CN108446712B - ODN network intelligent planning method, device and system - Google Patents

ODN network intelligent planning method, device and system Download PDF

Info

Publication number
CN108446712B
CN108446712B CN201810108108.0A CN201810108108A CN108446712B CN 108446712 B CN108446712 B CN 108446712B CN 201810108108 A CN201810108108 A CN 201810108108A CN 108446712 B CN108446712 B CN 108446712B
Authority
CN
China
Prior art keywords
obd
stationing
result data
data
distribution
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.)
Active
Application number
CN201810108108.0A
Other languages
Chinese (zh)
Other versions
CN108446712A (en
Inventor
李毅
杨莉
杨鹤鸣
林涛
陈华荣
赵松君
邓晓旭
林焜煜
赵勇
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.)
Guangdong Planning and Designing Institute of Telecommunications Co Ltd
Original Assignee
Guangdong Planning and Designing Institute of Telecommunications 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 Guangdong Planning and Designing Institute of Telecommunications Co Ltd filed Critical Guangdong Planning and Designing Institute of Telecommunications Co Ltd
Priority to CN201810108108.0A priority Critical patent/CN108446712B/en
Publication of CN108446712A publication Critical patent/CN108446712A/en
Application granted granted Critical
Publication of CN108446712B publication Critical patent/CN108446712B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—Network analysis or design using statistical or mathematical methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q11/00—Selecting arrangements for multiplex systems
    • H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062—Network aspects
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q11/00—Selecting arrangements for multiplex systems
    • H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062—Network aspects
    • H04Q2011/0079—Operation or maintenance aspects

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Pure & Applied Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention relates to an ODN intelligent planning method, a device and equipment, wherein the method comprises the following steps: acquiring an OBD distribution influence factor set of a region to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set; processing a historical OBD stationing sample data set through a neural network model to obtain a current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter; performing similarity clustering processing on the service sample data set to obtain OBD stationing clustering result data; carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data; and comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data, and generating current OBD stationing optimization data according to the comparison result. By adopting the method, the OBD distribution in the ODN network of the area to be planned can be optimized, and the accuracy and efficiency of the OBD distribution are improved.

Description

ODN network intelligent planning method, device and system
Technical Field
The invention relates to the technical field of communication, in particular to an ODN intelligent planning method, device and system.
Background
With the development of communication technology, although mobile communication is developed at a remarkable speed nowadays, due to the limited bandwidth, the terminal size cannot be too large and the display screen is limited, and for a fixed terminal with relatively superior performance, the implementation of FTTH (Fiber to the home) can enable a user to experience a better communication mode.
FTTH refers to the manner in which optical fiber is directly routed to a subscriber home, and all access modes are optical fiber lines from the communication office end to the subscriber home. An Optical Distribution Network (ODN) is a main part of the whole Optical Network construction, the ODN is composed of feeder Optical fibers, Optical splitters (OBD: On-Board diagnostics) and branch lines, and an Optical Line Terminal (OLT: Optical Line Terminal) and a plurality of Optical Network units (ONU: Optical Network Unit) are connected to provide bidirectional transmission of Optical signals. And the OBD stationing of the ODN networking planning is an important link in the FTTH construction, and the quality of the OBD stationing of the ODN networking planning directly affects the efficiency of the FTTH construction.
The planning process is complex and the time is long due to more influence factors of the OBD distribution; in addition, the actual construction scenes are complex and various, the number and the position of DP (electric distribution box) in the planning region, the user distribution characteristics and the cell properties are difficult to be comprehensively considered, the approximate user scale needing to be covered in the planning region is predicted, and the like, so that the equivalent user capacity and the corresponding OBD point distribution unit are difficult to evaluate. In the traditional technical scheme, manual planning is basically adopted in the whole process, on-site survey and checking and confirming of stock resources are carried out, judgment needs to be made by personal experience during planning and design, and detailed standard specifications do not exist. In the implementation process, the inventor finds that at least the following problems exist in the conventional technology: the traditional manual ODN network planning and point distribution has long resource construction period, non-standardization, waste investment and low ODN networking planning and point distribution efficiency.
Disclosure of Invention
Based on this, the invention provides an ODN network intelligent planning method, device and system, which are necessary to solve the problem of low point placement efficiency in ODN networking planning in the conventional technical scheme.
In order to achieve the above object, in one aspect, an embodiment of the present invention provides an ODN network intelligent planning method, including the following steps:
acquiring an OBD distribution influence factor set of a region to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set; processing a historical OBD stationing sample data set through a neural network model to obtain a current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter; wherein, the neural network model is a BP neural network model;
performing similarity clustering processing on the service sample data set to obtain OBD stationing clustering result data;
carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data;
and comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data, and generating current OBD stationing optimization data according to the comparison result.
In one embodiment, the step of obtaining the current OBD stationing training parameter by processing the historical OBD stationing sample data set through the neural network model includes:
acquiring an input layer vector and an input layer weight matrix of a historical OBD stationing sample data set;
performing linear processing on the input layer vector and the input layer weight matrix to obtain a hidden layer input vector; performing reserved mapping function processing on the hidden layer input vector to obtain a hidden layer output vector;
carrying out linear processing on the hidden layer output vector and a hidden layer weight matrix corresponding to the hidden layer output vector to obtain an output layer input vector; and carrying out reserved mapping function processing on the input vector of the output layer to obtain the current OBD stationing training parameter.
In one embodiment, the similarity cluster is an SNN cluster; the nearest neighbor classification is a KNN classification;
the method for obtaining OBD stationing clustering result data by clustering the service sample data through similarity comprises the following steps:
carrying out SNN clustering processing on the service sample data set to obtain OBD stationing clustering result data;
the step of obtaining the optimized OBD stationing and classifying result data by carrying out nearest-neighbor classification on the OBD stationing and clustering result data comprises the following steps of:
and carrying out KNN classification processing on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data.
In one embodiment, the step of obtaining OBD stationing clustering result data by performing similarity clustering on the service sample data set includes:
comparing the current OBD stationing training parameter with the OBD stationing clustering result data; obtaining OBD abnormal point layout data;
and carrying out nearest neighbor classification processing on the OBD abnormal point distribution data to obtain optimized OBD point distribution classification result data.
In one embodiment, the step of generating the current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data includes:
calibrating the current OBD stationing training parameters according to the comparison result to obtain calibrated current OBD stationing classification result data;
and comparing the calibrated current OBD stationing classification result data with the optimized OBD stationing classification result data, and updating the current OBD stationing optimization data.
In one embodiment, the step of obtaining the OBD placement impact factor set of the area to be planned comprises the following steps:
carrying out data preprocessing on the OBD distribution influence factor set to obtain a historical OBD distribution sample data set and a service sample data set;
the data preprocessing comprises data cleaning, data encoding and data reconstruction.
In one embodiment, the historical OBD stationing sample data set comprises a copper coverage influencing factor data set and a light coverage influencing factor data set;
the service sample data set comprises an influencing factor data set of the region grid segmentation unit.
On the other hand, the embodiment of the invention also provides an ODN intelligent planning device, which comprises:
the data acquisition unit is used for acquiring an OBD distribution influence factor set of an area to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set; the neural network processing unit is used for processing the historical OBD stationing sample data set through the neural network model to obtain current OBD stationing training parameters and current OBD stationing classification result data corresponding to the current OBD stationing training parameters; wherein, the neural network model is a BP neural network model;
the clustering unit is used for clustering the service sample data through similarity to obtain OBD stationing clustering result data;
the classification processing unit is used for carrying out nearest classification processing on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data;
and the comparison optimization unit is used for generating current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
On the other hand, the embodiment of the invention also provides an ODN intelligent planning system, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the steps of the ODN intelligent planning method when executing the computer program.
On the other hand, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned ODN network intelligent planning method are implemented.
One of the above technical solutions has the following advantages and beneficial effects:
inputting a historical OBD stationing sample data set of a region to be planned into a neural network model to obtain current OBD stationing classification result data; performing similarity clustering processing on a service sample data set of an area to be planned, and performing establishment verification on OBD distribution to obtain OBD distribution clustering result data; carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data; and generating current OBD distribution optimization data according to a comparison result obtained by comparing the current OBD distribution classification result data with the optimized OBD distribution classification result data, thereby optimizing the OBD distribution in the ODN network of the area to be planned and improving the accuracy and efficiency of the OBD distribution.
Drawings
FIG. 1 is a diagram of an application environment of an ODN intelligent planning method in one embodiment;
FIG. 2 is a diagram illustrating an exemplary scenario for implementing the ODN intelligent planning method;
FIG. 3 is a flow diagram illustrating an ODN intelligent planning method according to an embodiment;
FIG. 4 is a schematic flow chart diagram illustrating the processing steps of the neural network in one embodiment;
FIG. 5 is a schematic diagram showing the structure of the neural network processing step in one embodiment;
FIG. 6 is a schematic structural diagram of a KNN classification processing step in one embodiment;
FIG. 7 is a schematic flow chart of the iterative optimization step in one embodiment;
FIG. 8 is a flow chart illustrating an ODN intelligent planning method according to another embodiment;
FIG. 9 is a flow chart diagram of an ODN intelligent planning method in one embodiment;
FIG. 10 is a flowchart of an ODN intelligent planning method in one embodiment;
FIG. 11 is a block diagram illustrating the structure of an ODN intelligent planning apparatus according to an embodiment;
fig. 12 is an internal structural diagram of the ODN network intelligent planning system in one embodiment.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
The ODN intelligent planning method provided by the invention can be applied to the application environment shown in figure 1. The server 102 may obtain ODN network planning data of the area to be planned 104. The server 102 can obtain a historical OBD stationing sample data set and a service sample data set of an area to be planned; processing a historical OBD stationing sample data set through a neural network model to obtain current OBD stationing classification result data; performing similarity clustering and nearest neighbor classification on the service sample data to obtain optimized OBD stationing classification result data; and according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data, current OBD stationing optimization data can be generated. The server 102 may be, but not limited to, various personal computers, notebook computers, tablet computers, and the like, and the server 102 may also be implemented by an independent server or a server cluster formed by a plurality of servers. The area to be planned 104 may be an industrial park, a residential district, and so on.
The invention discloses an ODN intelligent planning method, a device and a system embodiment, wherein an application scene is as follows:
fig. 2 is a diagram of an application scenario of the ODN network intelligent planning method in an embodiment, and as shown in fig. 2, the OND network planning is generally divided according to the selection of the splitting number, the OBD installation position, and the splitting ratio. The conventional OBD planning distribution of the OND network of the area to be planned mainly includes two types: first, the cell two-level splitting mode. The primary OBD is intensively arranged in an optical cross connecting box or an optical cross connecting room in a cell, the secondary OBD is arranged in each building, and various combination modes such as 1:4+4 × 1:16, 1:8+8 × 1:8 or 1:16+16 × 1:4 are selected according to the number of users in the building. The networking mode is generally suitable for scenes with low user density, such as low-rise residential districts and industrial parks. Second, two-stage light splitting in the building. The first-level OBD is arranged on the first floor or basement of each building and adopts the light splitting ratio of 1:8 or 1: 16; the second-level OBD is arranged on a floor, and the splitting ratio is selected according to the splitting ratio of the first-level optical splitter, so that the total splitting ratio reaches 1: 64. The networking mode is also suitable for high-rise residential buildings or large-bay business office buildings, the initial investment of network construction is low, the utilization rate is high, and the later expansion is simple. The conventional technical scheme is that an OBD planning deployment scheme is manually formulated according to OBD equipment characteristics and port quantity by combining building forms, user distribution and road or pipeline distribution conditions, and a target planning area is subjected to simple spatial range distribution, so that the phenomena of long resource construction period, non-standardization, waste investment and the like occur, and the overall efficiency of an ODN networking mode is low.
According to the embodiment of the invention, the OBD planning deployment scheme can be quickly generated by analyzing the big data of the building scheme of the total-storage optical-width coverage cell according to the storage DP resource, the OBD resource, the client and building information and the vector distribution condition of the target planning area and combining a neural network model and a statistical (clustering and classifying) algorithm, so that the manual checking and intervention are reduced, the labor cost is saved, the construction period is shortened, and the accuracy and the efficiency of the OBD point distribution are improved.
In one embodiment, as shown in fig. 3, an ODN network intelligent planning method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
step S310, obtaining an OBD distribution influence factor set of an area to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set.
The area to be planned refers to an area where optical fibers to be planned enter a home; the area to be planned may be an industrial park or a residential district. The OBD distribution impact factor set refers to a resource data set which affects OBD distribution, and may include impact factors of OBD distribution of several to-be-planned areas, and each impact factor may include several given factors. The historical OBD stationing sample data set may be OBD stationing data already existing in the area to be planned. The service sample data set may be OBD stationing data determined according to a property characteristic of the area to be planned.
Specifically, an OBD stationing influence factor set of the to-be-planned region is obtained, wherein a historical OBD stationing sample data set can be obtained according to historical stationing data of the to-be-planned region, and a service sample data set can be obtained according to characteristics of the to-be-planned region, such as an actual environment.
And step S320, processing the historical OBD stationing sample data set through the neural network model to obtain the current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter.
The neural network model refers to a network model which is organized by a plurality of logic units according to different levels, wherein an output variable of each layer can be used as an input variable of the next layer. The current OBD stationing training parameter refers to an output parameter processed properly by a neural network. The current OBD stationing classification result data refers to classification result data processed by a historical OBD stationing sample data set through a neural network model.
Specifically, the historical OBD stationing sample data set is input into a neural network model, and current OBD stationing training parameters and current OBD stationing classification result data can be obtained through neural network processing.
And step S330, clustering the service sample data set through similarity to obtain OBD stationing clustering result data.
Wherein, the similarity clustering process refers to grouping similar things into one group (unsupervised learning training). The OBD stationing clustering result data refers to stationing optimization data obtained through similarity clustering processing.
Specifically, the service sample data set is subjected to similarity clustering processing, and the OBD stationing clustering result data after being divided again can be obtained through spatial clustering of different sizes, shapes, densities and the like on the service sample data set.
And step S340, carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data.
The nearest classification processing refers to classification into existing categories (supervised learning training) according to the features or attributes of the data. The optimized OBD stationing classification result data refers to classification result data obtained through nearest neighbor classification processing.
Specifically, the OBD stationing clustering result data is subjected to nearest neighbor classification, so that nearest neighbor classification can be performed on edge data points (service sparse zones) in the OBD stationing clustering result data, and optimized OBD stationing classification result data is obtained.
And step S350, generating current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
Specifically, comparing the current OBD stationing classification result data obtained by the neural network model processing with the optimized OBD stationing classification result data obtained by the nearest classification processing, for example, if OBD abnormal point data exists in the current OBD stationing classification result data, classifying the OBD abnormal point data (moving the OBD abnormal point position or deleting the OBD abnormal point) by comparing the optimized OBD stationing classification result data, thereby generating the current OBD stationing optimization data.
In the embodiment of the ODN intelligent planning method, the current OBD distribution classification result data is obtained by acquiring an OBD distribution influence factor set of the region to be planned and inputting a historical OBD distribution sample data set of the region to be planned into a neural network model; similarity clustering processing is carried out on the service sample data set of the area to be planned, and determination verification is carried out on OBD distribution to obtain OBD distribution clustering result data; carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data; and generating current OBD distribution optimization data according to a comparison result obtained by comparing the current OBD distribution classification result data with the optimized OBD distribution classification result data, thereby optimizing the OBD distribution in the ODN network of the area to be planned and improving the accuracy and efficiency of the OBD distribution.
It should be noted that, the neural network model, the similarity cluster, and the nearest neighbor classification processing process in the above embodiments may be repeatedly trained processing processes, for example, after the current OBD stationing optimization data is generated, the neural network model processing may be continued on the current OBD stationing optimization data, and the processing result is continuously compared with the optimization results of the similarity cluster and the nearest neighbor classification processing, so as to obtain suboptimal OBD stationing optimization data.
In one embodiment, step S320 includes:
step S410, obtaining an input layer vector and an input layer weight matrix of the historical OBD stationing sample data set.
Step S420, carrying out linear processing on the input layer vector and the input layer weight matrix to obtain a hidden layer input vector; and carrying out reserved mapping function processing on the hidden layer input vector to obtain a hidden layer output vector.
Step S430, carrying out linear processing on the hidden layer output vector and the hidden layer weight matrix corresponding to the hidden layer output vector to obtain an output layer input vector; and carrying out reserved mapping function processing on the input vector of the output layer to obtain the current OBD stationing training parameter.
In particular, the neural network model may have an input layer, a hidden layer, and an output layer, wherein output data of the input layer is input data of the hidden layer, and output data of the hidden layer is input data of the output layer. Acquiring an input layer vector and an input layer weight matrix of a historical OBD (on-board diagnostics) stationing sample data set, and performing linear processing on the input layer vector and the input layer weight matrix to obtain a hidden layer input vector; performing reserved mapping function processing on the hidden layer input vector to obtain a hidden layer output vector; carrying out linear processing on the hidden layer output vector and a hidden layer weight matrix corresponding to the hidden layer output vector to obtain an output layer input vector; and carrying out reserved mapping function processing on the input vector of the output layer to obtain the current OBD stationing training parameter. Preferably, the linear processing may be weighted linear processing.
TABLE 1 OBD Point impact factor set
Figure GDA0002529930150000091
In the above embodiment, the current OBD stationing training parameter can be obtained by processing three layers (an input layer, a hidden layer and an output layer) of the neural network on the historical OBD stationing sample data set, so as to provide portability for subsequent improvement of OBD stationing, and in addition, the current OBD stationing training parameter can be repeatedly trained through the neural network model, so as to obtain more optimized OBD stationing data.
In one embodiment, the set of OBD placement impact factors for the area to be planned may be as shown in table 1.
The grid cell is a pattern that uses a grid when a large-scale physical area is divided and managed, and each grid is a minimum cell for grid management after the grid is divided in a grid manner. Grid management is applied in area management, and grid cells can be community and cell waiting planning areas.
In one embodiment, the neural network model processing step is specifically as follows:
acquiring an influence factor data set x influencing OBD distribution in a region to be plannednCombining the characteristics of the neural network, the number of nodes of the hidden layer is hp. The output layer is OBD intelligent distribution point yqSpecifically, the results are shown in Table 2.
TABLE 2 input layer, hidden layer and output layer parameters
Number of neurons in input layer Number of hidden layer neurons Number of neurons in output layer
xn(k) hp(k) yq(k)
Where k represents the number of neurons (number of nodes). And dividing the influence factor data set into a test set and a plurality of training sets, and carrying out forward training on the test set and the plurality of training sets through a neural network model.
1. Initializing input layer, hidden layer, and output layer parameters (as shown in Table 3)
Note that others in the above table refer to conditions other than E (k +1) < E (k) and E (k +1) > 1.04E (k); w is anpRepresenting the weight of each variable in an input layer, wherein n refers to n row observed values, and q refers to q columns of variables; bpHidden layer offset vector coefficients, p refers to p row observations; bqOutputting a layer offset vector coefficient, wherein q refers to a q-row observation value; e (K) means the mean of the K-th variable, and E (K +1) means the mean of the K + 1-th variable.
2. Computing hidden layer input and output
The input vector of the hidden layer is calculated by performing weighted linear operation processing on the input layer vector and the corresponding input layer weight matrix:
Figure GDA0002529930150000101
wherein h isih(k) An input vector formula representing a hidden layer, h being 1 to p observations (h ═ 1,2 … p), i.e. different OBD devices. The role of the hidden layer threshold vector is: the offset is adjusted by linear superposition.
TABLE 3 input layer, hidden layer and output layer initialization
Figure GDA0002529930150000111
Inputting h of each hidden layerih(k) Processing with reserved mapping function to obtain output vector h of each hidden layeroh(k)：
Figure GDA0002529930150000112
Wherein h isoh(k) An output vector formula representing a hidden layer, h is 1 to p observations (h is 1,2 … p), i.e., different OBD devices. The hidden layer output vectors are expressions with respect to the input layer weight matrix, and then each hidden layer output vector hoh(k) As input to the output layer.
3. Computing output layer input and output
Performing linear processing through the hidden layer output vector and the corresponding output layer weight matrix, and calculating the input vector of the output layer as follows:
Figure GDA0002529930150000121
wherein, yio(k) An input vector formula representing the output layer, o refers to 1 to q input variables (o ═ 1,2 … q), i.e., the influencing factors of the OBD mesh cells, whoOutputting a vector h for each hidden layeroh(k) The output layer weight of (1).
Input y of each output layerio(k) Continuing to perform the retention mapping function processing to obtain the output vector y of each output layeroo(k)：
Figure GDA0002529930150000122
Wherein, yoo(k) An output vector formula representing an output layer, o represents 1 to q input variables (o is 1,2 … q), i.e., influencing factors of OBD mesh cells, whoOutputting a vector h for each hidden layeroh(k) Output layer weight of yio(k) The output layer input vectors.
And obtaining each classified output result of the output layer. And when the difference between the expected output and the actual output exceeds the preset error precision e defined by the error function, performing an error reverse training process.
Figure GDA0002529930150000123
Wherein d iso(k) Representing the desired output vector.
Based on the neural network model framework, the parameters of the neural network model are obtained by training the existing OBD stationing data, the rules of the existing OBD stationing are established, and the disadvantage of stationing under the rules is predicted. The historical OBD stationing sample data is processed through a neural network model, so that the current OBD stationing training parameters can be obtained, the accuracy and recall rate of OBD stationing prediction are enhanced, and the accuracy and effectiveness of OBD stationing are improved.
In one embodiment, as shown in fig. 5, a schematic structural diagram of a neural network processing step in one embodiment is shown. The input layer of the neural network model has 784 nodes, the hidden layer has 30 nodes, and the output layer has 10 nodes. Generating an input layer vector (corresponding to 784 nodes) by acquiring an OBD (on-board diagnostics) distribution influence factor set (as shown in Table 1); 30 node data can be obtained through the processing of the input layer, and the 30 node data are transmitted to the hidden layer; through hidden layer processing, 10 node data can be obtained, the 10 node data are transmitted to an output layer, and current OBD stationing training parameters can be obtained through output layer processing.
In one embodiment, the Similarity cluster is an SNN (Similarity-Nearest Neighbor clustering algorithm) cluster; the Nearest Neighbor classification is the KNN (K-Nearest Neighbor: K Nearest Neighbor) classification.
Step S330 includes: and carrying out SNN clustering processing on the service sample data set to obtain OBD stationing clustering result data.
In particular, SNN clustering refers to a shared nearest neighbor based clustering algorithm that processes variable density clusters by using the number of shared nearest neighbors between data points as a similarity. Therefore, spatial clusters with different sizes, shapes and densities can be found in the service sample data set. And through SNN clustering, the cluster of OBD point distribution of the area to be planned is subdivided, and the shapes and densities of various service data points are identified, so that the influence of the geographic environment is fully utilized.
Step S340 includes: and carrying out KNN classification processing on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data.
Specifically, the KNN classification refers to finding K nearest neighbors of a test sample in a training sample, and then determining the class of the test sample according to the classes of the K nearest neighbors. And performing KNN classification processing on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data. By processing KNN classification on the basis of SNN clustering, the check degree of OBD stationing can be effectively improved, the problem of coverage of optical resources in a service sparse area is greatly considered, and the destruction of an OBD stationing dead-angle area is realized.
In an embodiment, as shown in fig. 6, a schematic structural diagram of the KNN classification processing step in an embodiment is shown. The specific process of KNN classification treatment is as follows:
obtaining a test sample set to be classified and a classified training sample set Ci(where i ═ 1,2,3 … … n) and nearest neighbor K values; performing K-neighbor classification processing on the test sample set and the training sample set P, judging that Nk (P) ═ K (the training sample set 2, P and K are included), and if K neighbor point sets Nk (P) of the training sample set P all belong to the same class CiThe training sample set P is divided into the classes; if K nearest neighbors in the training sample set P are centered, finding the nearest shared neighbor of the training sample set P to obtain the point number Ki of each category in the K nearest neighbor centering. Comparing Nk (P) with the training sample set Q, judging Nk (Q) ═ k (wherein the training sample set 2, Q, k), and if the point distribution of P belongs to Nk (Q), obtaining the similarity between the training sample set P and the training sample set Q according to the size of the shared points. Calculating the average similarity N (C) of a certain distribution pointi) (ii) a P and CiIs equal to the sum nk (p) of the similarities divided by k; maximum value (q and C)iSimilarity of (d); and the training sample set P is divided into the classes, so that abnormal point distribution is optimized.
In one embodiment, step S330 is followed by:
comparing the current OBD stationing training parameter with the OBD stationing clustering result data; obtaining OBD abnormal point layout data;
and carrying out nearest neighbor classification processing on the OBD abnormal point distribution data to obtain optimized OBD point distribution classification result data.
Specifically, current OBD point distribution training parameters obtained through processing of the neural network model are compared with OBD point distribution clustering result data obtained through similarity clustering processing, and OBD abnormal point distribution data are obtained. And performing nearest neighbor classification processing on the OBD abnormal point distribution data to obtain optimized OBD distribution classification result data, thereby optimizing the OBD abnormal point distribution data and improving the OBD distribution accuracy.
In one embodiment, the current OBD stationing training parameter obtained by processing the neural network model may be subjected to similarity clustering again to obtain re-division of the current OBD stationing training parameter, a new type of grid cell stationing is established for some edge data of the neural network model which is not subjected to effective OBD stationing division, and OBD abnormal stationing data is obtained by comparing with the neural network output result. And performing nearest neighbor classification on the OBD abnormal distribution point data, namely calculating the average similarity of each OBD abnormal distribution point and different classes of K neighbors around the OBD abnormal distribution point, and finally distributing the OBD abnormal distribution point to the class of grid unit with the maximum similarity, thereby optimizing the OBD distribution point result under the clustering algorithm.
In one embodiment, as shown in FIG. 7, a flow diagram of the iterative optimization step in one embodiment is shown. Step S350 is followed by:
and step S710, calibrating the current OBD stationing training parameters according to the comparison result to obtain calibrated current OBD stationing classification result data.
And S720, comparing the calibrated current OBD stationing classification result data with the optimized OBD stationing classification result data, and updating the current OBD stationing optimization data.
Specifically, according to the optimization processing of the OBD stationing, after obtaining current OBD stationing optimization data, the current OBD stationing training parameters are continuously calibrated, and calibrated current OBD stationing classification result data is obtained. And comparing the calibrated current OBD stationing classification result data with the optimized OBD stationing classification result data again, and updating the current OBD stationing optimization data, so that more optimized OBD stationing data can be obtained, and the accuracy and efficiency of OBD stationing are improved.
In one embodiment, step S310 is followed by:
carrying out data preprocessing on the OBD distribution influence factor set to obtain a historical OBD distribution sample data set and a service sample data set; the data preprocessing comprises data cleaning, data encoding and data reconstruction.
Specifically, data preprocessing, such as data cleaning, data encoding and data reconstruction, is performed on the OBD distribution influence factor set, so that data of the OBD distribution influence factor set can be screened, and useless data can be deleted; and classifying the OBD distribution influence factor set to obtain a historical OBD distribution sample data set and a service sample data set. And further improve the OBD distribution optimization processing speed.
In one embodiment, the historical OBD stationing sample data set includes a copper coverage contributor data set and a light coverage contributor data set; the service sample data set comprises an influencing factor data set of the region grid segmentation unit.
Specifically, the copper coverage influencing factor dataset, the light coverage influencing factor dataset, and the area mesh dividing unit influencing factor dataset may be as shown in table 1 above.
In one embodiment, as shown in fig. 8, a schematic flow chart of an ODN network intelligent planning method in another embodiment is shown. The ODN intelligent planning method can comprise the following steps:
step S810, obtaining an OBD distribution influence factor set of an area to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set.
And S820, processing the historical OBD stationing sample data set through the neural network model to obtain the current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter.
And step S830, performing SNN clustering processing on the service sample data to obtain OBD stationing clustering result data.
And step 840, carrying out KNN classification processing on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data.
And step S850, generating current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
In the above embodiment, a model for prediction and verification is established according to the existing data set to solve the service foundation of the intelligent stationing OBD grid unit. The method comprises the steps of establishing an OBD intelligent stationing neural network model and establishing an OBD verification model based on combination of SNN clustering and KNN. The former establishes an OBD (on-board diagnostics) stationing prediction model through a training data set; the result of the OBD stationing prediction model is optimized and checked by the aid of the OBD stationing prediction model, parameters of the neural network model are corrected in a returning mode, and the neural network model is trained repeatedly, so that OBD stationing is optimized, and accuracy and efficiency of OBD stationing are improved.
In an embodiment, as shown in fig. 9, a flowchart of an ODN network intelligent planning method in an embodiment is shown. The existing OBD point distribution sample data (OBD historical point distribution sample data) is transmitted to the neural network model, and training parameters of the OBD point distribution are obtained through neural network processing. And carrying out establishment verification processing based on the spatial density OBD distribution position on sample data (service distribution sample data) without the OBD distribution through a SNN similarity clustering algorithm, wherein the sample data without the OBD distribution such as cell scale (number of households), floor space sense (number of floors), existing telecommunication service volume (number of quantity) and the like. And performing optimization classification on the edge data points (service sparse zones) again through a KNN classification algorithm, and comparing the result of the KNN classification algorithm with the result of the neural network, so as to calibrate the parameters of the neural network model, train the neural network model repeatedly, and optimize OBD distribution. The method has the advantages of scientifically clustering the geographic division of the existing service resources and the number of the users, improving the regularity and the integrity of the distribution, and simultaneously improving the accuracy and the efficiency of the OBD distribution position.
In one embodiment, as shown in fig. 10, a flowchart of an ODN network intelligent planning method in one embodiment is shown. The method comprises the steps of obtaining an OBD distribution influence factor set through the requirement on OBD position intelligent distribution, and calculating factors having the most influence on OBD distribution through preprocessing before modeling such as data cleaning, data reconstruction, characteristic engineering construction, correlation coefficient matrix analysis and the like, and modeling tests such as neural network establishment and KNN model establishment. And (4) clustering, classifying and the like are carried out on the influence factor data according to requirements through neural network model algorithm processing, so as to obtain the specific OBD point distribution position of the output layer.
It should be understood that although the various steps in the flowcharts of fig. 3, 4, 7 and 8 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 3, 4, 7, and 8 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 11, there is provided an ODN network intelligent planning apparatus, including: a data acquisition unit 110, a neural network processing unit 120, a clustering processing unit 130, a classification processing unit 140, and an alignment optimization unit 150. Wherein:
a data obtaining unit 110, configured to obtain an OBD distribution influence factor set of an area to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set.
And the neural network processing unit 120 is configured to process the historical OBD stationing sample data set through the neural network model to obtain the current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter.
And the clustering unit 130 is configured to perform similarity clustering on the service sample data to obtain OBD stationing clustering result data.
And the classification processing unit 140 is configured to perform nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data.
And the comparison optimization unit 150 is configured to generate current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
In one embodiment, the neural network processing unit 120 includes:
the input layer data acquisition unit is used for acquiring an input layer vector and an input layer weight matrix of a historical OBD stationing sample data set;
the hidden layer data acquisition unit is used for carrying out linear processing on the input layer vector and the input layer weight matrix to obtain a hidden layer input vector; performing reserved mapping function processing on the hidden layer input vector to obtain a hidden layer output vector;
the output layer data acquisition unit is used for carrying out linear processing on the hidden layer output vector and a hidden layer weight matrix corresponding to the hidden layer output vector to obtain an output layer input vector; and carrying out reserved mapping function processing on the input vector of the output layer to obtain the current OBD stationing training parameter.
In one embodiment, the ODN network intelligent planning apparatus further includes:
the OBD abnormal point acquisition unit is used for comparing the current OBD distribution training parameter with the OBD distribution clustering result data; obtaining OBD abnormal point layout data;
and the OBD abnormal point processing unit is used for carrying out nearest neighbor classification processing on the OBD abnormal point distribution data to obtain optimized OBD distribution classification result data.
In one embodiment, the ODN network intelligent planning apparatus further includes:
the parameter calibration unit is used for calibrating the current OBD stationing training parameters according to the comparison result to obtain calibrated current OBD stationing classification result data;
and the data updating unit is used for comparing the calibrated current OBD stationing classification result data with the optimized OBD stationing classification result data and updating the current OBD stationing optimization data.
In one embodiment, the ODN network intelligent planning apparatus further includes:
the data preprocessing unit is used for preprocessing data of the OBD distribution influence factor set to obtain a historical OBD distribution sample data set and a service sample data set; the data preprocessing comprises data cleaning, data encoding and data reconstruction.
For specific limitations of the ODN network intelligent planning apparatus, reference may be made to the above limitations of the ODN network intelligent planning method, which is not described herein again. All or part of the modules in the ODN intelligent planning device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, an ODN network intelligent planning system is provided, and the ODN network intelligent planning system may be a server, and its internal structure diagram may be as shown in fig. 12. The ODN intelligent planning system comprises a processor, a memory, a network interface and a database which are connected through a system bus. Wherein, the processor of the ODN network intelligent planning system is used for providing calculation and control capability. The memory of the ODN network intelligent planning system comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the ODN intelligent planning system is used for storing data such as an OBD distribution influence factor set, current OBD distribution training parameters, current OBD distribution classification result data, OBD distribution clustering result data, optimized OBD distribution classification result data and current OBD distribution optimization data. The network interface of the ODN network intelligent planning system is used for being connected and communicated with an external terminal through a network. The computer program is executed by a processor to implement an ODN intelligent planning method.
Those skilled in the art will appreciate that the structure shown in fig. 12 is only a block diagram of a part of the structure related to the present application, and does not constitute a limitation to the ODN network intelligent planning system to which the present application is applied, and a specific ODN network intelligent planning system may include more or less components than those shown in the figure, or combine some components, or have a different arrangement of components.
In one embodiment, an ODN network intelligent planning system is provided, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the following steps:
acquiring an OBD distribution influence factor set of a region to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set;
processing a historical OBD stationing sample data set through a neural network model to obtain a current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter;
performing similarity clustering processing on the service sample data set to obtain OBD stationing clustering result data;
carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data;
and comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data, and generating current OBD stationing optimization data according to the comparison result.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring an OBD distribution influence factor set of a region to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set;
processing a historical OBD stationing sample data set through a neural network model to obtain a current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter;
performing similarity clustering processing on the service sample data set to obtain OBD stationing clustering result data;
carrying out nearest neighbor classification on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data;
and generating current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. An ODN network intelligent planning method is characterized by comprising the following steps:
acquiring an OBD distribution influence factor set of a region to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set;
processing the historical OBD stationing sample data set through a neural network model to obtain a current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter; wherein the neural network model is a BP neural network model;
performing similarity clustering processing on the service sample data set to obtain OBD stationing clustering result data;
carrying out nearest neighbor classification on the OBD stationing and clustering result data to obtain optimized OBD stationing and classifying result data;
and generating current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
2. The ODN intelligent planning method according to claim 1, wherein the step of processing the historical OBD stationing sample data set through a neural network model to obtain current OBD stationing training parameters comprises:
acquiring an input layer vector and an input layer weight matrix of the historical OBD stationing sample data set;
performing linear processing on the input layer vector and the input layer weight matrix to obtain a hidden layer input vector; performing reserved mapping function processing on the hidden layer input vector to obtain a hidden layer output vector;
performing linear processing on the hidden layer output vector and a hidden layer weight matrix corresponding to the hidden layer output vector to obtain an output layer input vector; and performing reserved mapping function processing on the input vector of the output layer to obtain the current OBD stationing training parameter.
3. The ODN intelligent planning method according to claim 1 wherein the similarity clusters are SNN clusters; the nearest neighbor classification is a KNN classification;
the step of obtaining OBD stationing clustering result data by clustering the service sample data set through similarity comprises the following steps:
performing SNN clustering processing on the service sample data set to obtain OBD stationing clustering result data;
the step of obtaining the optimized OBD stationing and classifying result data by carrying out nearest-neighbor classification on the OBD stationing and clustering result data comprises the following steps of:
and carrying out KNN classification processing on the OBD stationing clustering result data to obtain the optimized OBD stationing classification result data.
4. The ODN intelligent planning method according to claim 1, wherein the step of obtaining OBD stationing clustering result data by performing similarity clustering processing on the service sample data set comprises the following steps:
comparing the current OBD stationing classification result data with the OBD stationing clustering result data; obtaining OBD abnormal point layout data;
and carrying out nearest neighbor classification processing on the OBD abnormal point distribution data to obtain the optimized OBD point distribution classification result data.
5. The ODN intelligent planning method according to claim 1, wherein the step of obtaining a comparison result according to the comparison between the current OBD stationing classification result data and the optimized OBD stationing classification result data comprises:
calibrating the current OBD stationing training parameters according to the comparison result to obtain calibrated current OBD stationing classification result data;
and comparing the calibrated current OBD stationing classification result data with the optimized OBD stationing classification result data, and updating the current OBD stationing optimization data.
6. The ODN network intelligent planning method according to any one of claims 1 to 5, wherein the step of obtaining the OBD distribution point influence factor set of the area to be planned comprises the following steps:
carrying out data preprocessing on the OBD distribution influence factor set to obtain a historical OBD distribution sample data set and the service sample data set;
the data preprocessing comprises data cleaning, data encoding and data reconstruction.
7. The ODN network intelligent planning method of claim 6,
the historical OBD stationing sample data set comprises an influence factor data set of copper coverage and an influence factor data set of light coverage;
the service sample data set comprises an influencing factor data set of a region grid segmentation unit.
8. An ODN network intelligent planning device is characterized by comprising:
the data acquisition unit is used for acquiring an OBD distribution influence factor set of an area to be planned; the OBD distribution influence factor set comprises a historical OBD distribution sample data set and a service sample data set; the neural network processing unit is used for processing the historical OBD stationing sample data set through a neural network model to obtain a current OBD stationing training parameter and current OBD stationing classification result data corresponding to the current OBD stationing training parameter; wherein the neural network model is a BP neural network model;
the clustering processing unit is used for clustering the service sample data set through similarity to obtain OBD stationing clustering result data;
the classification processing unit is used for carrying out nearest classification processing on the OBD stationing clustering result data to obtain optimized OBD stationing classification result data;
and the comparison optimization unit is used for generating current OBD stationing optimization data according to a comparison result obtained by comparing the current OBD stationing classification result data with the optimized OBD stationing classification result data.
9. An ODN network intelligent planning system, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the ODN network intelligent planning method according to any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the ODN network intelligent planning method according to any one of claims 1 to 7.
CN201810108108.0A 2018-02-02 2018-02-02 ODN network intelligent planning method, device and system Active CN108446712B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810108108.0A CN108446712B (en) 2018-02-02 2018-02-02 ODN network intelligent planning method, device and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810108108.0A CN108446712B (en) 2018-02-02 2018-02-02 ODN network intelligent planning method, device and system

Publications (2)

Publication Number Publication Date
CN108446712A CN108446712A (en) 2018-08-24
CN108446712B true CN108446712B (en) 2020-12-11

Family

ID=63191551

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810108108.0A Active CN108446712B (en) 2018-02-02 2018-02-02 ODN network intelligent planning method, device and system

Country Status (1)

Country Link
CN (1) CN108446712B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110008977B (en) * 2018-12-05 2023-08-11 创新先进技术有限公司 Clustering model construction method and device
CN111328084B (en) * 2018-12-14 2022-12-23 ***通信集团广东有限公司 Method and device for evaluating cell capacity
CN115243270B (en) * 2021-04-07 2023-09-22 ***通信集团设计院有限公司 5G network planning method, device, computing equipment and storage medium
CN113316038B (en) * 2021-05-31 2022-03-25 北京邮电大学 Optical access network ODN intelligent clustering planning method facing sparse user distribution

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103024762A (en) * 2012-12-26 2013-04-03 北京邮电大学 Service feature based communication service forecasting method
CN106022614A (en) * 2016-05-22 2016-10-12 广州供电局有限公司 Data mining method of neural network based on nearest neighbor clustering
US9577782B2 (en) * 2014-06-13 2017-02-21 Ciena Corporation Systems and methods for statistical multiplexing with OTN and DWDM

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103024762A (en) * 2012-12-26 2013-04-03 北京邮电大学 Service feature based communication service forecasting method
US9577782B2 (en) * 2014-06-13 2017-02-21 Ciena Corporation Systems and methods for statistical multiplexing with OTN and DWDM
CN106022614A (en) * 2016-05-22 2016-10-12 广州供电局有限公司 Data mining method of neural network based on nearest neighbor clustering

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于无源光网络的联合资源优化设计;李明;《中国优秀硕士学位论文全文数据库信息科技辑》;20150415(第04期);第43-52页 *

Also Published As

Publication number Publication date
CN108446712A (en) 2018-08-24

Similar Documents

Publication Publication Date Title
CN108446712B (en) ODN network intelligent planning method, device and system
CN111294812B (en) Resource capacity-expansion planning method and system
CN109992848A (en) A kind of press crown Robust Optimal Design method based on minus ideal result proximity distance
Han et al. Spectral clustering and genetic algorithm for design of district metered areas in water distribution systems
CN111523777A (en) Novel smart city system and application method thereof
CN109242150A (en) A kind of electric network reliability prediction technique
CN116050540B (en) Self-adaptive federal edge learning method based on joint bi-dimensional user scheduling
CN115755954A (en) Routing inspection path planning method and system, computer equipment and storage medium
CN111598457B (en) Method and device for determining quality of power wireless network
Khuntia et al. Risk‐based security assessment of transmission line overloading considering spatio‐temporal dependence of load and wind power using vine copula
CN111079930A (en) Method and device for determining quality parameters of data set and electronic equipment
CN115174416A (en) Network planning system, method and device and electronic equipment
CN112990776B (en) Distribution network equipment health degree evaluation method
CN112787833B (en) Method and device for deploying CDN (content delivery network) server
CN116541166A (en) Super-computing power scheduling server and resource management method
CN116523001A (en) Method, device and computer equipment for constructing weak line identification model of power grid
CN110705889A (en) Enterprise screening method, device, equipment and storage medium
CN113795032B (en) Method and device for judging invisible faults of indoor division, storage medium and equipment
CN110458432A (en) A kind of electric power Optical Transmission Network OTN reliability diagnostic method based on cloud model
CN115720212A (en) Network flow prediction and automatic optimization balancing method based on multi-source data fusion
CN114268625B (en) Feature selection method, device, equipment and storage medium
CN114423025A (en) Scene recognition method, device, equipment and storage medium
CN113541993A (en) Network evaluation method and device, network index processing method, equipment and medium
CN114757440B (en) Distributed wind power prediction method, model training method, device and medium
CN112147969A (en) Cold standby reliability redundancy allocation method and system based on group intelligent algorithm

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