Patent Application
20230412453
The present invention relates to a technology for supporting maintenance of a radio system.
In conventional maintenance of a radio system, a maintenance engineer collects a variety of information including a reception level, a waveform, an alarm from a radio device, and the like, estimates a failure location based on the collected information, and determines a maintenance action. Such the maintenance method greatly depends on individual knowhow and empirical rules, and thus requires a highly skilled maintenance engineer having years of experience.
NPL 1 discloses a technology for estimating a fault location in a carrier network based on rules.
A conventional maintenance method for a radio system greatly depends on individual knowhow and empirical rules, and thus requires a highly skilled maintenance engineer having years of experience. If such high-level maintenance engineers decrease due to retirement or the like, efficiency and accuracy of the maintenance of the radio system may decrease.
An object of the present invention is to provide a technique capable of efficiently performing maintenance of a radio system without relying on skill and experience of individual maintenance engineers.
A first aspect provides a radio system maintenance support device for supporting maintenance of a radio system.
The radio system maintenance device includes:
The system status information indicates at least one of a radio communication status in the radio system and a device status of a radio device included in the radio system.
The anomaly situation information indicates a situation of anomaly in the radio system.
The failure analysis result information indicates at least one of a failure location in the radio system and a maintenance action required for the failure location.
The failure analysis model is a trained model that receives at least the anomaly situation information and outputs the failure analysis result information, which is generated by learning based on the anomaly situation information and the failure analysis result information that are obtained in the past.
The one or more processors are configured to:
A second aspect provides a radio system maintenance support method for supporting maintenance of a radio system.
The radio system maintenance support method includes:
Failure analysis result information indicates at least one of a failure location in the radio system and a maintenance action required for the failure location.
A failure analysis model is a trained model that receives at least the anomaly situation information and outputs the failure analysis result information, which is generated by learning based on the anomaly situation information and the failure analysis result information that are obtained in the past.
The radio system maintenance support method further includes:
A third aspect provides a radio system maintenance support program. The radio system maintenance support program, when executed by a computer, causes the computer to executed the radio system maintenance support method stated described above. The radio system maintenance support program may be recorded on a non-transitory computer-readable recording medium. The radio system maintenance support program may be provided via a network.
According to the present invention, the failure analysis model is generated based on information on past maintenance records. Using the failure analysis model makes it possible to automatically acquire appropriate failure analysis result information according to a present situation. The acquired failure analysis result information is presented to a maintenance engineer. The maintenance engineer is able to perform a proper maintenance action by referring to the failure analysis result information. It is thus possible to efficiently perform the maintenance of the radio system without relying on skill and experience of individual maintenance engineers.
Embodiments of the present invention will be described below with reference to the accompanying drawings.
The radio system maintenance support device 100 supports maintenance of the radio system 1. Specifically, the radio system maintenance support device 100 acquires system status information indicating a status of the radio system 1. Further, the radio system maintenance support device 100 automatically estimates a failure location in the radio system 1 based on the system status information, and automatically determines a maintenance action required for repairing the failure location. The radio system maintenance support device 100 presents the failure location and/or the maintenance action to a maintenance engineer. The maintenance engineer is able to efficiently perform the maintenance of the radio system 1 by referring to the information presented from the radio system maintenance support device 100.
The radio system maintenance support device 100 is a computer. For example, the radio system maintenance support device 100 is a general-purpose computer such as a personal computer. Alternatively, the radio system maintenance support device 100 may be a computer dedicated to the maintenance of the radio system.
The processor 110 executes a variety of information processing. In particular, the processor 110 executes “failure analysis processing” which will be described later. For example, the processor 110 includes a central processing unit (CPU).
The storage device 120 stores a variety of information 200 necessary for the processing executed by the processor 110. Examples of the storage device 120 include a volatile memory, a non-volatile memory, a hard disk drive (HDD), a solid state drive (SSD), and the like.
The input/output interface 130 is an interface for communicating with the outside of the radio system maintenance support device 100. Examples of the input/output interface 130 include a communication interface, a Universal Serial Bus (USB) interface, a user interface, and the like.
The display device 140 displays a variety of of information. Examples of the display device 140 include a liquid crystal display and the like.
A radio system maintenance support program PROG is a computer program executed by the processor 110. Functions of the radio system maintenance support device 100 are implemented by the processor 110 executing the radio system maintenance support program PROG. The radio system maintenance support program PROG is stored in the storage device 120. The radio system maintenance support program PROG may be recorded on a non-transitory computer-readable recording medium. The radio system maintenance support program PROG may be provided via a network to the radio system maintenance support device 100. In any case, the radio system maintenance support program PROG is executed by the processor 110 to implement the functions of the radio system maintenance support device 100.
The system status information 210 is information indicating a status of the radio system 1. More specifically, the system status information 210 indicates at least one of a radio communication status in the radio system 1 and a device status of each radio device 10. The system status information 210 is acquired from the radio system 1.
The reception level information 211 indicates a reception level (received power) of a radio wave in the radio device 10. The reception level information 211 is acquired from the radio device 10.
The SINR information 212 indicates a signal-to-interference-noise ratio (SINR) of a received signal in the radio device 10. The SINR information 212 is acquired from the radio device 10.
The BER information 213 includes a bit error rate (BER) of a received signal in the radio device 10. The BER information 213 is acquired from the radio device 10.
The fading information 214 indicates fading which is a variation in the reception level in the radio device 10. The fading information 214 is acquired from the radio device 10.
The waveform information 215 indicates a waveform of the reception signal in the radio device 10. The waveform information 215 is acquired from a spectrum analyzer.
The device alarm information 216 indicates an alarm output from the radio device 10 in a case where an anomaly occurs in the radio device 10. The device alarm information 216 is acquired from the radio device 10.
It should be noted that it is not always necessary for the system status information 210 to include all the information exemplified in
The anomaly situation information 220 indicates a situation of anomaly (fault) in the radio system 1. The anomaly situation information 220 is acquired by analyzing the system status information 210 described above.
For example, in a case where the reception level indicated by the reception level information 211 is less than a predetermined reception level threshold, it is determined that the reception level is abnormal. As another example, in a case where the SINR indicated by the SINR information 212 is less than a predetermined SINR threshold, it is determined that the SINR is abnormal. As still another example, in a case where the BER indicated by the BER information 213 is equal to or greater than a predetermined BER threshold, it is determined that the BER is abnormal. As still another example, in a case where a fading level indicated by the fading information 214 is equal to or greater than a certain level, it is determined that abnormal fading occurs. As still another example, comparing the waveform indicated by the waveform information 215 with an ideal waveform makes it possible to determine whether or not the waveform is distorted. In a case where the waveform indicated by the waveform information 215 has “distortion”, it is determined that the waveform is abnormal.
The device alarm information 216 itself indicates that an anomaly occurs in the radio device 10. It should be noted that various types of anomalies can occur in the radio device 10. Examples of the anomaly in the radio device 10 include anomaly of an antenna, anomaly of a processor, and the like.
As shown in
The failure analysis result information 230 is information indicating at least one of a failure location in the radio system 1 and a maintenance action required for the failure location. The failure analysis result information 230 may indicate both the failure location and the maintenance action required for the failure location.
For example, a maintenance engineer who has advanced skill and long-term experience identifies a failure location by a try and error base on the anomaly situation information 220, and determines a maintenance action. However, in this case, the maintenance engineer bears enormous burden, and thus it is not always efficient.
As will be described later, according to the present embodiment, the failure analysis result information 230 can be also automatically acquired by the radio system maintenance support device 100.
The maintenance database 240 is a database storing past maintenance records for the radio system 1. More specifically, the maintenance database 240 indicates a correspondence relationship between the system status information 210, the anomaly situation information 220, and the failure analysis result information 230 that are acquired in the past. For example, the maintenance engineer performs the maintenance of the radio system 1 to acquire the failure analysis result information 230. The system status information 210, the anomaly situation information 220, and the failure analysis result information 230 at that time are associated with each other and stored in the maintenance database 240. Each time the maintenance engineer performs the maintenance, the maintenance database 240 is updated.
As will be described later, in some cases, the failure analysis result information 230 is automatically acquired by the radio system maintenance support device 100. In this case, the system status information 210, the anomaly situation information 220, and the failure analysis result information 230 may be added to the maintenance database 240. That is, the maintenance database 240 may be updated appropriately based on information automatically acquired by the radio system maintenance support device 100.
The failure analysis model 250 is a model that automatically generates and outputs the failure analysis result information 230 based on predetermined input information. The failure analysis model 250 is a trained model generated by learning such as deep learning. Learning of the failure analysis model 250 is performed by utilizing the maintenance database 240 described above.
In Step S111, the information acquisition unit 111 acquires current system status information 210 via the input/output interface 130. The current system status information 210 is stored in the storage device 120.
In Step S112, the anomaly determination unit 112 analyzes the current system status information 210 to acquire current anomaly situation information 220. The method of acquiring the anomaly situation information 220 from the system status information 210 is as described in the above Section 2-2. The current anomaly situation information 220 is stored in the storage device 120.
In Step S113, the failure analysis unit 113 automatically acquires the failure analysis result information 230 by using the failure analysis model 250. As described above, the failure analysis model 250 is generated in advance based on the maintenance database 240 and stored in the storage device 120. The failure analysis unit 113 reads out the failure analysis model 250 from the storage device 120.
In the first example of the failure analysis model 250 shown in
In the second example of the failure analysis model 250 shown in
In Step S114, the result output unit 114 presents the acquired failure analysis result information 230 to the maintenance engineer (user). Specifically, the result output unit 114 displays the acquired failure analysis result information 230 on the display device 140. The failure analysis result information 230 indicates at least one of the failure location in the radio system 1 and the maintenance action required for the failure location. The maintenance engineer is able to perform a proper maintenance action by referring to the failure analysis result information 230.
As described above, according to the present embodiment, the failure analysis model 250 is generated based on information on past maintenance records stored in the maintenance database 240. The radio system maintenance support device 100 is able to automatically acquire the appropriate failure analysis result information 230 according to a current situation by using the failure analysis model 250. The radio system maintenance support device 100 presents the acquired failure analysis result information 230 to the maintenance engineer (user). The maintenance engineer is able to perform a proper maintenance action by referring to the failure analysis result information 230.
Consequently, according to the present embodiment, it is possible to perform the maintenance of the radio system 1 efficiently and with high accuracy, without relying on skill and experience of individual maintenance engineers. Even if high-level maintenance engineers having advanced skill and years of experience decrease due to retirement or the like, high-quality maintenance can be maintained. Further, the number of times of try and errors in maintenance is reduced, and wasteful work is also reduced. Moreover, the time and costs required for training of maintenance engineers can be saved.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2021/003586 | 2/1/2021 | WO |
