5g network coverage evaluation method and apparatus, computer readable medium, and electronic device

Abstract

The application belongs to the technical field of network coverage planning, and particularly relates to a 5G network coverage evaluation method, device, readable medium and electronic equipment; the application obtains 5G network coverage data of a to-be-evaluated area by taking 4G network coverage data of the to-be-evaluated area and difference feature data of 4G network and 5G network as input through a 5G network coverage conversion model; the application adds the important feature of difference feature data when performing 5G network coverage prediction, reduces the error between the 5G network coverage prediction result and the actual result, and the obtained 5G network coverage result is more accurate; moreover, the evaluation steps of the application are simple, no additional steps are needed, and the efficiency is high; meanwhile, the application directly uses road test data and measurement report data when training the model, does not need to take the form of a simulation map, and is low in cost, which is conducive to accurate deployment of 5G network in the to-be-evaluated area.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of network coverage evaluation and planning, and particularly relates to a 5G network coverage evaluation method, a 5G network coverage evaluation device, a computer readable medium and an electronic device. BACKGROUND

[0002] In the current 5G network planning process, the commonly used method is to use 5G simulation software running on a server with high hardware configuration to perform 5G network coverage prediction and simulation, and output the planning results.

[0003] However, the above method for 5G network coverage evaluation has problems such as high cost, low efficiency, large error, etc., and thus cannot effectively guide the accurate deployment of 5G network.

[0004] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present application, and thus can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The purpose of the present application is to provide a 5G network coverage evaluation method, a 5G network coverage evaluation device, a computer readable medium and an electronic device, which at least partially overcome the technical problems of high cost, low efficiency, large error, etc. in the related art 5G network coverage evaluation.

[0006] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0007] According to one aspect of an embodiment of the present application, a 5G network coverage evaluation method is provided, comprising:

[0008] extracting 4G network coverage data, 5G network coverage data and difference feature data of road testing, the 4G network coverage data being used to evaluate 4G network coverage quality, the 5G network coverage data being used to evaluate 5G network coverage quality, and the difference feature data including a set of feature differences between 4G network and 5G network in network coverage;

[0009] combining the 4G network coverage data and the difference feature data as input samples, and using 5G network coverage data corresponding to the 4G network coverage data as output samples, training a pre-set machine learning model to obtain a 5G network coverage conversion model;

[0010] The 4G measurement report data is used to extract 4G network coverage data of the to-be-evaluated area, difference feature data of the to-be-evaluated area is extracted, the 4G network coverage data of the to-be-evaluated area and the difference feature data of the to-be-evaluated area are input into the 5G network coverage conversion model, and 5G network coverage data of the to-be-evaluated area is obtained.

[0011] According to an aspect of an embodiment of the present application, a 5G network coverage evaluation device is provided, comprising:

[0012] The extraction module is configured to extract 4G network coverage data, 5G network coverage data and difference feature data of road testing, the 4G network coverage data is used to evaluate 4G network coverage quality, the 5G network coverage data is used to evaluate 5G network coverage quality, and the difference feature data includes a set of feature differences between 4G network and 5G network in network coverage;

[0013] The model establishment module is configured to use the 4G network coverage data and the difference feature data combination as an input sample, use 5G network coverage data corresponding to the 4G network coverage data as an output sample, train a preset machine learning model, and obtain a 5G network coverage conversion model;

[0014] The evaluation module is configured to extract 4G network coverage data of a to-be-evaluated area through 4G measurement report data, extract difference feature data of the to-be-evaluated area, input the 4G network coverage data of the to-be-evaluated area and the difference feature data of the to-be-evaluated area into the 5G network coverage conversion model, and obtain 5G network coverage data of the to-be-evaluated area.

[0015] In some embodiments of the present application, based on the above technical solutions, the 4G measurement report data includes data positioning and 4G network coverage data content, and the evaluation module includes:

[0016] The data acquisition unit is configured to acquire data positioning of the 4G measurement report data, and the data positioning is used to represent position information of the to-be-evaluated area.

[0017] The 4G network coverage data content corresponding to the data positioning is acquired.

[0018] In some embodiments of the present application, based on the above technical solutions, the data acquisition unit includes a position acquisition unit, the position acquisition unit is configured to acquire a map and a fingerprint library corresponding to the map, and the fingerprint library is used to establish a mapping relationship between a received signal strength sample and map position information.

[0019] According to the 4G measurement report data, the received signal strength of the to-be-evaluated area is acquired.

[0020] The received signal strength of the to-be-evaluated area is matched with the received signal strength samples in the fingerprint library, and the map position information corresponding to the received signal strength that is successfully matched is taken as the position information of the to-be-evaluated area.

[0021] In some embodiments of the present application, based on the above technical solution, the position obtaining unit comprises a matching unit, which is configured to, when the received signal strength of the to-be-evaluated area is the same as the received signal strength sample in the fingerprint library or the similarity of the received signal strength of the to-be-evaluated area and the received signal strength in the fingerprint library is within a preset threshold range, take the position information corresponding to the received signal strength in the fingerprint library as the position information of the to-be-evaluated area.

[0022] In some embodiments of the present application, based on the above technical solution, the to-be-evaluated area comprises a primary cell area and a neighbor cell area, and the matching unit comprises a similarity calculation unit, which is configured to calculate the similarity of the received signal strength of the to-be-evaluated area and the received signal strength in the fingerprint library according to the following formula, and the specific formula is as follows:

[0023]

[0024] wherein, s i (i = 1, …, M) represents the intensity error of the i-th neighbor cell area received signal strength in the 4G measurement report data and the received signal strength in the fingerprint library, and the calculation formula of the intensity error is as follows:

[0025]

[0026] wherein, g(i) represents the weight of different neighbor cell areas relative to the primary cell area in the 4G measurement report data, P i represents the received signal strength of the i-th neighbor cell area, represents the received signal strength corresponding to the area in the fingerprint library that is the same as the i-th neighbor cell area in terms of the absolute radio frequency channel number and the physical layer cell identification respectively, and σ represents a preset value, when there is no area in the fingerprint library that is the same as the i-th neighbor cell area in terms of the absolute radio frequency channel number and the physical layer cell identification respectively, s i The value of s

[0027] In some embodiments of the present application, based on the above technical solution, the model establishing module comprises a data receiving unit, a data preprocessing unit and a model training unit,

[0028] The data receiving unit is configured to receive 4G reference signal receiving power, 4G transmission power, 4G antenna gain, 4G frequency band, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmission power, 5G antenna gain, 5G frequency band, and 5G mobile phone receiving antenna quantity of road testing;

[0029] The data preprocessing unit is configured to preprocess the 4G reference signal receiving power, 4G transmission power, 4G antenna gain, 4G frequency band, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmission power, 5G antenna gain, 5G frequency band, and 5G mobile phone receiving antenna quantity for inputting into a machine learning model for training.

[0030] The model training unit is configured to combine the preprocessed 4G reference signal receiving power and 5G transmission power difference, 4G antenna gain and 5G antenna gain difference, 4G frequency band and 5G frequency band difference, and 4G mobile phone receiving antenna quantity and 5G mobile phone receiving antenna quantity difference as input samples, and the preprocessed 5G reference signal receiving power as output samples, to train a preset machine learning model to obtain a 5G network coverage conversion model.

[0031] In some embodiments of the present application, based on the above technical solutions, the data preprocessing unit includes a normalization unit and an encoding unit,

[0032] The normalization unit is configured to normalize 4G reference signal receiving power, 4G transmission power, 4G antenna gain, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmission power, 5G antenna gain, and 5G mobile phone receiving antenna quantity.

[0033] The encoding unit is configured to use one-hot encoding for 4G frequency band and 5G frequency band.

[0034] According to an aspect of an embodiment of the present application, a computer readable medium having a computer program stored thereon is provided, the computer program being executed by a processor to implement the 5G network coverage evaluation method in the above technical solutions.

[0035] According to an aspect of an embodiment of the present application, an electronic device is provided, including a processor and a memory for storing executable instructions of the processor, wherein the processor is configured to execute the executable instructions to perform the 5G network coverage evaluation method in the above technical solutions.

[0036] According to an aspect of an embodiment of the present application, a computer program product or computer program is provided, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the 5G network coverage evaluation method as in the above technical solution.

[0037] In the technical solution provided in the embodiments of the present application, the 4G network coverage data of road testing and the difference feature data of the 4G network and the 5G network are combined as input samples, the 5G network coverage data of road testing is extracted as output samples, a preset machine learning model is trained to obtain a 5G network coverage conversion model, the 4G network coverage data of the to-be-evaluated area and the difference feature data combination of the to-be-evaluated area of the 4G network and the 5G network are input into the 5G network coverage conversion model to obtain the result of the 5G network coverage data of the to-be-evaluated area. The present application adds the important feature of difference feature data when predicting the 5G network coverage, reduces the error between the prediction result and the actual result of the 5G network coverage, and when the 5G network coverage conversion model is trained, the 4G network coverage data and the difference feature data are directly input to obtain the 5G network coverage data of the to-be-evaluated area, without the need for additional steps, which is efficient, and the model training directly uses the data and measurement report data of road testing, without the need for a simulation map, which is low in cost and conducive to the accurate deployment of the 5G network in the to-be-evaluated area.

[0038] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0039] The drawings herein are incorporated into the specification and form a part of the specification, show embodiments consistent with the present application, and together with the specification serve to explain the principles of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained from these drawings without creative labor for those skilled in the art.

[0040] Figure 1 An exemplary system architecture block diagram to which the technical solution of the present application is applied is schematically shown.

[0041] Figure 2 A flowchart of the 5G network coverage evaluation method of the present application is schematically shown.

[0042] Figure 3 A flowchart of the acquisition method of the difference feature data of the present application is schematically shown.

[0043] Figure 4 A 4G and 5G transmission power comparison table is schematically shown.

[0044] Figure 5 A flowchart of a method for establishing a 5G network coverage conversion model of the present application is schematically shown.

[0045] Figure 6 A flowchart of a method for obtaining data positioning of 4G measurement report data of the present application is schematically shown.

[0046] Figure 7 An effect diagram of the present application is schematically shown for combining 5G reference signal received power of each neighbor cell area and the primary cell area.

[0047] Figure 8 An enlarged view of the area grid to be evaluated in the present application is schematically shown. Figure 7

[0048] Figure 9 A structural block diagram of a 5G network coverage evaluation device of the present application is schematically shown.

[0049] Figure 10 A computer system structural block diagram of an electronic device for implementing the embodiments of the present application is schematically shown. DETAILED DESCRIPTION

[0050] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations can be implemented in any

[0051] Moreover, described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the

[0052] The block diagrams in the drawings show only the functional entities and not necessarily the physical separation of the functional entities. That is, the functional entities can be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0053] ​The flowchart shown in the drawing is only an exemplary illustration, not necessarily including all contents and operations / steps, and not necessarily executed in the order described. For example, some operations / steps can be further decomposed, and some operations / steps can be combined or partially combined, so that the actual execution order can be changed according to actual conditions.

[0054] With the development of 5G network, the planning of 5G network for each area is particularly important. The focus of the planning of 5G network is to evaluate the 5G network coverage of the area, and to make different planning and deployment according to the coverage.

[0055] The 5G network coverage can be obtained by using 5G simulation software running on a server with high hardware configuration to perform coverage prediction and simulation, and output the planning result. The specific steps of the method are as follows:

[0056] First step: map analysis, import the three-dimensional map information of the area where the network is located in the simulation tool. Generally, the imported information includes: terrain elevation map, feature classification map and vector map;

[0057] Second step: import of basic data, import related basic data in the simulation tool, mainly including engineering parameter data, antenna data, test data, etc.

[0058] Third step: selection and correction of propagation model: according to the frequency of the network and the network propagation topography, select the appropriate propagation model, such as standard propagation model, ray tracing model. Model correction needs to conduct typical test according to specific wireless environment, obtain some actual propagation loss data, and then correct the original propagation prediction model based on these data;

[0059] Fourth step: wireless network simulation, use the simulation tool to calculate the path loss of the simulation area, and then obtain the coverage simulation, interference simulation and other results of the area according to the transmission power, antenna gain, etc.

[0060] Fifth step: simulation result output, statistics of each coverage index of the simulation area and output of related rendering layers.

[0061] There are the following problems in obtaining the 5G network coverage situation by using the above method: first, a dedicated simulation map is needed, which makes the cost higher, and many rural areas lack simulation maps and cannot be planned at all; second, the imported static three-dimensional map has errors, and the domestic infrastructure speed is very fast now, and the landscape of some places may change greatly in a few months, such as building a building or a flyover, or demolishing a shantytown. Therefore, using an old map to calculate will bring an error of at least 3-6dB in some places, resulting in a large error in the obtained 5G network coverage result; third, the propagation model has errors, and the standard deviation of the propagation model in urban environment is 8-10dB, and even if the road test is used to correct the propagation model, there will still be an error of about 6dB, which will also lead to a large error in the 5G network coverage result; fourth, the whole process is very tedious and inefficient. Therefore, using the above method to obtain the 5G network coverage situation has the problems of large error, high cost and low efficiency, which cannot effectively guide the precise deployment of 5G network and is not conducive to the deployment and development of 5G network.

[0062] To solve the above technical problems, the present application provides a 5G network coverage evaluation method, a 5G network coverage evaluation device, a computer readable medium and an electronic device. The specific content of each aspect will be disclosed in detail below.

[0063] Figure 1 An exemplary system architecture block diagram to which the technical solutions of the present application are applied is schematically shown.

[0064] As shown in Figure 1 The system architecture 100 can include a terminal device 110, a network 120 and a server 130. The terminal device 110 can include various electronic devices such as smartphones, tablets, notebooks, desktop computers, etc. The server 130 can be a standalone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The network 120 can be various connection types of communication media that can provide a communication link between the terminal device 110 and the server 130, such as a wired communication link or a wireless communication link.

[0065] According to the implementation needs, the system architecture in the embodiments of the present application can have any number of terminal devices, networks and servers. For example, the server 130 can be a server group composed of multiple server devices. In addition, the technical solutions provided by the embodiments of the present application can be applied to the terminal device 110, or can be applied to the server 130, or can be jointly implemented by the terminal device 110 and the server 130, and the present application does not make special limitations on this.

[0066] The above part introduces the composition of the system architecture of the present application, and the specific implementation method of the present application will be introduced next.

[0067] According to one aspect of the embodiments of the present application, the present application discloses a 5G network coverage evaluation method, as shown in Figure 2 Figure 2 The flow chart of the 5G network coverage evaluation method of the present application is schematically shown, including steps S210-S230, as follows:

[0068] In step S210: Extract the 4G network coverage data, 5G network coverage data and difference feature data of road testing, the 4G network coverage data is used to evaluate the 4G network coverage quality, the 5G network coverage data is used to evaluate the 5G network coverage quality, and the difference feature data includes a set of feature differences of 4G network and 5G network in network coverage.

[0069] Among them, the 4G network coverage data and the 5G network coverage data are data used to evaluate the network coverage quality of the 4G network and the 5G network respectively. In actual use, the reference signal received power (RSRP) can be used as the evaluation data of the network coverage quality. The reference signal received power (RSRP) can be obtained by road testing or measurement report. The reference signal received power is one of the key parameters that can represent the wireless signal strength in the LTE (Long Term Evolution) network and the physical layer measurement requirement, which is the average value of the signal power received on all REs (resource particles) carrying the reference signal within a certain symbol. According to the size of the reference signal received power, the corresponding network coverage quality can be obtained. When the value of the reference signal received power is greater than -85dBm, it is considered that the network coverage quality of the location is good, and the data service of medium or above rate can be obtained. Indoor can initiate various services, and can obtain data service of low rate or above. When the value of the reference signal received power is less than -85dBm, it indicates that the network coverage quality is poor, and the drop rate is high. If the value of the reference signal received power is less than -110m, it indicates that the area is equivalent to no network coverage, and the service is basically unable to call.

[0070] The 4G network coverage data and the 5G network coverage data of the road testing of the present application can come from the data obtained by pre-testing, such as the 4G reference signal received power and the 5G reference signal received power of a specific area obtained by the operator through road testing. It can also be the 4G network coverage data and the 5G network coverage data obtained by actual road testing. The range of road testing should be determined as a region as large as possible, which can be taken as a reference with an area of fifteen kilometers. In this way, the 4G network coverage data and the 5G network coverage data obtained by road testing have the value of establishing the conversion model, which can avoid the large conversion deviation of the final 5G conversion model caused by too small area.

[0071] ​The difference feature data in this application refers to the set of feature differences between 4G and 5G networks in terms of network coverage. From the perspective of network coverage, co-located 4G and 5G base stations have almost the same wireless environment, and the differences in coverage between the two mainly come from factors such as frequency band, power, terminal, and beam gain. Therefore, based on the premise that 4G and 5G base stations are basically co-located in the early stage of network construction, this application focuses on the differences in network coverage between 4G and 5G networks mainly in terms of frequency band, power, terminal, and beam gain. Other difference feature data are not considered in this application when 4G and 5G base stations are co-located.

[0072] The following section analyzes the differences in network coverage between 4G and 5G networks and discloses the process of obtaining the difference feature data. In one embodiment of this application, such as... Figure 3 As shown, Figure 3 A flowchart illustrating the method for obtaining differential feature data according to this application is shown. The method for obtaining differential feature data includes steps S310-S340.

[0073] Step S310: Obtain the 4G frequency band and the 5G frequency band, calculate the difference range between the 4G frequency band and the 5G frequency band, and use the difference range between the 4G frequency band and the 5G frequency band as the first difference feature data.

[0074] A frequency band is the range of radio waves used in wireless communication. Frequency bands are generally determined by the configuration of base stations. If 4G and 5G operate on the same frequency band, their propagation path losses are essentially the same, so differences between frequency bands do not need to be considered. However, if 4G and 5G operate on different frequency bands, the fundamental differences between them must be taken into account. Currently, each operator has its own frequency band standards and names. Although the names of the frequency bands differ among operators, their corresponding frequency ranges often overlap. For example, China Telecom uses a 4G frequency band named 1800, which corresponds to a frequency range of 1860-1880MHz; the 5G frequency band used by China Telecom has a frequency range of 3300-3400MHz. Therefore, there is a difference between these two frequency bands, and the corresponding difference characteristic data is the frequency band difference range. There are several ways to calculate the difference range. One way is to subtract the minimum value from the maximum value in the corresponding frequency band range of the 5G frequency band to obtain the frequency band difference range. For example, the frequency band difference range between China Telecom's 4G and 5G frequency bands is 3400-1860=1540MHz. Therefore, the frequency band difference range can be obtained through the above method, and the frequency band difference range can be used as the first difference characteristic data.

[0075] Step S320: Obtain the 4G transmission power and 5G transmission power, calculate the difference between the 4G transmission power and 5G transmission power, and use the difference between the 4G transmission power and 5G transmission power as the second difference feature data.

[0076] The transmission power of the base station in the network is averaged to each subcarrier, which corresponds to the equal division of the transmission power of the base station for each subcarrier, so the transmission power of each subcarrier is affected by the configured system bandwidth, the larger the bandwidth, the smaller the power of each subcarrier, and the transmission power is obtained through the configuration of the base station. As shown in Figure 4 Figure 4 A 4G and 5G transmission power comparison table is schematically shown. For the 4G network, the LTE (Long Term Evolution) system is used, the bandwidth is 20MHz, the corresponding subcarrier spacing is 15KHz, the system bandwidth is 20MHz, the effective subcarrier number is 1200, the antenna power is 40w, and the single subcarrier power is 15.2dBm. The above 4G network transmission power data is only for the LTE (Long Term Evolution) system with a bandwidth of 20MHz. For other 4G networks, the corresponding transmission power can be obtained by checking the network form.

[0077] The 5G network uses the NR (New Radio) system, which is divided into two types, NR1 and NR2. The transmission power data of NR1 is that the subcarrier spacing is 30KHz, the system bandwidth is 100MHz, the subcarrier spacing configuration is 1, the effective subcarrier number is 3264, the antenna power is 200w, and the single subcarrier power is 17.9dBm. The transmission power data of NR2 is that the subcarrier spacing is 60KHz, the system bandwidth is 100MHz, the subcarrier spacing configuration is 2, the effective subcarrier number is 1620, the antenna power is 200w, and the single subcarrier power is 20.9dBm. Therefore, from the above data, it can be concluded that for the 5G network of NR1 system, the difference between 4G transmission power and 5G transmission power is 2.7dBm, and for the 5G network of NR2 system, the difference between 4G transmission power and 5G transmission power is 5.7dBm. Therefore, the difference between the calculated 4G transmission power and 5G transmission power can be used as the second difference feature data. Therefore, the difference between the actual road test data corresponding to the 4G transmission power and the 5G transmission power can be obtained by referring to the above method, and the difference between the 4G transmission power and the 5G transmission power is obtained after obtaining the single subcarrier power of the 4G network and the 5G network.

[0078] Step S330: Obtain the number of 4G receiving antennas and the number of 5G receiving antennas, calculate the antenna difference value of the number of 4G receiving antennas and the number of 5G receiving antennas, calculate the receiving level difference value according to the antenna difference value, and take the receiving level difference value as the third difference feature data.

[0079] ​4G network and 5G network receive antenna number is generally different, the number of receiving antennas is a preset fixed value, for LTE terminal, the number of receiving antennas is two, and for 5GNR terminal, the number of receiving antennas is four, and the corresponding antenna difference is 2, and the function of receiving antenna is to receive diversity, so as to increase the level and quality of uplink received signal, therefore, considering the receiving gain of multiple antennas, the receiving level difference is calculated according to the antenna difference, and it is concluded that the receiving level of 5GNR terminal is 3dB higher than that of 4GLTE terminal, that is, the receiving level difference is 3dB.

[0080] Step S340: obtaining 4G antenna gain and 5G antenna gain, calculating the difference between 4G antenna gain and 5G antenna gain, and taking the difference between 4G antenna gain and 5G antenna gain as the fourth difference feature data.

[0081] In order to determine the difference between 4G antenna gain and 5G antenna gain, it is necessary to determine the difference between 4G wave speed and 5G wave speed, which refers to the distance propagated by a certain vibration state per unit time. Due to the introduction of massive MIMO and beam management technology in 5G network, the network coverage provides stronger coverage ability and flexibility in horizontal and vertical dimensions compared with 4G. Among them, 5G uses narrow wave speed, and 4G uses wide wave speed. The coverage ability of multiple narrow beams of 5G is stronger than that of single wide beam of 4G.

[0082] The calculation method of antenna gain is determined by the horizontal and vertical patterns of actual antenna. First, the horizontal angle and vertical angle of any point in space in the horizontal and vertical pattern coordinates of the antenna are calculated by coordinate transformation. The horizontal gain and vertical gain are read according to the antenna pattern, and then the gain of any point in space can be calculated by three-dimensional linear interpolation method, and the 4G antenna gain is obtained. In actual use, the 4G antenna gain can be directly obtained by obtaining the 4G antenna gain calculated.

[0083] Due to the difference between 5G and 4G, the 5G antenna gain is calculated first. The gain Gn of each narrow beam n is calculated, and the calculation method is the same as that of 4G antenna gain. Then, according to the calculated gain, the beam with the maximum gain is selected as the 5G coverage wave speed of the point, and the maximum Gn is the 5G antenna gain of the point. In actual use, the 5G antenna gain can be directly obtained by obtaining the 5G antenna gain calculated, and the difference between 4G antenna gain and 5G antenna gain is taken as the fourth difference feature data.

[0084] After extracting the 4G network coverage data, 5G network coverage data and difference feature data of road test by step S210, the data can be trained, and the specific training method is as shown in step S220.

[0085] In step S220: the 4G network coverage data and the difference feature data combination is taken as the input sample, the 5G network coverage data corresponding to the 4G network coverage data is taken as the output sample, and the preset machine learning model is trained to obtain the 5G network coverage conversion model.

[0086] The corresponding data in step S210 is substituted into the 4G network coverage data, the difference feature data and the 5G network coverage data, and the specific 5G network coverage conversion model establishment method is as shown in Figure 5 Figure 5 The flow chart of the 5G network coverage conversion model establishment method of the present application is schematically shown. It includes steps S510-S530.

[0087] Step S510: receiving the 4G reference signal receiving power, 4G transmitting power, 4G antenna gain, 4G frequency band and 4G mobile phone receiving antenna number, 5G reference signal receiving power, 5G transmitting power, 5G antenna gain, 5G frequency band and 5G mobile phone receiving antenna number of road test.

[0088] Among them, the 4G reference signal receiving power and the 5G reference signal receiving power correspond to the 4G network coverage data and the 5G network coverage data respectively. As for the difference feature data, in combination with the above steps S310-S340, the difference range of 4G frequency band and 5G frequency band is taken as the first difference feature data; the difference value of 4G transmitting power and 5G transmitting power is taken as the second difference feature data; the antenna difference value of 4G receiving antenna number and 5G receiving antenna number is calculated, the receiving level difference value is calculated according to the antenna difference value, and the receiving level difference value is taken as the third difference feature data; the difference value of 4G antenna gain and 5G antenna gain is taken as the fourth difference feature data, and the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data are combined to form the difference feature data.

[0089] Step S520: 4G reference signal receiving power, 4G transmitting power, 4G antenna gain, 4G frequency band and 4G mobile phone receiving antenna number, 5G reference signal receiving power, 5G transmitting power, 5G antenna gain, 5G frequency band and 5G mobile phone receiving antenna number are preprocessed for inputting the machine learning model for training.

[0090] ​In an embodiment of the present application, before the above data is preprocessed, the first difference feature data can be obtained according to the 4G frequency band and the 5G frequency band; the second difference feature data can be obtained according to the 4G transmit power and the 5G transmit power; the third difference feature data can be obtained according to the antenna difference value of the 4G receiving antenna quantity and the 5G receiving antenna quantity; and the fourth difference feature data can be obtained according to the difference value of the 4G antenna gain and the 5G antenna gain. Therefore, the data actually preprocessed can be: the 4G reference signal received power, the 5G reference signal received power, the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data.

[0091] Before the above data is preprocessed, the data is generally cleaned.

[0092] In an embodiment of the present application, before the 4G reference signal received power, the 4G transmit power, the 4G antenna gain, the 4G frequency band and the 4G mobile phone receiving antenna quantity, the 5G reference signal received power, the 5G transmit power, the 5G antenna gain, the 5G frequency band and the 5G mobile phone receiving antenna quantity are preprocessed, the above data is also cleaned, and the specific data cleaning method can include:

[0093] Missing value processing: in the process of obtaining the above data through road test data, due to environmental factors, many areas may be missing specific data. Therefore, the idea of missing value processing is to fill in the missing values with the most possible values. The present application can use regression analysis, Bayesian calculation formula or decision tree to infer the maximum possible value of the specific attribute of the record, so as to fill in the corresponding value. For example, through road test in a certain area, due to specific reasons, the value of 4G reference signal received power cannot be obtained. At this time, the maximum possible value of the specific attribute of the record can be inferred by regression analysis, Bayesian calculation formula or decision tree, so as to directly fill in the missing value and avoid the error of the subsequent 5G network coverage conversion model.

[0094] Abnormal value processing: in the process of obtaining the above data through road test data, some data may be obviously wrong, so these wrong data need to be excluded. The method used by the present application is to use clustering algorithm to help find abnormal data. Similar or adjacent data are aggregated together by clustering algorithm to form various clustering sets, and those data objects located outside the clustering sets are naturally considered as abnormal data and can be excluded.

[0095] In addition to the above data cleaning method, other methods can be used to clean the above data to further improve the accuracy of the 5G network coverage conversion model and avoid errors. In particular, it is necessary to avoid situations where the 5G network coverage conversion model cannot recognize the data or cannot obtain corresponding output data even if there is data input.

[0096] After the above data is cleaned, data preprocessing can be performed, and the method of data preprocessing is different based on different data.

[0097] In an embodiment of the present application, the method of data preprocessing includes:

[0098] The 4G reference signal received power, 4G transmit power, 4G antenna gain, and 4G mobile phone receiving antenna number, 5G reference signal received power, 5G transmit power, 5G antenna gain, and 5G mobile phone receiving antenna number are standardized.

[0099] In a multi-index evaluation system, due to the different nature of each evaluation index, there are usually different dimensions and orders of magnitude. When the levels of each index differ greatly, if the original index value is directly used for analysis, the index with higher value will be highlighted in the comprehensive analysis, and the index with lower value level will be relatively weakened. Therefore, in order to ensure the reliability of the results, it is necessary to standardize the original index data. The corresponding 5G network coverage conversion model of the present application is also a multi-index evaluation system, which contains 4G reference signal received power, 4G transmit power, 4G antenna gain, and 4G mobile phone receiving antenna number, 5G reference signal received power, 5G transmit power, 5G antenna gain, and 5G mobile phone receiving antenna number. Therefore, it is also necessary to standardize these values, and the present application specifically standardizes the values.

[0100] Standardization is to scale the data in proportion so that it falls within a small specific interval. It is converted into a dimensionless pure value, which facilitates the comparison and weighting of indicators with different units or orders of magnitude. The specific standardization formula is as follows:

[0101] x = (X-mean) / std

[0102] Where x is the standardized value, X is the original value, mean is the mean, and std is the standard deviation.

[0103] Taking the 4G reference signal received power as an example, the present application collects one thousand 4G reference signal received powers through road testing, and the mean can represent the average of the one thousand 4G reference signal received powers, and the std represents the standard deviation of the one thousand 4G reference signal received powers. The standardized value is x, so x=(X-mean) / std, and thus all the one thousand 4G reference signal received powers can be standardized by using the formula. Similarly, the present application also processes the corresponding one thousand first difference feature data in the same way to achieve standardization. Finally, the 4G reference signal received power, the 5G reference signal received power, the second difference feature data, the third difference feature data and the fourth difference feature data are all within a small specific interval.

[0104] The pre-processing method for the frequency band is different. The present application pre-processes the frequency band by using one-hot encoding for the 4G frequency band and the 5G frequency band. One-hot encoding is to use an N-bit state register to encode N states, each state has its independent register bit, and at any time, only one bit is valid. After one-hot encoding processing of the frequency band, the first difference feature data can be obtained.

[0105] Through the above method, the 4G reference signal received power, the 5G reference signal received power, the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data after data cleaning and preprocessing can be obtained. Step S530 can be performed.

[0106] Step S530: Taking the pre-processed 4G reference signal received power and the difference between the 4G transmit power and the 5G transmit power, the difference between the 4G antenna gain and the 5G antenna gain, the difference between the 4G frequency band and the 5G frequency band, and the difference between the 4G mobile phone receiving antenna number and the 5G mobile phone receiving antenna number as input samples, taking the pre-processed 5G reference signal received power as output samples, using a pre-set machine learning model for training, and obtaining a 5G network coverage conversion model.

[0107] The present application can also combine the pre-processed 4G reference signal received power, the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data as input samples, and the pre-processed 5G reference signal received power as output samples, and use a pre-set machine learning model for training to obtain a 5G network coverage conversion model.

[0108] The machine learning model can use a LightGBM model, which has faster training efficiency; the lightGBM algorithm is used for model prediction, and the sparkMLlib supports the model, which can be used to train and call the model by using the spark engine. Spark is a fast general-purpose computing engine designed for large-scale data processing. It is easy to implement in engineering, and the model training can be offline. After training, batch prediction can be achieved.

[0109] In an embodiment of the present application, the present application can also introduce a grid position to realize the binding relationship of data. The grid position is the position converted from the actual map latitude and longitude with height to the Mercator coordinate system. One grid position corresponds to a region on the map, and a set of 4G reference signal receiving power, first difference feature data, second difference feature data, third difference feature data, fourth difference feature data and 5G reference signal receiving power exist in one grid position. Moreover, the 5G reference signal receiving power corresponds to the combination of the 4G reference signal receiving power, the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data. Therefore, the introduction of the grid position can quickly locate the geographical position, and facilitate the connection between the 5G reference signal receiving power and the combination of the 4G reference signal receiving power, the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data.

[0110] The above will be further illustrated by examples. The collected road test data of an area of fifteen kilometers may obtain ten thousand grid positions, corresponding to ten thousand sets of 4G reference signal receiving power, first difference feature data, second difference feature data, third difference feature data, fourth difference feature data and 5G reference signal receiving power data. The ten thousand sets of 4G reference signal receiving power, first difference feature data, second difference feature data, third difference feature data, fourth difference feature data and 5G reference signal receiving power data are input into the lightGBM model for training, with the 4G reference signal receiving power, first difference feature data, second difference feature data, third difference feature data and fourth difference feature data as input samples, and the corresponding 5G reference signal receiving power data as output samples. The 5G network coverage conversion model of the present application can be obtained. The corresponding relationship of the 5G reference signal receiving power data is based on the grid position, and the combination of the 4G reference signal receiving power, the first difference feature data, the second difference feature data, the third difference feature data and the fourth difference feature data in the same grid position corresponds to the 5G reference signal receiving power data.

[0111] After the model is trained by the above steps, the 5G network coverage data can be evaluated, specifically as step S230.

[0112] In step S230: 4G network coverage data of the to-be-evaluated area is extracted through the 4G measurement report data, difference feature data of the to-be-evaluated area is extracted, the 4G network coverage data of the to-be-evaluated area and the difference feature data of the to-be-evaluated area are input into the 5G network coverage conversion model, and 5G network coverage data of the to-be-evaluated area is obtained.

[0113] The measurement report refers to information that sends data once every 480 ms (470 ms on a signaling channel) on a service channel, and the data can be used for network evaluation and optimization. The measurement report data mainly comes from user equipment and base stations. The measurement data is reported to the OMC-R in the form of statistical data (which can be realized on the base station OMC-R) for storage or directly reported to the OMC-R in the form of sample data for storage.

[0114] The 4G network coverage data of the to-be-evaluated area can be obtained by extracting the 4G network coverage data of the to-be-evaluated area through the 4G measurement report data, and the difference feature data of the to-be-evaluated area is calculated in steps S310-S340, which will not be described here. The 4G reference signal received power corresponding to the to-be-evaluated area is obtained, and the difference feature data of the to-be-evaluated area is input into the 5G network coverage conversion model, and the 5G network coverage data of the to-be-evaluated area is obtained. In this step, the 4G reference signal received power of the to-be-evaluated area and the difference feature data of the to-be-evaluated area are also subjected to data cleaning and data preprocessing. The 5G reference signal received power of the to-be-evaluated area can be obtained by using this step, and the 5G network coverage quality of the to-be-evaluated area can be determined by using the 5G reference signal received power.

[0115] The 4G measurement report data is stored in the form of statistical data, so there are many location information and data content corresponding to the location information in the 4G measurement report data. The location information refers to data positioning, and the 4G network coverage data content is saved in the 4G measurement report.

[0116] In an embodiment of the present application, the method for extracting 4G network coverage data of the to-be-evaluated area through 4G measurement report data comprises the following steps:

[0117] Obtaining the data positioning of the 4G measurement report data, the data positioning is used to represent the location information of the to-be-evaluated area;

[0118] Obtaining the 4G network coverage data content corresponding to the data positioning.

[0119] For example, the 5G network coverage of area A needs to be evaluated, the data positioning corresponding to area A in the 4G measurement report data can be obtained, and then the 4G network coverage data content is obtained according to the data positioning to obtain the 4G reference signal receiving power. As for the position information, the latitude and longitude can be used to determine the position, or the grid position can be used. When the grid position is used for matching, the specific method is: the latitude and longitude and height of the map of the area to be evaluated are converted into the position in the Mercator coordinate system as the grid position of the area to be evaluated, the grid position of the area to be evaluated is matched with the grid position in the 4G measurement report data, and the grid position in the 4G measurement report data that matches successfully is taken as the grid position of the area to be evaluated. Using the above method, the data positioning of the area to be evaluated can be obtained, and then the 5G reference signal receiving power can be obtained according to the data positioning.

[0120] In an embodiment of the present application, the method for obtaining the data positioning of the 4G measurement report data is a method for using the measurement report fingerprint library. Figure 6 The flowchart of the method for obtaining the data positioning of the 4G measurement report data is schematically shown. As shown in Figure 6 The method for obtaining the data positioning of the 4G measurement report data includes steps S610-S630.

[0121] Step S610: Obtain a map and a fingerprint library corresponding to the map, and the fingerprint library is used to establish a mapping relationship between the received signal strength sample and the map position information.

[0122] The map can use a high-precision map or an existing ordinary map. The fingerprint library corresponding to the map is to associate a position in the map with a certain "fingerprint", and one position corresponds to one unique fingerprint. This fingerprint can be single-dimensional or multi-dimensional, for example, the positioning device receives or transmits information, and the fingerprint can be one feature or multiple features (most commonly signal strength) of the information or signal. The fingerprint used in the present application corresponds to the received signal strength.

[0123] The received signal strength (RSS) depends on the position of the receiver. The RSS is easy to obtain because it is necessary for the normal operation of most wireless communication devices. Many communication systems need RSS information to sense the quality of the link, implement handover, adapt the transmission rate, and the like. RSS is not affected by the bandwidth of the signal, and it is not necessary to have a high bandwidth, so RSS is a very popular signal feature and is widely used in positioning and can be directly obtained through the measurement report.

[0124] The application establishes a mapping relationship between the received signal strength sample and the map location information, that is, one received signal strength information corresponds to one map location information, and the reason for such mapping is based on the characteristic of the multipath structure of the signal. The multipath structure of the signal refers to that when the radio signal propagates, the "rays" can be reflected on smooth planes (such as the walls and floors of buildings), diffracted when encountering sharp edges, and scattered when encountering small objects (such as leaves). The radio signal emitted by the transmitting source can propagate through multiple paths to the same position, so multiple rays will be received at one position, each ray having different energy intensity and time delay. Each ray reaching the receiver is called a multipath component, and the multipath structure of the channel refers to this group (multiple rays) of signal intensity and time delay. If the bandwidth of the signal is large enough (such as using direct sequence spread spectrum technology or ultra-wideband technology), each multipath component can be decomposed and processed at the receiver. Therefore, the multipath structure obtained at a certain position depends on the actual environment and can be used as a location fingerprint. That is, the received signal strength at each position is unique, so a mapping relationship between the received signal strength and the location information can be established to obtain the fingerprint library.

[0125] After obtaining the fingerprint library, the data positioning can be acquired, and step S620 is performed.

[0126] Step S620: According to the 4G measurement report data, the received signal strength of the area to be evaluated is acquired.

[0127] The received signal strength of the area to be evaluated can be directly obtained from the 4G measurement report data, and the received signal strength can also be acquired by the following formula:

[0128] RSS = Pt-K-10a log10d

[0129] Wherein, a is called the path loss index, Pt is the transmission power, K is a constant depending on the environment and frequency, and RSS represents the received signal strength.

[0130] After obtaining the received signal strength, the data can be input into the fingerprint library, and the data positioning can be acquired, as in step S630.

[0131] Step S630: The received signal strength of the area to be evaluated is matched and detected with the received signal strength sample in the fingerprint library, and the map location information corresponding to the received signal strength matched successfully is taken as the location information of the area to be evaluated.

[0132] Since the fingerprint library has established a mapping relationship between a lot of location information and received signal strength, when the received signal strength in the fingerprint library matches the received signal strength of the area to be evaluated, the location information corresponding to the matching received signal strength can be used as the location information of the area to be evaluated. The location information here corresponds to grid location information.

[0133] For example, the fingerprint library contains 10,000 mapping relationships, one of which has a received signal strength RSS of -65dbm and corresponding grid location information of (1, 0). If the received signal strength of the area to be evaluated is also -65dbm, the grid location of the area to be evaluated can be obtained as (1, 0). Converting the grid location to latitude and longitude position, the location information of the area to be evaluated can be obtained.

[0134] For the matching of received signal strength, the method used in the present application is as follows. In an embodiment of the present application, the method for matching and detecting the received signal strength of the area to be evaluated with the received signal strength samples in the fingerprint library comprises:

[0135] When the received signal strength of the area to be evaluated is the same as the received signal strength sample in the fingerprint library or the similarity of the received signal strength of the area to be evaluated and the received signal strength in the fingerprint library is within a preset threshold range, the location information corresponding to the received signal strength in the fingerprint library is used as the location information of the area to be evaluated.

[0136] The setting of the preset threshold range can be based on the number of data samples of the fingerprint library. If the number of data samples of the fingerprint library is small, it means that the matching success probability is small, so the preset threshold range can be set to a larger range. Conversely, if the number of data samples of the fingerprint library is large, the threshold range can be lowered to improve the accuracy of matching.

[0137] Using the above method, the location information of the area to be evaluated can be obtained. However, in actual use, the location information obtained by step S630 is not accurate enough, because the area to be evaluated includes a primary cell area and a neighbor cell area, and the location information obtained by step S630 is usually only the location information of the primary cell area.

[0138] For a region, in order to improve the network coverage of the region, a wireless access point (wireless AP) is installed in the region, which is used as a wireless switch of the wireless network. The wireless AP is an access point for mobile computer users to enter the wired network, mainly used in broadband homes, building interiors and park interiors, and can cover tens of meters to hundreds of meters. In the process of generating the 4G measurement report, the received signal strength obtained is often the maximum received signal strength of each cell in the evaluation area, that is, the best position for connection with the wireless access point. The area with the best connection with the wireless access point is usually called the primary cell area, and the other cells in the evaluation area are all neighbor cell areas. Therefore, the position information obtained through step S630 can only represent the position information of the primary cell area. In actual network coverage deployment, in order to better provide network services, the network coverage of the neighbor cell area also needs to be considered. Therefore, the application considers the signal conditions of the neighbor cell area in the process of received signal strength field strength matching, specifically, the similarity of the received signal strength of the evaluation area and the received signal strength in the fingerprint library is calculated, and the specific formula is as follows:

[0139]

[0140] Wherein, s i (i = 1,... M) represents the strength error of the received signal strength of the i-th neighbor cell area in the 4G measurement report data and the received signal strength in the fingerprint library. By multiplying the strength error of each neighbor cell area, the similarity of the received signal strength of the evaluation area and the received signal strength in the fingerprint library is obtained, and the matching judgment can be made based on the similarity.

[0141] Wherein, the calculation formula of the strength error is as follows:

[0142]

[0143] Wherein, g(i) represents the weight of different neighbor cell areas relative to the primary cell area in the 4G measurement report data, which is a preset value, and different values can be set according to different cells, wherein the weight of the neighbor area is higher. For example, g(1) is greater than g(2).

[0144] Pi represents the received signal strength of the i-th neighbor cell area, which can be calculated according to the corresponding received signal strength formula in step S620.

[0145] represents the received signal strength corresponding to the area in the fingerprint library that has the same absolute radio frequency channel number and physical layer cell identification as the i-th neighbor cell area, that is, in the fingerprint library, whether there is an area whose absolute radio frequency channel number (earfcn) and physical layer cell identification are the same, and the received signal strength corresponding to the area is taken as The radio frequency channel number (PCI) and the physical layer cell identification are both determined values, which can be obtained by querying the operator. Therefore, using this method, the value of the strength error s i can be obtained.

[0146] σ represents a preset value, and when there is no area in the fingerprint library that has the same absolute radio frequency channel number and physical layer cell identification as the i-th neighbor cell area, the value of s i is σ. The value of σ is generally a very small value.

[0147] Using the above method, more accurate position information of the area to be evaluated can be obtained, so that more accurate 4G reference signal received power and difference characteristic data can be obtained based on the accurate position information, and the output 5G reference signal received power is more accurate.

[0148] In an embodiment of the present application, the method for obtaining 5G network coverage data of the area to be evaluated further comprises:

[0149] The data positioning of the neighbor cell area in the area to be evaluated is obtained according to the calculation formula of the similarity of the received signal strength and the received signal strength in the fingerprint library. The specific calculation method is to determine different signal reception strengths according to different s i information, and then to obtain the data positioning of different neighbor cell areas by comparing the signal reception strengths with the fingerprint library.

[0150] The 4G reference signal received power and difference characteristic data of each neighbor cell area in the area to be evaluated are obtained according to the data positioning. When the data positioning of each neighbor cell area is determined, the 4G reference signal received power and difference characteristic data of each neighbor cell area can be measured.

[0151] The 4G reference signal received power and difference characteristic data of the neighbor cell area are input into the 5G network coverage conversion model to obtain the 5G reference signal received power of the neighbor cell area.

[0152] The 5G reference signal received power of each neighbor cell area and the 5G reference signal received power of the main cell area are combined to obtain the 5G reference signal received power of the area to be evaluated.

[0153] The 5G reference signal received power of the main cell area is the 5G reference signal received power calculated by step S230. Combining the 5G reference signal received power of each neighbor cell area and the 5G reference signal received power of the main cell area means marking the corresponding 5G reference signal received power on the main cell area and each neighbor cell area of the to-be-evaluated area, as shown in Figure 7 , Figure 7 The effect diagram of the application combining the 5G reference signal received power of each neighbor cell area and the 5G reference signal received power of the main cell area is schematically shown. By combining the 5G reference signal received power of each neighbor cell area and the 5G reference signal received power of the main cell area on the corresponding grid of the to-be-evaluated area. As shown in Figure 8 , Figure 8 The enlarged view of the to-be-evaluated area grid in the application Figure 7 is schematically shown. As can be seen from the figure, the grid corresponding to one to-be-evaluated area contains the main cell area A and four neighbor cell areas B, C, D and E, wherein the main cell area A is closest to the AP device. For 5G reference signal received power combination, the 5G reference signal received power of the main cell area A and the four neighbor cell areas B, C, D and E can be marked on the area in the corresponding grid, or the 5G network coverage of the to-be-evaluated area can be directly represented by a column chart or the like.

[0154] The following specifically discloses an actual application embodiment. The applicant collects 4G measurement report data of a test area according to the method of the application, and on this basis, calculates 5G coverage capability to obtain 5G network coverage data, wherein the test area corresponds to the to-be-evaluated area of the application.

[0155] The main information of the test area is as follows: the number of test area macro stations: 102, the coverage area: 18.3Km2, the number of covered buildings: 4189, the average building height: 18.43m, and the average station spacing: 455m.

[0156] After collecting the information of the test area, the evaluation of the 5G network coverage is started. First, the 4G measurement report data of the test area for 3 days is collected after cleaning and stored in the database; the fingerprint library data is generated, and the fingerprint library is corrected by road test data; the positioning algorithm of the fingerprint library is used to obtain the positioning measurement report data, and the 4G reference signal received power and the difference characteristic data in the measurement report data are extracted, and then the 4G reference signal received power and the difference characteristic data are input into the pre-trained 5G network coverage conversion model to obtain the 5G reference signal received power, and according to the 5G reference signal received power, the network coverage of the test area can be obtained.

[0157] The final network coverage result of the specific above test area is as follows: total network coverage rate: 97%, the total network coverage rate is to calculate the size of the 5G reference signal received power, locate the area with network coverage in the area where the 5G reference signal received power is greater than the set value, and the total network coverage rate is the area ratio of the area with network coverage to the whole area to be evaluated. The indoor coverage rate is 95%, and the outdoor coverage rate is 100%. The calculation method of the indoor coverage rate and the outdoor coverage rate is similar to that of the total network coverage rate. The indoor position and the outdoor position are determined by measuring the report data, and the average value of all indoor 5G reference signal received power or all outdoor 5G reference signal received power is calculated. The average coverage level is -86.49 dBm, which is the average value of all 5G reference signal received power. The indoor average coverage level is -95.40 dBm, and the outdoor average coverage level is -77.85 dBm. The calculation method of the indoor average coverage level and the outdoor average coverage level is similar to that of the average coverage level. The indoor position and the outdoor position are determined by measuring the report data, and the average value of all indoor 5G reference signal received power or all outdoor 5G reference signal received power is calculated. Among them, when the application is counted, the coverage rate is greater than -110 dBm as network coverage. Therefore, by using the 5G network coverage evaluation method of the application, the network coverage of the test area can be obtained.

[0158] The application extracts the 4G network coverage data of road testing and the difference feature data of 4G network and 5G network to combine as input samples, extracts the 5G network coverage data of road testing as output samples, trains the preset machine learning model, obtains the 5G network coverage conversion model, and inputs the 4G network coverage data of the to-be-evaluated area and the difference feature data of the to-be-evaluated area into the 5G network coverage conversion model to obtain the result of the 5G network coverage data of the to-be-evaluated area. The application adds the important feature of difference feature data when predicting the 5G network coverage, reduces the error between the 5G network coverage prediction result and the actual result, and when the 5G network coverage conversion model is trained, the 4G network coverage data and the difference feature data of the to-be-evaluated area can be directly input to obtain the 5G network coverage data of the to-be-evaluated area, which is very efficient and does not need additional steps. At the same time, the application directly uses the data of road testing and the measurement report data during model training, without the form of a simulation map, which is low in cost. The application uses the present network 4G data combined with the machine learning algorithm to make the result more real and effective, and the obtained coverage data is more accurate than the original pure simulation method. Moreover, the application can still perform 5G coverage estimation in the area without a simulation map, and the coverage effect is wider. It is beneficial to accurately deploy the 5G network in the to-be-evaluated area.

[0159] The above discloses the content of the 5G network coverage evaluation method of the present application. Next, other aspects of the present application are disclosed.

[0160] According to an aspect of an embodiment of the present application, the present application provides a 5G network coverage evaluation device 900, as shown in Figure 9 Figure 9 The structural block diagram of the 5G network coverage evaluation device of the present application is schematically shown. It includes:

[0161] The extraction module 910 is configured to extract 4G network coverage data, 5G network coverage data and difference feature data of road testing, the 4G network coverage data is used to evaluate the 4G network coverage quality, the 5G network coverage data is used to evaluate the 5G network coverage quality, and the difference feature data includes a set of feature differences of 4G network and 5G network in network coverage;

[0162] The model establishing module 920 is configured to train a preset machine learning model by taking the combination of 4G network coverage data and difference feature data as input samples and taking the 5G network coverage data corresponding to the 4G network coverage data as output samples, to obtain a 5G network coverage conversion model;

[0163] The evaluation module 930 is configured to extract 4G network coverage data of the area to be evaluated by 4G measurement report data, extract difference feature data of the area to be evaluated, input the 4G network coverage data of the area to be evaluated and the difference feature data of the area to be evaluated into the 5G network coverage conversion model, and obtain 5G network coverage data of the area to be evaluated.

[0164] In some embodiments of the present application, based on the above technical solution, the 4G measurement report data includes data positioning and 4G network coverage data content, and the evaluation module 930 includes:

[0165] The data acquisition unit is configured to acquire the data positioning of the 4G measurement report data, and the data positioning is used to represent the position information of the area to be evaluated;

[0166] Acquire the 4G network coverage data content corresponding to the data positioning.

[0167] In some embodiments of the present application, based on the above technical solution, the data acquisition unit includes a position acquisition unit, and the position acquisition unit is configured to acquire a map and a fingerprint library corresponding to the map, and the fingerprint library is used to establish a mapping relationship between a received signal strength sample and map position information;

[0168] According to the 4G measurement report data, the received signal strength of the area to be evaluated is acquired;

[0169] ​The received signal strength of the to-be-evaluated area is matched with the received signal strength samples in the fingerprint library, and the map position information corresponding to the received signal strength that is successfully matched is taken as the position information of the to-be-evaluated area.

[0170] In some embodiments of the present application, based on the above technical solutions, the position obtaining unit comprises a matching unit, which is configured to take the position information corresponding to the received signal strength in the fingerprint library as the position information of the to-be-evaluated area when the received signal strength of the to-be-evaluated area is the same as the received signal strength sample in the fingerprint library or the similarity of the received signal strength of the to-be-evaluated area and the received signal strength in the fingerprint library is within a preset threshold range.

[0171] In some embodiments of the present application, based on the above technical solutions, the to-be-evaluated area comprises a primary cell area and a neighbor cell area, and the matching unit comprises a similarity calculation unit, which is configured to calculate the similarity of the received signal strength of the to-be-evaluated area and the received signal strength in the fingerprint library according to the following formula:

[0172]

[0173] wherein s i (i = 1, …, M) represents the intensity error of the i-th neighbor cell area received signal strength in the 4G measurement report data and the received signal strength in the fingerprint library, and the calculation formula of the intensity error is as follows:

[0174]

[0175] wherein g(i) represents the weight of different neighbor cell areas relative to the primary cell area in the 4G measurement report data, P i represents the received signal strength of the i-th neighbor cell area, represents the received signal strength corresponding to the area in the fingerprint library that is the same as the i-th neighbor cell area in terms of the absolute radio frequency channel number and the physical layer cell identifier, respectively, and σ represents a preset value, s i is equal to σ when there is no area in the fingerprint library that is the same as the i-th neighbor cell area in terms of the absolute radio frequency channel number and the physical layer cell identifier, respectively.

[0176] In some embodiments of the present application, based on the above technical solutions, the model establishing module 920 comprises a data receiving unit, a data preprocessing unit and a model training unit,

[0177] The data receiving unit is configured to receive the 4G reference signal received power, the 4G transmit power, the 4G antenna gain, the 4G frequency band and the 4G mobile phone receiving antenna quantity, the 5G reference signal received power, the 5G transmit power, the 5G antenna gain, the 5G frequency band and the 5G mobile phone receiving antenna quantity of the road test.

[0178] The data preprocessing unit is configured to preprocess the 4G reference signal receiving power, the 4G transmitting power, the 4G antenna gain, the 4G frequency band, and the 4G mobile phone receiving antenna quantity, the 5G reference signal receiving power, the 5G transmitting power, the 5G antenna gain, the 5G frequency band, and the 5G mobile phone receiving antenna quantity so as to be input into the machine learning model for training.

[0179] The model training unit is configured to combine the preprocessed 4G reference signal receiving power and 4G transmitting power difference, 4G antenna gain and 5G antenna gain difference, 4G frequency band and 5G frequency band difference, and 4G mobile phone receiving antenna quantity and 5G mobile phone receiving antenna quantity difference as input samples, and the preprocessed 5G reference signal receiving power as an output sample, to train a preset machine learning model to obtain a 5G network coverage conversion model.

[0180] In some embodiments of the present application, based on the above technical solutions, the data preprocessing unit comprises a standardization unit and an encoding unit,

[0181] The standardization unit is configured to standardize the 4G reference signal receiving power, the 4G transmitting power, the 4G antenna gain, and the 4G mobile phone receiving antenna quantity, the 5G reference signal receiving power, the 5G transmitting power, the 5G antenna gain, and the 5G mobile phone receiving antenna quantity.

[0182] The encoding unit is configured to use one-hot encoding processing for the 4G frequency band and the 5G frequency band.

[0183] The specific details of the 5G network coverage evaluation device provided in the embodiments of the present application have been described in detail in the corresponding method embodiments, which will not be repeated here.

[0184] According to an aspect of an embodiment of the present application, a computer readable medium having a computer program stored thereon is provided, the computer program being executed by a processor to implement the 5G network coverage evaluation method in the above technical solutions.

[0185] According to an aspect of an embodiment of the present application, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the 5G network coverage evaluation method in the above technical solutions by executing the executable instructions.

[0186] According to an aspect of an embodiment of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device performs the 5G network coverage evaluation method as in the above technical solutions.

[0187] Figure 10 A computer system structure block diagram of an electronic device for implementing embodiments of the present application is schematically shown.

[0188] It should be noted that, Figure 10 The computer system 1000 of the electronic device shown is only an example, and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0189] As Figure 10 shown, the computer system 1000 includes a central processing unit 1001 (CPU), which can perform various appropriate actions and processes according to programs stored in a read-only memory 1002 (ROM) or loaded from a storage portion 1008 into a random access memory 1003 (RAM). Various programs and data required for system operation are also stored in the random access memory 1003. The central processing unit 1001, the read-only memory 1002, and the random access memory 1003 are connected to each other through a bus 1004. An input / output interface 1005 (I / O interface) is also connected to the bus 1004.

[0190] The following components are connected to the input / output interface 1005: an input portion 1006 including a keyboard, a mouse, and the like; an output portion 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage portion 1008 including a hard disk, and the like; and a communication portion 1009 including a network interface card such as a local area network card, a modem, and the like. The communication portion 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output interface 1005 as necessary. A removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 1010 as necessary, so that a computer program read therefrom is installed in the storage portion 1008 as necessary.

[0191] In particular, according to embodiments of the present application, the processes described in the various method flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 1009, and / or installed from the removable media 1011. When the computer program is executed by the central processing unit 1001, various functions defined in the system of the present application are executed.

[0192] It should be noted that the computer readable medium shown in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium, which can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or apparatus. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination of the above.

[0193] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0194] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0195] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0196] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0197] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for 5G network coverage evaluation, characterized in that, The method comprises the following steps: extracting 4G network coverage data, 5G network coverage data and difference feature data of the road test, the 4G network coverage data being used to evaluate the 4G network coverage quality, the 5G network coverage data being used to evaluate the 5G network coverage quality, and the difference feature data comprising first difference feature data, second difference feature data, third difference feature data and fourth difference feature data; wherein the first difference feature data comprises a difference range of 4G frequency bands and 5G frequency bands, the second difference feature data comprises a difference value of 4G transmission power and 5G transmission power, the third difference feature data comprises a reception level difference value calculated according to an antenna difference value of 4G reception antennas and 5G reception antennas, and the fourth difference feature data comprises a difference value of 4G antenna gain and 5G antenna gain; using the 4G network coverage data and the difference feature data as input samples, using the 5G network coverage data corresponding to the 4G network coverage data as output samples, and training a preset machine learning model to obtain a 5G network coverage conversion model; extracting 4G network coverage data of an evaluation area through 4G measurement report data, extracting difference feature data of the evaluation area, inputting the 4G network coverage data of the evaluation area and the difference feature data of the evaluation area into the 5G network coverage conversion model, and obtaining 5G network coverage data of the evaluation area.

2. The method of 5G network coverage evaluation according to claim 1, characterized in that, The 4G measurement report data comprises data positioning and 4G network coverage data content. The 4G network coverage data of the evaluation area is extracted through the 4G measurement report data, and the method comprises the following steps: obtaining data positioning of the 4G measurement report data, wherein the data positioning is used to represent position information of the evaluation area; obtaining 4G network coverage data content corresponding to the data positioning.

3. The 5G network coverage evaluation method of claim 2, wherein, The method for obtaining data positioning of the 4G measurement report data comprises the following steps: obtaining a map and a fingerprint library corresponding to the map, wherein the fingerprint library is used to establish a mapping relationship between a reception signal strength sample and map position information; obtaining a reception signal strength of the evaluation area according to the 4G measurement report data; performing matching detection on the reception signal strength of the evaluation area and a reception signal strength sample in the fingerprint library, and taking map position information corresponding to the reception signal strength which is successfully matched as the position information of the evaluation area.

4. The 5G network coverage evaluation method according to claim 3, wherein the matching detection on the reception signal strength of the evaluation area and the reception signal strength sample in the fingerprint library comprises the following steps: when the reception signal strength of the evaluation area is the same as the reception signal strength sample in the fingerprint library or the similarity of the reception signal strength of the evaluation area and the reception signal strength sample in the fingerprint library is within a preset threshold range, taking position information corresponding to the reception signal strength in the fingerprint library as the position information of the evaluation area. The evaluation area comprises a primary cell area and a neighbor cell area, and the calculation formula of the similarity of the reception signal strength of the evaluation area and the reception signal strength sample in the fingerprint library is as follows:

5. The method of 5G network coverage evaluation according to claim 4, characterized in that, ​ ； wherein s i represents the strength error of the received signal strength of the i-th neighboring cell area in the 4G measurement report data and the received signal strength in the fingerprint library, i = 1,..., M, M represents the total number of neighboring cell areas in the 4G measurement report data, and the calculation formula of the strength error is as follows: ； wherein g(i) represents a weight of a different neighbor cell area with respect to a primary cell area in the 4G measurement report data, P i represents a received signal strength of an i-th neighbor cell area, k represents a received signal strength of an area corresponding to the i-th neighbor cell area in the fingerprint library, wherein the area has the same absolute radio frequency channel number and physical layer cell identification as the i-th neighbor cell area, respectively, and σ represents a preset value, wherein when there is no area in the fingerprint library having the same absolute radio frequency channel number and physical layer cell identification as the i-th neighbor cell area, respectively, s i is equal to σ.

6. The method of 5G network coverage evaluation according to claim 1, characterized in that, The 4G network coverage data and the difference feature data are combined as input samples, 5G network coverage data corresponding to the 4G network coverage data is used as output samples, and a preset machine learning model is trained to obtain a 5G network coverage conversion model; comprising: Receiving 4G reference signal receiving power, 4G transmit power, 4G antenna gain, 4G frequency band, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmit power, 5G antenna gain, 5G frequency band, and 5G mobile phone receiving antenna quantity of road testing; The 4G reference signal receiving power, 4G transmit power, 4G antenna gain, 4G frequency band, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmit power, 5G antenna gain, 5G frequency band, and 5G mobile phone receiving antenna quantity are preprocessed for inputting into a machine learning model for training; The 4G reference signal receiving power and the 4G transmit power and 5G transmit power difference, the 4G antenna gain and the 5G antenna gain difference, the 4G frequency band and the 5G frequency band difference, and the 4G mobile phone receiving antenna quantity and the 5G mobile phone receiving antenna quantity difference are combined as input samples, the 5G reference signal receiving power after preprocessing is used as output samples, and a preset machine learning model is trained to obtain a 5G network coverage conversion model.

7. The method of 5G network coverage evaluation according to claim 6, characterized in that, The 4G reference signal receiving power, 4G transmit power, 4G antenna gain, 4G frequency band, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmit power, 5G antenna gain, 5G frequency band, and 5G mobile phone receiving antenna quantity are preprocessed; Comprising: The 4G reference signal receiving power, 4G transmit power, 4G antenna gain, and 4G mobile phone receiving antenna quantity, 5G reference signal receiving power, 5G transmit power, 5G antenna gain, and 5G mobile phone receiving antenna quantity are standardized processed; The 4G frequency band and the 5G frequency band are processed by one-hot encoding. 8.A 5G network coverage evaluation device, characterized in that, Comprising: The extraction module is configured to extract 4G network coverage data, 5G network coverage data, and difference feature data of road testing, the 4G network coverage data is used to evaluate 4G network coverage quality, the 5G network coverage data is used to evaluate 5G network coverage quality, and the difference feature data includes first difference feature data, second difference feature data, third difference feature data, and fourth difference feature data; wherein the first difference feature data includes a difference range of 4G frequency band and 5G frequency band, the second difference feature data includes a difference value of 4G transmit power and 5G transmit power, the third difference feature data includes a receiving level difference value calculated according to an antenna difference value of 4G receiving antenna quantity and 5G receiving antenna quantity, and the fourth difference feature data includes a difference value of 4G antenna gain and 5G antenna gain; The model establishment module is configured to combine the 4G network coverage data and the difference feature data as input samples, use 5G network coverage data corresponding to the 4G network coverage data as output samples, train a preset machine learning model, and obtain a 5G network coverage conversion model. The evaluation module is configured to extract 4G network coverage data of the to-be-evaluated area through 4G measurement report data, extract difference feature data of the to-be-evaluated area, input the 4G network coverage data of the to-be-evaluated area and the difference feature data of the to-be-evaluated area into the 5G network coverage conversion model, and obtain 5G network coverage data of the to-be-evaluated area. 9.A computer readable medium having stored thereon a computer program which, when executed by a processor, implements the 5G network coverage evaluation method of any one of claims 1 to 7.

10. An electronic device, comprising: comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to implement the 5G network coverage evaluation method of any one of claims 1 to 7 via execution of the executable instructions.

Images