CN110809280A - Detection and early warning method and device for railway wireless network quality - Google Patents

Detection and early warning method and device for railway wireless network quality Download PDF

Info

Publication number
CN110809280A
CN110809280A CN201911001575.4A CN201911001575A CN110809280A CN 110809280 A CN110809280 A CN 110809280A CN 201911001575 A CN201911001575 A CN 201911001575A CN 110809280 A CN110809280 A CN 110809280A
Authority
CN
China
Prior art keywords
quality
data
early warning
network
module
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911001575.4A
Other languages
Chinese (zh)
Other versions
CN110809280B (en
Inventor
高一凡
贾利民
刘强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
BEIJING JINHONG XIDIAN INFORMATION TECHNOLOGY Co Ltd
Original Assignee
BEIJING JINHONG XIDIAN INFORMATION TECHNOLOGY Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by BEIJING JINHONG XIDIAN INFORMATION TECHNOLOGY Co Ltd filed Critical BEIJING JINHONG XIDIAN INFORMATION TECHNOLOGY Co Ltd
Priority to CN201911001575.4A priority Critical patent/CN110809280B/en
Publication of CN110809280A publication Critical patent/CN110809280A/en
Application granted granted Critical
Publication of CN110809280B publication Critical patent/CN110809280B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/08—Testing, supervising or monitoring using real traffic
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L27/00—Central railway traffic control systems; Trackside control; Communication systems specially adapted therefor
    • B61L27/40—Handling position reports or trackside vehicle data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/04—Arrangements for maintaining operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H04W4/42—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for mass transport vehicles, e.g. buses, trains or aircraft

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mechanical Engineering (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a detection and early warning method for railway wireless network quality, which comprises the following steps: the invention also provides a detection and early warning device for the quality of the railway wireless network, and the detection and early warning device is used for realizing the method. The invention has the advantages of being capable of rapidly detecting the network quality of each position of the railway, being capable of predicting the change trend of the network quality and early warning the places where the network quality is abnormal in a period of time in the future.

Description

Detection and early warning method and device for railway wireless network quality
Technical Field
The invention relates to a method and a device for detecting the quality of a wireless network, in particular to a method and a device for detecting and early warning the quality of a railway wireless network, and belongs to the field of detection.
Background
The railway wireless network is different from other wireless networks and is specially designed for railway communication, the existing GSM-R system runs over twenty spring and autumn along with the sound of the train, most of the equipment enters the failure high-occurrence period, the dispatching, controlling and other services in the locomotive depend on the GSM-R network system, the quality of the existing GSM-R network at each position of the railway needs to be detected in order to ensure the driving safety,
at present, a railway network system based on the 4G LTE technology is being built in China, new equipment and new technology are debugged in the building process, and a method for effectively detecting the quality of a railway wireless network is also urgently needed.
The detection method adopted by the existing railway network quality is mostly used for collecting network information of each point of a railway to observe the railway network quality in real time, the detection method can be found only after the railway network has problems, the obtained information is relatively backward, only sheep death and reinforcement can be realized, and the railway safety is not facilitated.
In addition, the fault of the railway network equipment is different from the fault of other equipment, the use effect of part of the network equipment is attenuated along with the change of time, for example, the transmission and receiving power of the equipment is weakened along with the increase of the use time, in addition, the phenomena of signal weakening, signal connection strength reduction and the like can be caused by the change of regional environment, although the communication can still be carried out, the communication quality is deteriorated, and certain potential safety hazard is also provided.
Therefore, a method capable of predicting the railway network quality trend is urgently needed to be researched, and the method reasonably predicts the network quality trends of different places, so that key inspection and maintenance are performed on the network equipment related to the places, and the quality of the railway wireless network is improved.
Disclosure of Invention
In order to overcome the above problems, the present inventors have conducted intensive research and developed a method and an apparatus for detecting and warning quality of a railway wireless network, the method including:
s1, collecting data packets;
s2, analyzing the data packet;
s3, analyzing and predicting;
and S4, outputting the prediction result.
According to the present invention, in step S1, the collecting of the data packets is to collect data packets of different locations of the railway for a period of time,
the data packet comprises a geographic position, recording time and network performance data, further, the geographic position is information capable of determining a data packet collection position, preferably latitude and longitude information, the network performance data can be one or more of signal strength, wireless link packet loss rate, time delay, bandwidth-delay product or voice call connection speed, or other related data capable of reflecting network quality,
in a preferred embodiment, the data packet may further include a running speed of the train and a name of a network device communicating with the running speed.
According to the present invention, in step S2, the parsing the data packet is to disassemble and analyze the collected data packet, determine whether the data packet contains valid usable data, and if yes, classify the data according to a predetermined category, including the following sub-steps:
s21, judging whether the data packet contains preset information;
if the judgment result is negative, abandoning the data packet, processing the next data packet,
if yes, S22 is executed.
The preset information comprises information types and information formats required to be collected, wherein the information types comprise one or more of network performance data besides geographic positions and recording time, and more preferably comprise train machine models and equipment names for providing networks.
S22, storing the effective information in the data packet in a classified manner;
the effective information is information meeting preset information conditions, and is classified, stored and arranged into a data table.
And S22, repeating the steps S21 and S22, classifying and storing the information in all the collected data packets, and finishing the analysis process.
In step S3, the analyzing and predicting refers to analyzing and predicting the co-located network performance data in the data network performance data sorted in step S2, and preferably includes the following sub-steps:
s31, merging and sorting the related data of different time and same place obtained in the step S2, and storing to obtain a plurality of groups of arrays with the same place, different time and network performance data;
s32, establishing an autoregressive model for the array obtained in the step S31, and predicting network performance data after a period of time in the future, wherein the method comprises the following sub-processes:
s321, arranging the network performance data in the array into a sequence Xi according to the recording time sequence;
s322, determining a control upper limit UCL and a control lower limit LCL according to the characteristics of different types of network performance data;
s323, judging sequence XiWhether the last value exceeds a control range formed by a control upper limit UCL and a control lower limit LCL,
if the judgment result is over, the early warning is performed, the step S32 is ended, the step S33 is executed,
if the determination result is that the voltage is not exceeded, executing S324;
s324, judging the sequence X in S321iIf the sequence is a non-stationary sequence, preprocessing the sequence, and performing a difference method formula: y isi＝Xi+1-XiPerforming pretreatment until the unit root is tested YiThe result is stable;
s325, for smooth XiEstablishing an autoregressive model for the sequence, wherein the autoregressive model is as follows:
wherein, XtFor the corresponding sequence at time t, WtIs white noise, p is the order of the model, akTo estimate for needThe parameters of (1);
s326, determining an autoregressive model akAnd the autoregressive model is adaptively checked to confirm a valid model,
wherein, akThe value satisfies the minimum sum of the mean square errors of the forward prediction error and the backward prediction error;
s327, predicting the network quality value of the place in a future period of time through the determined autoregressive model, and further obtaining the network quality condition;
and S328, comparing the network quality value predicted in the step S327 with the control upper limit UCL and the control lower limit LCL, and if the network quality value exceeds the range of the control upper limit UCL or the control lower limit LCL, giving an early warning.
And S33, repeating the step S32, processing other groups obtained in the step S31, and predicting the network performance data of different places after a period of time.
On the other hand, the inventor provides a detection and early warning device for the quality of the railway wireless network,
the device comprises an acquisition module 501, a parameter module 502, an analysis module 503, a storage module 504, a prediction module 505 and an input/output device 506;
the obtaining module 501 is configured to obtain data packets of different locations of a railway, where the data packets include geographic locations, recording times, and network performance data.
The parameter module 502 is configured to set and store preset information, where the preset information is the same as the preset information in step S2.
A parsing module 503 for parsing the data packet collected by the obtaining module 501,
the storage module 504 is configured to store the data analyzed by the analysis module 503.
A prediction module 505 for predicting the network quality, which can analyze the prediction according to the method of the above step S3 based on the preset information in the parameter module 502.
And the input and output device 506 is used for inputting preset information in the parameter module 502 and outputting a result predicted by the prediction module 505.
The detection and early warning method and the detection and early warning equipment for the quality of the railway wireless network have the following beneficial effects that:
1. according to the detection and early warning method and the detection and early warning device for the quality of the railway wireless network, the network quality of each position of a railway can be rapidly detected, and an alarm is given for abnormal equipment;
2. according to the detection and early warning method and the detection and early warning device for the quality of the railway wireless network, the change trend of the network quality can be predicted, and early warning can be performed on places where network quality abnormity possibly occurs in a period of time in the future;
3. according to the detection and early warning method and the detection and early warning device for the quality of the railway wireless network, which are provided by the invention, the detection and early warning device can be connected with vehicle-mounted communication equipment, so that the network quality diagnosis can be quickly realized.
Drawings
FIG. 1 is a schematic flow chart of a detection and early warning method for the quality of a railway wireless network according to a preferred embodiment of the invention;
fig. 2 is a schematic diagram illustrating a flow of parsing a data packet in a detection and early warning method for quality of a railway wireless network according to a preferred embodiment of the invention;
FIG. 3 is a schematic diagram illustrating an analysis and prediction flow in a detection and early warning method for the quality of a railway wireless network according to a preferred embodiment of the invention;
fig. 4 is a schematic diagram illustrating an analysis and prediction flow in a detection and early warning method for the quality of a railway wireless network according to a preferred embodiment of the invention.
Detailed Description
The invention is explained in more detail below with reference to the drawings and preferred embodiments. The features and advantages of the present invention will become more apparent from the description.
The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
On one hand, the inventor develops a detection and early warning method for the quality of a railway wireless network, as shown in fig. 1, the method comprises the following steps:
s1, collecting data packets;
s2, analyzing the data packet;
s3, analyzing and predicting;
and S4, outputting the prediction result.
Specifically, in step S1, the collecting of the data packets is to collect data packets of different locations of the railway within a period of time,
in the present invention, the method for collecting the data packets is not particularly limited, and may be performed by a periodic collection method, a fixed distance collection method, or a method of directly reading the log of the vehicle-mounted communication device, the collection method may be freely changed according to the actual requirement,
for example, in a train run, packets are collected every 10s, or every 50 meters of train run,
in the invention, the data packet contains geographic position, recording time and network performance data,
further, the geographical location is information, preferably latitude and longitude,
the network performance data is data capable of representing network quality, and may be one or more of signal strength, packet loss rate, time delay, bandwidth-time delay product or voice call connection speed, or other relevant data capable of representing network quality,
in a preferred embodiment, the data packet may further include a train machine model of the train and a device name providing a network, so as to perform multi-level analysis later,
the vehicle type refers to trains with different running speeds, such as a general express train, a rapid train, a motor train, a high-speed rail and the like.
According to the present invention, in step S2, the parsing the data packet is to disassemble and analyze the collected data packet, determine whether the data packet contains valid usable data, and if yes, classify the data according to the preset information, as shown in fig. 2, including the following sub-steps:
s21, judging whether the data packet contains preset information;
if the judgment result is negative, abandoning the data packet, processing the next data packet,
if yes, S22 is executed.
According to the invention, the preset information includes information category and information format which need to be collected, wherein the information category includes one or more of network performance data besides geographical position and recording time, and more preferably includes train machine model and equipment name providing network, and the information format is a preset fixed format, for example, the geographical position data format may be "113.041, 36.011", the signal strength format may be "-85 dBm", and the like.
Further, when judging whether the data packet contains the preset information, the judgment condition is considered to be met under the condition that the data packet simultaneously has the geographic position, the recording time and the network performance data, and the judgment result is yes, otherwise, the judgment result is no.
S22, storing the effective information in the data packet in a classified manner;
the effective information is the information meeting the preset information condition, the effective information is classified and stored, preferably, the effective information is arranged into a data table form,
illustratively, the table format is as in table 1.
TABLE 1
Figure BDA0002241494960000071
Figure BDA0002241494960000081
And S23, repeating the steps S21 and S22, classifying and storing the information in all the collected data packets, and finishing the analysis process.
In step S3, the analyzing and predicting refers to analyzing and predicting the co-located network performance data in the data network performance data sorted in step S2, and preferably includes the following sub-steps, as shown in fig. 3:
s31, merging and sorting the related data with different time and same place obtained in the step S2, and storing to obtain a plurality of groups of arrays with the same place, different time and network performance data;
the same place is a place with the same or similar longitude and latitude, the range of the same place can be determined according to the actual requirement, for example, the change of the longitude and latitude within 0.10 or 0.01 can be defined as the same place, such as the longitude and latitude 113.011, 36.011-113.111 and 36.111 as the same place, or the longitude and latitude 113.011, 36.011-113.012 and 36.012 as the same place,
preferably, the places with the longitude and latitude changes within 0.01 are the same, the data of the same places are sorted according to the time sequence, and are preferably sorted in the same table, as shown in table 2:
TABLE 2
Figure BDA0002241494960000082
Figure BDA0002241494960000091
S32, building an autoregressive model with the set of arrays obtained in step S31, and predicting network performance data after a period of time in the future, as shown in fig. 4, the method may include the following sub-processes:
s321, arranging the network performance data in the array into a sequence X according to the recording time sequencei；
Further, if the network performance data comprises a plurality of types, each type of data is arranged into a sequence Xi。
S322, determining a control upper limit and a control lower limit according to the characteristics of the performance data of different types of networks;
the upper control limit UCL and the lower control limit LCL are values of which the quality of the wireless network is abnormal, for example, when the network performance data is signal strength, since communication can be guaranteed only when the signal strength is below-100 dBm, the upper control limit UCL is 0dBm, the lower control limit LCL is-100 dBm, and for example, when the network performance data is packet loss rate, the upper control limit UCL is 2%, and the lower control limit is 0; when the network performance data is time delay, the upper control limit UCL is 300ms, and the lower control limit CLC is 0.1 ms; when the network performance data is a bandwidth delay product, the upper control limit UCL is 125kB, and the lower control limit LCL is 75 kB; when the network performance data is the voice call connection speed, the upper limit UCL is 18s, and the lower limit CLC is 2 s.
When the network performance data contains a plurality of data, the sequence X is applied to each group according to different data typesiDesigning different upper control limits UCL and lower control limits LCL,
in a preferred embodiment, the regulation upper limit UCL ═ μ +3 σ; lower regulatory limit LCL ═ μ -3 σ, where μ is sequence Xiσ is the sequence XiAccording to the statistical principle, in the normal distribution, the probability that the fluctuation value of the network quality is normal falls within μ ± 3 σ is 99.73%, and the probability that the value of the fluctuation value exceeds one side, i.e., is greater than μ -3 σ or less than μ +3 σ, is 0.27%/2 ≈ 0.135% ≈ 1 ‰, i.e., it is impossible to be an event.
S323, judging sequence XiWhether the last value exceeds a control range formed by a control upper limit UCL and a control lower limit LCL or not;
if the judgment result is that the network equipment is out of the control range, an alarm is sent out to remind the network equipment of the fault;
if the determination result is that the control range is not exceeded, go to step S324,
sequence XiThe last value is the last detected network performance data, which can represent the current network situation, and if the network situation exceeds the control range, namely the network quality is in a problem, the network device is in a fault.
S324, judging the sequence X in S321iIf the sequence is a non-stationary sequence, preprocessing the sequence into a stationary sequence;
the stationary sequence refers to a joint probability distribution function not along with timeChanged sequence, e.g. if a sequence { X }iI ≧ 0} is a stationary sequence, the joint distribution function of its random variables is:
F(X1,X2,…,Xk)＝F(X1+h,X2+h,…，Xk+h)；(k≥2)
wherein F is represented as a joint distribution function; h belongs to R, R represents goodness of fit, h represents any time, and h is greater than 0; x1,X2,…,XkIs { XiAnd i is more than or equal to 0, otherwise, the random variable is a non-stationary sequence.
In the present invention, it is preferable to use a differential method: y isi＝Xi+1-XiAnd (4) preprocessing is carried out until the unit root test result is stable.
S325, for smooth XiEstablishing an autoregressive model for the sequence;
if a system responds X at time ttAnd only with respect to the correspondence of its previous time instant, and not with respect to the perturbation of its previous time instant into the system, then this system is an autoregressive system, the corresponding model being said autoregressive model,
preferably, the autoregressive model is:
Figure BDA0002241494960000111
wherein, X in the formulatFor the corresponding sequence at time t, WtIs white noise, p is the order of the model, akAre parameters that need to be estimated. The process of establishing the autoregressive model is the process of estimating the model parameters, namely selecting proper parameters to enable the residual error w obtained from the regression model to be obtainedtIs a white noise sequence.
S326, determining an autoregressive model akPerforming adaptive test on the autoregressive model to confirm a valid model;
preferably, akThe value satisfies the minimum sum of the forward prediction error and the backward prediction error mean square error,
preferably, the adaptive test of the autoregressive model includes calculating an estimated value of the original sequence by using the estimation parameters, and calculating to obtain an estimated residual sequence, and if the residual sequence is white noise after test, the autoregressive model can be judged to be valid, that is, a reliable autoregressive model is obtained through the adaptive test, so that the model can effectively predict network performance data at the next moment.
S327, predicting the numerical value of the network performance of the place in a period of time in the future through the determined autoregressive model, and further obtaining the condition of the network quality;
the prediction uses sequence XtObserved value X at time t versus time t + lt-1Predicting with t as the origin and Xt+lThe condition (2) is desirably Xt-lThereby obtaining a numerical value of the network quality and further indicating the condition of the network quality.
And S328, comparing the network quality value predicted in the step S327 with the control upper limit UCL and the control lower limit LCL, and if the network quality value exceeds the ranges of the control upper limit UCL and the control lower limit LCL, giving an early warning.
And S33, repeating the step S32, processing other groups of data in the step S31, and predicting network performance data of different places after a period of time.
According to the detection and early warning method for the quality of the railway wireless network, the prediction accuracy is reduced along with the increase of the prediction time, the prediction accuracy within 10 days can reach more than 70%, the prediction accuracy within 20 days can reach more than 58%, and the accuracy after 25 days is lower.
In a preferred embodiment, the data of different locomotive models in the step S31 are respectively repeated in the step S32 to obtain different network quality prediction results, so as to show the adaptability of the same network to different locomotives,
due to the fact that the different car machines have large running speed difference, if the running speed of a high-speed rail is 3 times that of a common train, the network connection quality of the high-speed rail has difference, data of different car machine models are processed respectively, and the network quality situation of a certain place can be reflected more accurately.
In step S4, the output prediction result is to present the prediction result to the user,
the output mode can be to output the network quality values of all the longitudes and latitudes of the railway in a future period in a tabulated mode, or to mark the prediction result on a railway map in different colors according to different network qualities, or to only display the railway location needing early warning.
In a preferred embodiment, when the prediction result is output, the name of the device providing the network may also be output together, so as to facilitate the inspection and maintenance of the network device with high probability of abnormality.
On the other hand, the inventor provides a detection and early warning device for the quality of the railway wireless network,
the device comprises an acquisition module 501, a parameter module 502, an analysis module 503, a storage module 504, a prediction module 505 and an input/output device 506;
the obtaining module 501 is configured to obtain data packets of different locations of a railway, where the data packets include geographic locations, recording times, and network performance data.
Preferably, the obtaining module 501 is connected to the vehicle-mounted communication device and collects data packets from the vehicle-mounted communication device.
The parameter module 502 is configured to store preset information, where the preset information is the same as the preset information in step S2.
Preferably, the parameter module 502 further has a parameter setting function, and the parameter module 502 can select a data collection method, select the type of the network performance data to be processed, and control parameters such as an upper control limit UCL and a lower control limit LCL,
for example, when the network performance data acquired at 501 includes signal strength, wireless link packet loss rate, time delay, bandwidth delay product, and voice call connection rate, the parameter module 502 may specify that only the voice call connection rate and the packet loss rate are predicted in the subsequent processing process, and the signal strength, the time delay, and the bandwidth delay product are not analyzed and predicted any more, so as to increase the prediction speed and reduce the data processing amount.
And an analyzing module 503, configured to analyze the data packet collected by the obtaining module 501.
The parsing module 503 determines whether the data in the data packet includes preset information, if so, the data is classified according to the format in the preset information to make a corresponding table, and the parsing process is completed after all the collected data packets are classified and made.
The storage module 504 is configured to store the data analyzed by the analysis module 503.
Preferably, the device according to the present invention reserves data in the storage module 504 after each operation, and the reserved data may be used as a part of a data packet to further participate in the process of wireless network quality prediction, so that the prediction result is more accurate.
A prediction module 505 for predicting the network quality, which can analyze the prediction according to the method of the above step S3 based on the preset information in the parameter module 502.
Further, the prediction module 505 can also arrange the wireless network quality values in the data of the same place into a sequence X according to the recording time sequencei，
Optionally, the prediction module 505 can also determine a control upper limit and a control lower limit according to the characteristics of different types of network performance data,
further, the prediction module 505 can determine the sequence XiIf the sequence is a non-stationary sequence, preprocessing the sequence to form a stationary sequence,
in the invention, the prediction module 505 can also be used for smooth XiThe sequence builds an autoregressive model, determines autoregressive model parameters, and performs an adaptive check on the autoregressive model to determine if it is a valid model,
furthermore, the prediction module 505 can predict the network quality value of the location in a future period of time through the determined autoregressive model, so as to obtain the network quality condition,
in a preferred embodiment, the prediction module 505 is further configured to compare the predicted network quality value with a control upper limit UCL and a control lower limit LCL, and perform an early warning if the network quality value exceeds the control upper limit UCL or the control lower limit LCL.
And the input and output device 506 is used for inputting preset information in the parameter module 502 and outputting a result predicted by the prediction module 505.
In the present invention, the input/output device 506 may be a device capable of realizing interaction, such as a keyboard, a mouse, or a touch screen, or may be a simple input/output device, such as a printer.
Examples
Example 1
Firstly, implementation:
and (3) carrying out detection prediction early warning on the signal strength in the wireless network quality along the T209 by the following method.
S1, collecting data packets on the T209 trains;
s2, analyzing the collected data packet to obtain the data as shown in the table 3:
TABLE 3
Figure BDA0002241494960000141
Figure BDA0002241494960000151
S31, sorting the data of the same place with the latitudes of 115.788, 39.256-115.788 and 39.257 in the table 3 according to the chronological order to obtain the data shown in the table 4:
Figure BDA0002241494960000152
s32, sorting the data in the table 4 into a sequence Xi: -50, -63, … …, -81, -72, … …, and determining that the upper limit UCL is 0dBm and the lower limit LCL is-100 dBm according to the signal strength characteristics.
Due to the sequence XiIs a non-stationary sequence, is pre-processed into a stationary sequence to obtain Xi：-50，-56，……，-78，-81，……
For stationary sequence XiEstablishing an autoregressive model:
Figure BDA0002241494960000162
finally determining a by comparing the mean square error of the forward prediction error and the backward prediction error to be minimumk7.3165, and the autoregressive model is determined to be effective after the adaptability test,
and substituting t-240 into the autoregressive model to obtain 115.788 after 240 hours, wherein the signal intensity at 39.256-115.788 is-86 dBm.
S33, repeating the processes of S31 and S32, processing the network performance data of different locations in table 3, obtaining the network quality prediction situation of different locations after 240 hours, as shown in table 5,
TABLE 5
Figure BDA0002241494960000171
And S4, printing the obtained table 5 and specially marking the place needing early warning.
Secondly, effect evaluation:
after waiting for 240 hours, the early warning results obtained in table 5 were compared with the actual test results, and the prediction accuracy was 83%.
In the description of the present invention, it should be noted that the terms "upper", "lower", "inner", "outer", "front", "rear", and the like indicate orientations or positional relationships based on operational states of the present invention, and are only used for convenience of description and simplification of description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The invention has been described in detail with reference to the preferred embodiments and illustrative examples. It should be noted, however, that these specific embodiments are only illustrative of the present invention and do not limit the scope of the present invention in any way. Various modifications, equivalent substitutions and alterations can be made to the technical content and embodiments of the present invention without departing from the spirit and scope of the present invention, and these are within the scope of the present invention. The scope of the invention is defined by the appended claims.

Claims (10)

1. A detection and early warning method for railway wireless network quality comprises the following steps:
s1, collecting data packets;
s2, analyzing the data packet;
s3, analyzing and predicting;
and S4, outputting the prediction result.
2. The detection and early warning method of the quality of the railway wireless network according to claim 1,
in step S1, the collecting of the data packets, in order to collect the data packets of different locations of the railway in a period of time,
the data packet contains geographic location, recording time, and network performance data.
3. The detection and early warning method of the quality of the railway wireless network as claimed in claim 2,
the geographic location is a latitude and longitude,
the network performance data is one or more of signal strength, packet loss rate, time delay, bandwidth-time delay product or voice call connection speed,
the data packet may further include a train machine model of the train and a device name for providing a network.
4. The detection and early warning method of the quality of the railway wireless network according to claim 1,
the step S2 includes the following sub-steps:
s21, judging whether the data packet contains preset information;
if the judgment result is negative, abandoning the data packet, processing the next data packet,
if the determination result is yes, step S22 is executed,
s22, storing the effective information in the data packet in a classified manner;
s23, repeating the steps S21 and S22, classifying and storing the information in all the collected data packets, and completing the analysis process;
the preset information comprises information types and information formats which need to be collected, and the information types comprise one or more of geographic positions, recording time and network performance data.
5. The detection and early warning method of the quality of the railway wireless network according to claim 1,
the step S3 includes the following sub-steps:
s31, merging and sorting the related data with different time and same place obtained in the step S2, and storing to obtain a plurality of groups of arrays with the same place, different time and network performance data;
s32, establishing an autoregressive model by using the array obtained in the step S31, and predicting network performance data after a period of time in the future;
and S33, repeating the step S32, processing other groups in the step S31, and predicting the network performance data of different places after a period of time.
6. The detection and early warning method of the quality of the railway wireless network as claimed in claim 5,
in step S31, the same location is a location with the same or similar latitude and longitude, and preferably a location with the latitude and longitude variation within 1' is the same location.
7. The detection and early warning method of the quality of the railway wireless network as claimed in claim 5,
the step S32 includes the following sub-steps:
s321, arranging the network performance data in the array into a sequence X according to the recording time sequencei；
S322, determining a control upper limit UCL and a control lower limit LCL according to the characteristics of different types of network performance data;
s323, judging sequence XiWhether the last value exceeds a control range formed by a control upper limit UCL and a control lower limit LCL or not;
if the judgment result is over, the early warning is performed, the step S32 is ended, the step S33 is executed,
if the determination result is that the voltage is not exceeded, executing S324;
s324, judging the sequence X in S321iIf the sequence is a non-stationary sequence, preprocessing the sequence into a stationary sequence;
s325, for smooth XiEstablishing an autoregressive model for the sequence, wherein the autoregressive model is as follows:
Figure FDA0002241494950000031
wherein, XtFor the corresponding sequence at time t, WtIs white noise, p is the order of the model, akIs the parameter to be estimated;
s326, determining an autoregressive model akPerforming adaptive test on the autoregressive model to confirm a valid model;
wherein, akThe value satisfies the minimum sum of the forward prediction error and the backward prediction error mean square error,
s327, predicting the numerical value of the network quality of the place in a period of time in the future through the determined autoregressive model, and further obtaining the condition of the network quality;
and S328, comparing the network quality value predicted in the step S327 with the control upper limit UCL and the control lower limit LCL, and if the network quality value exceeds the ranges of the control upper limit UCL and the control lower limit LCL, giving an early warning.
8. The detection and early warning method of the quality of the railway wireless network as claimed in claim 7,
in step S33, repeating step S32 for the data of different car models in step S31 to obtain different network quality prediction results,
the future period of time is 1-20 days.
9. A detection early warning device for railway wireless network quality comprises: an obtaining module 501, a parameter module 502, an analyzing module 503, a storage module 504, a prediction module 505, and an input/output device 506, wherein:
an obtaining module 501, configured to obtain data packets of different locations of a railway,
a parameter module 502, configured to store preset information, which is the same as the preset information in the step S2,
a parsing module 503 for parsing the data packet collected by the obtaining module 501,
a storage module 504 for storing the data parsed by the parsing module 503,
a prediction module 505 for predicting the network quality, which can analyze the prediction according to the method of the above step S3 based on the preset information in the parameter module 502,
and the input and output device 506 is used for inputting preset information in the parameter module 502 and outputting a result predicted by the prediction module 505.
10. The detection and early warning device for the quality of the railway wireless network according to claim 9,
the acquisition module 501 is connected with the vehicle-mounted communication equipment, collects data packets from the vehicle-mounted communication equipment,
the parameter module 502 has a parameter setting function, and the parameter module 502 selects a data collection method, selects the type of network performance data to be processed, and controls upper limit UCL and lower limit LCL parameters.
CN201911001575.4A 2019-10-21 2019-10-21 Detection and early warning method and device for railway wireless network quality Active CN110809280B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911001575.4A CN110809280B (en) 2019-10-21 2019-10-21 Detection and early warning method and device for railway wireless network quality

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911001575.4A CN110809280B (en) 2019-10-21 2019-10-21 Detection and early warning method and device for railway wireless network quality

Publications (2)

Publication Number Publication Date
CN110809280A true CN110809280A (en) 2020-02-18
CN110809280B CN110809280B (en) 2022-10-18

Family

ID=69488699

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911001575.4A Active CN110809280B (en) 2019-10-21 2019-10-21 Detection and early warning method and device for railway wireless network quality

Country Status (1)

Country Link
CN (1) CN110809280B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111654417A (en) * 2020-06-08 2020-09-11 中车青岛四方机车车辆股份有限公司 Evaluation method and device, storage medium, processor and train
CN113543188A (en) * 2021-07-22 2021-10-22 中移（杭州）信息技术有限公司 Wireless network signal quality detection method, terminal device and storage medium
CN116074215A (en) * 2022-12-30 2023-05-05 中国联合网络通信集团有限公司 Network quality detection method, device, equipment and storage medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101170729A (en) * 2007-11-30 2008-04-30 中国移动通信集团重庆有限公司 A GSM network alarming analysis system and network alarm analysis method
CN104091070A (en) * 2014-07-07 2014-10-08 北京泰乐德信息技术有限公司 Rail transit fault diagnosis method and system based on time series analysis
CN104573864A (en) * 2015-01-08 2015-04-29 浪潮软件集团有限公司 Data analysis alarm method based on autoregressive prediction
US20160191336A1 (en) * 2013-09-06 2016-06-30 Huawei Technologies Co., Ltd. Network performance prediction method and apparatus
CN106131865A (en) * 2016-07-19 2016-11-16 浪潮软件集团有限公司 Network quality analysis method based on high-speed rail line
CN107613505A (en) * 2016-07-11 2018-01-19 河南蓝信科技股份有限公司 High-speed railway GSM-R network dynamics detection method and its system
CN110209999A (en) * 2018-02-28 2019-09-06 广州地铁集团有限公司 A kind of mobile unit failure trend prediction method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101170729A (en) * 2007-11-30 2008-04-30 中国移动通信集团重庆有限公司 A GSM network alarming analysis system and network alarm analysis method
US20160191336A1 (en) * 2013-09-06 2016-06-30 Huawei Technologies Co., Ltd. Network performance prediction method and apparatus
CN104091070A (en) * 2014-07-07 2014-10-08 北京泰乐德信息技术有限公司 Rail transit fault diagnosis method and system based on time series analysis
CN104573864A (en) * 2015-01-08 2015-04-29 浪潮软件集团有限公司 Data analysis alarm method based on autoregressive prediction
CN107613505A (en) * 2016-07-11 2018-01-19 河南蓝信科技股份有限公司 High-speed railway GSM-R network dynamics detection method and its system
CN106131865A (en) * 2016-07-19 2016-11-16 浪潮软件集团有限公司 Network quality analysis method based on high-speed rail line
CN110209999A (en) * 2018-02-28 2019-09-06 广州地铁集团有限公司 A kind of mobile unit failure trend prediction method

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111654417A (en) * 2020-06-08 2020-09-11 中车青岛四方机车车辆股份有限公司 Evaluation method and device, storage medium, processor and train
CN113543188A (en) * 2021-07-22 2021-10-22 中移（杭州）信息技术有限公司 Wireless network signal quality detection method, terminal device and storage medium
CN113543188B (en) * 2021-07-22 2023-10-27 中移（杭州）信息技术有限公司 Wireless network signal quality detection method, terminal equipment and storage medium
CN116074215A (en) * 2022-12-30 2023-05-05 中国联合网络通信集团有限公司 Network quality detection method, device, equipment and storage medium
CN116074215B (en) * 2022-12-30 2024-04-19 中国联合网络通信集团有限公司 Network quality detection method, device, equipment and storage medium

Also Published As

Publication number Publication date
CN110809280B (en) 2022-10-18

Similar Documents

Publication Publication Date Title
CN110809280B (en) Detection and early warning method and device for railway wireless network quality
TWI691420B (en) Anomaly diagnosis device, anomaly diagnosis method, and computer program
CN107600110B (en) A kind of vehicle-mounted train groups prognostic and health management system
KR101617349B1 (en) Diagnostic system and method for the analysis of driving behavior
US20090216399A1 (en) Vehicle diagnosing apparatus, vehicle diagnosing system, and diagnosing method
US20230013544A1 (en) Method, Apparatus and System for Detecting Abnormal Operating States of a Device
CN105116870B (en) Fault analysis method, device and system of air conditioning unit
CN104331066A (en) Remote vehicle fault diagnosis method
KR101944070B1 (en) Platform screen door management system and method with big data
CN101272580B (en) Self-adapting mobile base station system reliability estimation method based on feedback
CN108645634A (en) A kind of rail vehicle trouble-shooter
CN108829088A (en) Vehicle diagnosis method, device and storage medium
CN109507992B (en) Method, device and equipment for predicting faults of locomotive brake system components
CN105335599A (en) Vehicle failure diagnosis rate detection method and system
CN114633709A (en) Vehicle fault processing system and method
CN114859871A (en) Vehicle fault diagnosis system and method capable of configuring and modifying fault judgment rules
KR102516227B1 (en) A system for predicting equipment failure in ship and a method of predicting thereof
KR20160062259A (en) Method, system and computer readable medium for managing abnormal state of vehicle
CN112329341B (en) Fault diagnosis system and method based on AR and random forest model
CN115114124A (en) Host risk assessment method and device
RU89248U1 (en) AUTOMATED SYSTEM OF ACCOUNTING, MONITORING AND FORECASTING OF WEAR OF COLLECTORS OF TRACTION LOCOMOTIVE ELECTRIC MOTORS
CN110633314A (en) Internet of vehicles data processing method and device
CN116164921B (en) Method, apparatus and medium for testing rack vibration of fuel cell stack
CN109625025B (en) BTM equipment early warning system
CN110198354A (en) A kind of passenger on public transport comfort based on smart phone and cloud platform and healthy evaluating system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant