of waveforms as the original waveform, near (1), L, -C.

Therefore, let T be the length of this same weight, )f iJ, and add f = (-) to ij8, as shown in the frequency spectrum by Fourier summer conversion, shown in >j i1j (f-to be precise, Fig. 3 fc).
shows a peak zl' at 0 on the circumference J of t.

Here, II is Maru Ayu/α old 5th round in T Moebunai. Now, 1.1 general ((- is the faulty number, ; when it is fading, the indication of alarm 1ii of this- is the quality (spectrum) U, and the call is 'JQ+Q'n1
No) Riai, 6 + 541 Δ (shown in al), alarm No. 16 on the first hand, that is, conversion to the 3rd state of life ^I, i; i
4. Since the frequency is 4, the frequency spectrum becomes a distribution with a peak near f-interval, as shown in E+441<1(b). On the other hand, in the case of fading, Figure 4 (c
As shown in Figure 1, the frequency of alarms is continuous at intervals of several seconds to several seconds (the minimum value depends on the hardware), and the frequency spectrum is As shown in Figure (a), the distribution has a peak at a frequency of f=(,4) (n is the number of occurrences/the number of bales in the time T; nal). Therefore, put C1T in a suitable position and "Kura"
), it is possible to quantitatively distinguish between the two.

Utilizing the above characteristics, in the present invention, the data for a predetermined time interval T from the time when the first alarm signal is generated is collected, and the alarm within that time is the occurrence of a bow.
Find the number of restoration times ■1 and calculate the peak frequency (h).
,,This value is compared to a predetermined value, and if it is larger than that, it is assumed to be a -zinc, and a small illness is considered to be a late [・complete birth]. However, if the alarm condition that occurs at the end of the aggregation occurs only in 4 cases, the correction of (phantom) is smaller than the predetermined 14. Also, in the former case, it was determined to be fading.It is assumed that an alarm occurred at the end of the contract, and data collection for the next 7 hours shall begin from 11.

Next, FIG. 1 zi! shows an embodiment of the present invention. This will be explained using FIGS. 5 and 6. In the figure, wireless device RA,
RB, RC7) h et al. I) 7 alarm signal is sensor SE
Detected by N, the computer (CPU) in the monitoring station SV
will be forwarded to.

Total I! - When the alarm signal generation/recovery information is input from the sensor SEN, the machine CPU calculates the annual income of the alarm signal information sent from point h for a certain period of time T, as shown in Figures 1 and 45. , the aggregation t, and all of the alarm status numbers are used to determine a failure. In addition, alarm signal generation/recovery continues, '4n' of alarm signals before and after the first aggregation! :+/ Perform the next aggregation from the recovery time. For the data collected in this way, the computer CPU handles each alarm after the end of each aggregation time.
The number of restorations is counted, and after an aggregation time of 1'' has elapsed, the counter value is divided by 2T to obtain (2-T). and,
This 'l1M is determined by the predetermined directivity f (Fig. 4 (1))
, Compare with (a)) 0 Comparison result , (II/2
1") is smaller than f, then at the time of the mulberry deal and the end of aggregation, the Fummu's bow is ii (the old power, or not, the national power has also been restored). 14 was not considered to be an instantaneous failure, but it was recovered at the end of the operation, and it was determined that the failure was due to failure.

On the other hand, when the contract is open, the alarm @ is activated and the Aseem signal is replaced when the C op1 group is completed. If a rare alarm has occurred (f), assuming that there is no change in the situation as the instant ILJ mountain old, it is assumed that the failure has occurred since 'w51a.On the other hand, 1]/2T is f. The alarm signal was restored at the beginning of the game, and the alarm signal was fading easily. If the name is 1 夛 1, then it is considered to be old, and the result is
) When the firstborn Aji-mu signal is generated, the state L is QL, and L-, 15iJl is the state W when the pieces are assembled. Zuro.

If these criteria are set to 1, the result will be as shown in Table 1. However, when the aggregation ends, an alumm occurs and the next aggregation begins.

To explain Table 1, [Recovery→occurrence J indicates that the alarm signal is in the restored state at the beginning of the aggregation time, and that the alarm signal is in the generated state at the end of the aggregation time. The second aggregation in FIG. i5 corresponds to this case.

Further, "1 occurrence→recovery" indicates that the alarm signal is in the generated state at the beginning of the aggregation time, and the alarm signal is in the recovery state at the end of the aggregation time.

This case corresponds to the third aggregation in FIG. It is possible to prevent a erroneous determination that the unit is affected by a malfunction, and to provide more accurate information to the operator. In addition, the number of times it is incorrectly determined to be a failure is reduced, and overall the computer CP
The effect of reducing the load on U can also be expected.

[Brief explanation of the drawing]

FIG. 1 is a diagram of a monitoring system for wireless communication equipment, FIG. 2 is a diagram for explaining a conventional fading determination method, FIGS. 3 and 4 are diagrams showing the principle of the present invention, and FIG. 5 is a diagram of the present invention. FIG. 6, which is a diagram for explaining the fading determination method in the invention, is a flowchart of the invention. In the figure, RA, RB:+RC is a wireless device, SEN is a sensor-8Vtj: a monitoring station.

Claims (1)

[Claims] In a system that captures the alarm signal emitted from communication equipment as 21 direct current, and monitors all failure states of the communication equipment using a computer, the alarm number 1 within a certain time is replaced by Fourier, and the peak of the spectrum is 11? This fading judgment processing method is characterized in that an approximation lli of the frequency value shown is obtained, and then the original failure, the effect due to no aging, and kA are distinguished from each other.