Intelligent terminal automatic survey data analysis and processing method and system based on deep learning

Abstract

The application relates to the technical field of base station survey, and discloses an intelligent terminal automatic survey data analysis and processing method and system based on deep learning, which comprises a server, an intelligent terminal and a monitoring device; the server is used for receiving base station information corresponding to each base station collected by the monitoring device; the intelligent terminal comprises a data acquisition module, a data calling module, a plant identification module and a processing module; the processing module is used for determining the base station influence degree of main plants corresponding to each base station according to surrounding environment data corresponding to each base station, plant standard growth demand information, base station basic information and base station surrounding plant growth video information; and the analysis module is used for adjusting and optimizing base station equipment corresponding to each base station based on a preset base station equipment adjustment strategy according to the base station influence degree of main plants corresponding to each base station, and outputting a corresponding base station equipment adjustment scheme.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of base station survey, in particular to an intelligent terminal automatic survey data analysis and processing method and system based on deep learning. BACKGROUND

[0002] In the field of communication, the rationality of base station equipment setting survey evaluation has long been focused on the performance of the equipment itself and the communication coverage demand. On the one hand, the performance indicators of the base station equipment such as signal transmission power, receiving sensitivity and transmission rate are concerned to ensure that the equipment can stably and efficiently realize the transceiving and transmission of signals to meet the growing demand for communication data flow. On the other hand, according to geographical area, population density, building distribution and other factors, the layout and coverage range of the base station are optimized to strive to achieve extensive and uniform coverage of communication signals and reduce signal blind area and interference. However, this traditional evaluation mode has significant limitations. It completely ignores the influence of base station operation on the surrounding ecological environment, especially on plant growth.

[0003] Therefore, there is an urgent need for an intelligent terminal automatic survey data analysis and processing method and system based on deep learning, which can solve the problem that the influence of base station operation on surrounding plants is ignored in the prior art, resulting in the growth of surrounding plants being affected, and ensure the stable operation of communication services while reducing the negative impact of base stations on plant growth. SUMMARY

[0004] One of the purposes of the present application is to provide an intelligent terminal automatic survey data analysis and processing method and system based on deep learning, which can solve the problem that the influence of base station operation on surrounding plants is ignored in the prior art, resulting in the growth of surrounding plants being affected, and ensure the stable operation of communication services while reducing the negative impact of base stations on plant growth.

[0005] In order to achieve the above purpose, an intelligent terminal automatic survey data analysis and processing method and system based on deep learning are provided, which include a server, an intelligent terminal for surveying the surrounding area of each base station site, and a monitoring device for collecting base station basic information corresponding to each base station.

[0006] The server is used to receive the base station information collected by the monitoring device, and the base station information includes base station basic information and base station surrounding plant growth video information.

[0007] The intelligent terminal includes:

[0008] The data collection module is configured to collect surrounding environment data within a preset range of each base station, wherein the surrounding environment data includes meteorological data, temperature and humidity data, soil pH data, light intensity data, and gas concentration data.

[0009] The data retrieval module is configured to retrieve base station basic information and base station surrounding plant growth video information corresponding to each base station from the server.

[0010] The plant identification module is configured to identify the plant type and the current growth state of the main plant corresponding to each base station based on the retrieved base station surrounding plant growth video information, and retrieve plant standard growth requirement information corresponding to the plant type in the current growth state from the server.

[0011] The processing module is configured to determine the base station influence degree of the main plant corresponding to each base station based on the surrounding environment data, the plant standard growth requirement information, the base station basic information, and the base station surrounding plant growth video information corresponding to each base station, and based on a preset base station influence degree determination strategy.

[0012] The analysis module is configured to adjust and optimize the base station equipment corresponding to each base station based on a preset base station equipment adjustment strategy, and output a corresponding base station equipment adjustment scheme.

[0013] The technical principle and effect of the present scheme are as follows: In the present scheme, the monitoring device is responsible for collecting base station basic information (such as base station location, equipment parameters, etc.) and base station surrounding plant growth video information, and transmitting these information to the server for storage. The data collection module of the intelligent terminal collects various surrounding environment data (meteorological data, temperature and humidity data, soil pH data, light intensity data, and gas concentration data) within a preset range of the base station installation, providing basic environmental data for subsequent analysis.

[0014] The base station basic information and the base station surrounding plant growth video information are obtained from the server, and are integrated with the surrounding environment data collected by the intelligent terminal to form a comprehensive data set about each base station and its surrounding environment.

[0015] The type and current growth state of the main plant surrounding the base station are identified by using the base station surrounding plant growth video information and deep learning technology (such as convolutional neural network). After identifying the plant type, the plant standard growth requirement information (such as suitable temperature range, light intensity, etc.) corresponding to the plant in the current growth state is retrieved from the server.

[0016] According to the integrated surrounding environment data, plant standard growth requirement information, base station basic information and plant growth video information, the influence degree of each base station on the surrounding main plants is evaluated according to a preset base station influence degree determination strategy (which may involve analyzing the heat, electromagnetic radiation and the like generated by the operation of the base station equipment and the satisfaction or interference of the plant growth requirements).

[0017] According to the evaluation result of the base station influence degree, the equipment in each base station is adjusted and optimized according to a preset base station equipment adjustment strategy (which may include adjusting the power, heat dissipation mode, installation position and the like of the base station equipment), so as to reduce the negative influence of the base station on the surrounding plant growth, and output a specific base station equipment adjustment scheme.

[0018] Traditional survey and evaluation of the rationality of base station equipment setting mainly focuses on the performance of the equipment itself and the communication coverage requirement, and ignores the influence on the surrounding ecological environment, especially the plants. The present scheme takes the influence of the base station on the surrounding plants as the key breakthrough point. By comprehensively collecting multi-dimensional environmental data such as meteorological data, temperature and humidity, soil pH, light intensity and gas concentration around the base station, combining with the accurate identification of plant types and growth states by deep learning technology, and the basic information of the base station, a new evaluation system is constructed. This innovative evaluation method can clearly understand how the heat, electromagnetic radiation and the like generated by the operation of the base station equipment interfere with the standard growth requirements of the plants.

[0019] The data collection dimension of the present scheme is rich, covering multiple key aspects of the environment around the base station. Meteorological data reflect the changes in macro climate conditions, temperature and humidity data reflect the comfort level of local micro environment, soil pH is related to the growth basis of plants, light intensity directly affects plant photosynthesis, and gas concentration is related to plant respiration and survival quality. These multi-dimensional data are interrelated and mutually verified, greatly enriching the description of the environmental state around the base station. Through comprehensive analysis of these massive complex data by a deep learning model, the complex cause-and-effect relationship between the operation of the base station and the growth of the plants can be deeply mined

[0020] Based on accurate evaluation and deep analysis, the present scheme can effectively reduce the influence of the base station on the surrounding plants. By adjusting the base station equipment according to the preset strategy, such as optimizing the heat dissipation structure to control the local temperature, reasonably adjusting the electromagnetic radiation intensity and range, adjusting the equipment installation angle to reduce the shading of light and the like, the operation of the base station equipment is coordinated with the growth requirements of the plants. In the natural reserve area around the base station, after the equipment adjustment, the growth state of the rare plants is obviously improved, and the degree of interference of the base station is greatly reduced, which effectively protects the local ecological environment.

[0021] The scheme reduces the negative impact of the base station on plant growth while ensuring stable operation of communication services. By optimizing the base station device settings, not only the device operation efficiency is improved and energy waste is reduced, but also the harmonious coexistence of the base station and the surrounding ecological environment is promoted. In some ecologically fragile areas, reasonable adjustment of the base station equipment not only meets the local communication demand, but also avoids damage to the ecological system, realizing the coordinated and sustainable development of communication infrastructure construction and ecological environment protection, and providing a feasible example for future communication network construction and ecological environment-friendly development.

[0022] Further, the preset base station influence degree determination strategy is:

[0023] According to the obtained meteorological data, temperature and humidity data, soil pH data, light intensity data, gas concentration data and plant standard growth requirement information around the base station, based on a preset element influence factor calculation formula, the element influence factor corresponding to each data is calculated;

[0024] The preset element influence factor calculation formula is:

[0025]

[0026] In the formula, is the element influence factor corresponding to data i, n is the total number of surrounding environment data, is the current data value corresponding to data i, is the standard minimum data value corresponding to data i, is the standard maximum data value corresponding to data i, is the first weight value and the second weight value corresponding to data i;

[0027] According to the plant standard growth requirement information and the plant growth video information around the base station, based on a preset plant recognition model, the plant morphological feature data corresponding to the main plants is identified, and the element influence factor corresponding to each plant morphological feature data is calculated; the plant morphological feature data includes plant height, leaf number and leaf color;

[0028] The plant recognition model adopts a convolutional neural network model including an input layer, multiple convolutional layers, a pooling layer, a fully connected layer and an output layer; ResNet50 or VGG16 is used as a basic network, and a custom classification layer is added at the end thereof; after collecting and labeling plant image data, the model is input for iterative training, a cross-entropy loss function is used, training parameters are set, and the model converges until the model converges;

[0029] ​According to the element influence factor corresponding to each data and the base station basic information, the base station influence degree of the main plant corresponding to the base station is calculated based on a preset base station influence degree calculation formula.

[0030] The preset base station influence degree calculation formula is:

[0031]

[0032] In the formula, is the base station influence degree of the main plant corresponding to the base station, is a corresponding weighting coefficient, is the element influence factor corresponding to the plant height, is the element influence factor corresponding to the leaf number, is the element influence factor corresponding to the leaf color, are respectively a sensitive coefficient of the main plant to the base station transmit power and a sensitive coefficient of the main plant to the antenna height, is the current base station transmit power corresponding to the base station, is the maximum transmit power corresponding to the same type of base station, is the current antenna height corresponding to the base station, is the maximum antenna height corresponding to the same type of base station.

[0033] Beneficial effects: through the preset element influence factor calculation formula, various environmental data such as weather, temperature and humidity, soil pH, light intensity, and gas concentration around the base station can be accurately quantified into corresponding element influence factors. This quantification method fully considers the standard range of data. When the data is between the minimum and maximum values of the standard, the element influence factor is 0, which means that the environmental factor is within the suitable growth range of the plant and has no negative impact on the growth of the plant. When the data exceeds this range, the first weight value or the second weight value is used to calculate the corresponding influence factor according to the linear relationship, which accurately reflects the influence degree when the environmental factor deviates from the standard growth requirement of the plant.

[0034] The preset plant recognition model is used to recognize plant morphological feature data such as plant height, leaf number, and leaf color, and calculate the corresponding element influence factor. This process includes the growth state of the plant in the consideration range of the base station influence degree. Different plant morphological features can directly reflect the health status of the plant and the degree of influence by external factors.

[0035] The weighting coefficient ​Reasonably allocate the weight of different factors in the base station influence degree calculation, so that the final base station influence degree can comprehensively and reasonably reflect the comprehensive influence of the base station on the main plants. At the same time, considering the sensitivity coefficient of the main plants to the base station transmitting power and the antenna height, the influence evaluation of the base station equipment parameters on the plant growth is further refined.

[0036] Further, the preset base station equipment adjustment strategy is:

[0037] S100, retrieve the latitude and longitude coordinate information and the altitude information corresponding to each base station from the server;

[0038] S200, according to the base station influence degree of the main plants corresponding to each base station, the meteorological data around the base station, the latitude and longitude coordinate information and the altitude information, based on the preset base station clustering strategy, the base stations are clustered and divided to form a plurality of base station clusters;

[0039] S300, according to the formed plurality of base station clusters, the adjacent other base station clusters corresponding to a base station cluster are determined, and each first outermost base station corresponding to the base station cluster and the adjacent other base station cluster and each outermost base station corresponding to the base station cluster are determined. Adjacent second outermost base station;

[0040] S400, according to the base station influence degree corresponding to each base station, the station influence degree difference value between each first peripheral base station corresponding to the base station cluster and the station influence degree difference value between each first outermost base station corresponding to the base station cluster and each second peripheral base station corresponding to the adjacent other base station cluster are calculated;

[0041] The calculation of the station influence degree difference value is as follows:

[0042]

[0043] In the formula, The station influence degree difference value is The historical average base station influence degree corresponding to the jth first outermost base station is The current base station influence degree corresponding to the jth first outermost base station is The total number of first peripheral base stations corresponding to a base station cluster is The total number of base stations corresponding to a base station cluster is The normalization coefficient of the station influence degree difference is The historical average base station influence degree of the first outermost base station is

[0044] The calculation of the station influence degree difference value is as follows:

[0045]

[0046] In the formula, is the station-outside influence degree difference value between a certain base station aggregation class and a certain other base station aggregation class adjacent thereto, is the total number of second peripheral base stations corresponding to the other base station aggregation class adjacent thereto;

[0047] S500, it is judged whether the station-influence degree difference value and the station-outside influence degree difference value both satisfy the corresponding preset difference threshold value, if yes, it is judged that the base station aggregation classes at this time are feasible division, and the corresponding multiple base station aggregation classes are output, otherwise, S200 is re-executed;

[0048] S600, according to the multiple base station aggregation classes output, a base station influence degree corresponding to a base station selected at random from each base station aggregation class is selected, and based on a preset influence degree division threshold value, a base station influence level corresponding to each base station aggregation class is judged, the base station influence level including a first level, a second level and a third level;

[0049] S700, according to the base station aggregation classes corresponding to the first level and the base station aggregation classes corresponding to the third level, based on a preset device adjustment table, a part of the base station devices corresponding to each base station in the base station aggregation classes corresponding to the first level are selected and added to each base station in the base station aggregation classes corresponding to the third level, and a corresponding base station device adjustment scheme is output.

[0050] Beneficial effects: In the present scheme, in steps S100 and S200, the latitude and longitude coordinate information and the altitude information of the base stations are called, and the base station influence degree and the meteorological data around the base stations are combined for clustering division. This multi-source data fusion method makes full use of the information contained in different types of data. Compared with single-dimensional data clustering, this method can more accurately group base stations with similar characteristics into a group, laying a foundation for subsequent targeted analysis and adjustment.

[0051] By calculating the station-influence degree difference value and the station-outside influence degree difference value, the change of the base station influence degree within the same base station aggregation class and between different base station aggregation classes can be accurately evaluated. The calculation of the station-influence degree difference value considers the historical average base station influence degree and the current base station influence degree of the first outermost base station, reflecting the fluctuation of the base station influence degree in the time dimension and the difference within the same aggregation class. The station-outside influence degree difference value measures the relative size of the influence degree between different base station aggregation classes. This accurate difference evaluation helps to find the abnormal situation of the base station influence degree and the imbalance between regions.

[0052] According to the base station aggregation class of different influence levels, device adjustment is performed by using a preset device adjustment table, and part of the devices in the first-level base station aggregation class are added to the base stations in the third-level base station aggregation class. In this way, the device resources of the base stations are optimally configured, and the device resources of the low-influence-level base stations are fully utilized to improve the influence of the high-influence-level base stations on plant growth. Through accurate device adjustment, on the one hand, the negative influence of the high-influence-level base stations on the surrounding plants can be reduced without a large increase in device investment, and the compatibility of the base stations with the environment is improved; on the other hand, the problem of resource waste and low efficiency caused by indiscriminate adjustment of all base stations is avoided, and the overall resource utilization efficiency and device adjustment effect are improved.

[0053] Further, the preset base station clustering strategy is:

[0054] S10, according to the base station influence degree of the main plants corresponding to each base station, the meteorological data around the base station, the latitude and longitude coordinate information and the altitude information, the base station similarity between each base station is calculated, and the base station with the smallest base station similarity corresponding to each base station is selected, and the selected base station and the corresponding base station are integrated as the first batch of base station set class; the number of base stations in the first batch of base station set class is greater than or equal to 2;

[0055] S20, a base station is randomly selected from the first batch of base station set class as the representative base station corresponding to the first batch of base station set class, the base station similarity between the representative base station and other remaining base stations, and the base station similarity between the representative base station and other representative base stations are called, and arranged in descending order of base station similarity, and selected from the first one, R base stations are selected as the related base stations corresponding to the representative base station;

[0056] S30, the number of each base station corresponding to the related base stations corresponding to each representative base station in the first batch of base station set class is counted, and the first number of each base station as a related base station is counted;

[0057] When the first number is greater than the preset number threshold, if the base station corresponding to the first number is other remaining base station, the other remaining base station is excluded from the related base station, and returns to the other remaining base station;

[0058] If the base station corresponding to the first number is other representative base station, the first batch of base station set class corresponding to the other representative base station is not merged with the first batch of base station set class corresponding to other representative base stations;

[0059] When the first number is less than or equal to the preset number threshold, if the base station corresponding to the first number is other remaining base station, the other remaining base station is added to the first batch of base station set class corresponding to the representative base station with the largest base station similarity, to form a new first batch of base station set class.

[0060] If the first quantity corresponds to other representative base stations, the first set of base station classes corresponding to other representative base stations is merged into the first set of base station classes corresponding to other representative base stations with the largest base station similarity, to form a new first set of base station classes.

[0061] S40, re-performs S20 until the other remaining base stations are completely added, judges whether the number of classes corresponding to the first set of base station classes at this time reaches the preset class threshold, if yes, the first set of base station classes at this time is the final base station aggregation class, if not, re-performs S20 until the number of classes reaches the preset class threshold.

[0062] Beneficial effect: A representative base station is randomly selected from the first set of base station classes, and then the base station similarity between the representative base station and other remaining base stations and other representative base stations is retrieved, and R base stations are selected as related base stations in order of similarity from large to small. This operation is like accurately screening closely related nodes from a complex base station relationship network. This screening method greatly improves the efficiency of base station relationship analysis and avoids blind integration, laying a foundation for subsequent formation of accurate base station aggregation classes. By focusing on the similarity between the representative base station and other base stations, the relationship between base stations with similar comprehensive characteristics can be highlighted. Base stations with too high similarity often have similar influence patterns on surrounding plants, meteorological environments, and geographical location characteristics. Grouping these base stations can strengthen the consistency of the clustering results and help develop uniform management and adjustment strategies for such characteristics.

[0063] In the clustering process, according to the comparison result of the first quantity of base stations as related base stations and the preset quantity threshold, the base stations are reasonably added or removed. When the first quantity is greater than the preset quantity threshold and the base station is other remaining base station, it is removed from the related base station, which avoids the mixing of too many irrelevant base stations into the same set class and prevents resource waste. When the first quantity is less than or equal to the preset quantity threshold, the remaining base stations are added to the set class corresponding to the appropriate representative base station, which makes full use of dispersed base station resources and optimizes the composition of the base station set class.

[0064] Further, the calculation formula corresponding to the base station similarity between each base station is:

[0065]

[0066] In the formula, is the base station similarity between base station g and base station l, , , is the corresponding weight coefficient, respectively, are historical average temperature data, historical average humidity data and historical average wind speed data corresponding to the base station g; , , is a corresponding weight value, respectively, are longitude and latitude coordinate information corresponding to the base station g, respectively, are altitude information corresponding to the base station g and the base station l.

[0067] Beneficial effect: By combining the geographical similarity, meteorological similarity and base station influence degree difference related items in a weighted manner, the key dimensions affecting the base station similarity are comprehensively covered. The geographical similarity considers the differences in longitude, latitude and altitude between base stations, and accurately depicts the proximity of their geographical positions.

[0068] Further, the value of R is associated with the number of other remaining base stations and representative base stations at this time and the number of categories corresponding to the first batch of base station set classes at this time;

[0069] The corresponding association formula is:

[0070]

[0071] In the formula, is the value of R corresponding to the t-th clustering cycle, is the number of other remaining base stations and representative base stations at the t-th clustering cycle, is the number of categories corresponding to the first batch of base station set classes at this time corresponding to the t-th clustering cycle, is the attenuation coefficient corresponding to the t-th clustering cycle, and the size of the attenuation coefficient is negatively related to the number of clustering cycles.

[0072] Beneficial effect: By associating the value of R with the number of remaining base stations and representative base stations, the number of categories of the first batch of base station set classes and the attenuation coefficient, the number of related base stations selected each time can be dynamically adjusted according to the actual situation during the clustering cycle. As the number of clustering cycles increases, the number of remaining base stations and the number of categories are constantly changing, and this formula can flexibly adapt to these changes. As the clustering proceeds, the number of remaining base stations decreases, the categories gradually stabilize, and the attenuation coefficient becomes smaller, so that the value of R decreases accordingly, avoiding excessive association of unrelated base stations and improving the accuracy of clustering. This dynamic adjustment mechanism makes the clustering process more in line with the actual situation, improving the adaptability of the clustering algorithm to base station data of different sizes and distributions.

[0073] By dynamically adjusting the R value, base stations in similar geographical environments and with similar influence degrees can be more accurately classified into a category, clustering errors caused by blindly expanding the clustering range in the early stage or insufficient adjustment in the later stage are avoided, and finally the quality of the clustering result is improved, providing a more reliable basis for subsequent base station equipment adjustment, resource allocation and other decisions.

[0074] The application further provides an intelligent terminal automatic survey data analysis and processing method based on deep learning. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 The figure is a logic block diagram of the intelligent terminal automatic survey data analysis and processing system based on deep learning in embodiment one of the application. DETAILED DESCRIPTION

[0076] The following will be further described in detail through specific embodiments:

[0077] Embodiment one

[0078] The intelligent terminal automatic survey data analysis and processing system based on deep learning basically comprises a server, an intelligent terminal for surveying around each base station site, and a monitoring device for collecting base station basic information corresponding to each base station. Figure 1

[0079] The server is used for receiving base station information corresponding to each base station collected by the monitoring device; the base station information comprises base station basic information and plant growth video information around the base station; in this embodiment, the base station information further comprises animal growth video information around the base station, and the influence of the base station on the animals around the base station can be evaluated to reduce the influence on the animals around the base station.

[0080] The intelligent terminal comprises:

[0081] A data collection module is used for collecting surrounding environment data within a preset range installed on each base station, and the surrounding environment data comprises meteorological data, temperature and humidity data, soil pH data, light intensity data and gas concentration data around the base station.

[0082] A data retrieval module is used for retrieving base station basic information and plant growth video information around the base station corresponding to each base station from the server.

[0083] ​The plant recognition module is configured to recognize the plant type and the current plant growth state of the main plant corresponding to the base station according to the plant growth video information around the base station, and to retrieve the plant standard growth demand information corresponding to the plant type in the current plant growth state from the server; the plant standard growth demand information includes the temperature range, the humidity range, the light intensity range, the soil pH range, the gas concentration range, and the water demand parameter required by the plant type in the current growth state.

[0084] The processing module is configured to determine the base station influence degree of the main plant corresponding to each base station based on the surrounding environment data corresponding to each base station, the plant standard growth demand information, the base station basic information, and the plant growth video information around the base station, and based on a preset base station influence degree determination strategy.

[0085] The preset base station influence degree determination strategy is:

[0086] According to the obtained base station surrounding meteorological data, temperature and humidity data, soil pH data, light intensity data, gas concentration data, and plant standard growth demand information, the element influence factor corresponding to each data is calculated based on a preset element influence factor calculation formula.

[0087] The preset element influence factor calculation formula is:

[0088]

[0089] In the formula, is the element influence factor corresponding to data i, n is the total number of surrounding environment data, is the current data value corresponding to data i, is the standard minimum data value corresponding to data i, is the standard maximum data value corresponding to data i, , is the first weight value and the second weight value corresponding to data i;

[0090] According to the plant standard growth demand information and the plant growth video information around the base station, the plant morphological feature data corresponding to the main plant is recognized based on a preset plant recognition model, and the element influence factor corresponding to each plant morphological feature data is calculated; the plant morphological feature data includes plant height, leaf number, and leaf color.

[0091] In the embodiment, the plant recognition model adopts an existing neural network model, for example, a convolutional neural network model (CNN) or a recurrent neural network model (RNN). Taking the convolutional neural network model as an example, a convolutional neural network (CNN) architecture based on deep learning is adopted, including an input layer, multiple convolutional layers, a pooling layer, a fully connected layer, and an output layer; specifically, ResNet50 or VGG16 can be used as a basic network, and a custom classification layer is added at the end thereof; image data of different plant types in different growth stages (seedling stage, growth stage, and mature stage) is collected, including overall plant images and leaf close-up images; after the collected image data is labeled, the image data is input into the model for iterative training, a cross-entropy loss function is adopted, training parameters (for example, the learning rate is set to 0.001, the batch size is 32, and the number of training iterations is not less than 100 times) are set, and the model converges until the model converges; the model output includes plant type recognition results, current growth state evaluation results (healthy, sub-healthy, and unhealthy), and quantitative values of plant morphological feature data (plant height, leaf number, and leaf color).

[0092] According to the element influence factor corresponding to each data and the base station basic information, the base station influence degree of the main plant corresponding to the base station is calculated based on a preset base station influence degree calculation formula.

[0093] The preset base station influence degree calculation formula is:

[0094]

[0095] In the formula, is the base station influence degree of the main plant corresponding to the base station, is a corresponding weighting coefficient, is an element influence factor corresponding to the plant height, is an element influence factor corresponding to the leaf number, is an element influence factor corresponding to the leaf color, , are a sensitive coefficient of the main plant to the base station transmit power and a sensitive coefficient of the main plant to the antenna height, respectively, is the current base station transmit power corresponding to the base station, is the maximum transmit power corresponding to the base station of the same type, is the current antenna height corresponding to the base station, is the maximum antenna height corresponding to the base station of the same type.

[0096] The analysis module is configured to adjust and optimize the base station equipment corresponding to each base station based on a preset base station equipment adjustment strategy according to the base station influence degree of the main plant corresponding to each base station, and output a corresponding base station equipment adjustment scheme.

[0097] The preset base station device adjustment strategy is:

[0098] S100, retrieve the longitude and latitude coordinate information and the altitude information corresponding to each base station from the server;

[0099] S200, according to the base station influence degree of the main plant corresponding to each base station, the meteorological data around the base station, the longitude and latitude coordinate information and the altitude information, based on the preset base station clustering strategy, the base stations are clustered and divided to form a plurality of base station aggregation classes;

[0100] The preset base station clustering strategy is:

[0101] S10, according to the base station influence degree of the main plant corresponding to each base station, the meteorological data around the base station, the longitude and latitude coordinate information and the altitude information, the base station similarity between each base station is calculated, and the base station with the smallest base station similarity corresponding to each base station is selected, and the selected base station and the corresponding base station are integrated as the first batch of base station set class; The number of base stations in the first batch of base station set class is greater than or equal to 2;

[0102] S20, randomly select a base station from the first batch of base station set class as the representative base station corresponding to the corresponding first batch of base station set class, retrieve the base station similarity between the representative base station and the other remaining base stations, and the base station similarity between the representative base station and other representative base stations, and arrange them in descending order of base station similarity, and select R base stations as the relevant base stations corresponding to the representative base station from the first one;

[0103] The calculation formula corresponding to the calculation of the base station similarity between each base station is:

[0104]

[0105] In the formula, The base station similarity between base station g and base station l, , , The corresponding weighting coefficient, The historical average temperature data, the historical average humidity data and the historical average wind speed data corresponding to base station g respectively; , , The corresponding weight value, The longitude and latitude coordinate information corresponding to base station g respectively, The altitude information corresponding to base station g and base station l respectively.

[0106] The value of R is related to the number of other remaining base stations and representative base stations at this time and the number of categories corresponding to the first batch of base station set class at this time;

[0107] The association formula corresponding to the specific association is:

[0108]

[0109] In the formula, Rtis the value of R corresponding to the tth clustering cycle, Ntis the number of other remaining base stations and the number of representative base stations corresponding to the tth clustering cycle, Ktis the number of categories corresponding to the first batch of base station set classes at this time corresponding to the tth clustering cycle, αtis the attenuation coefficient corresponding to the tth clustering cycle, and the size of the attenuation coefficient is negatively related to the number of clustering cycles.

[0110] S30, the number of each base station corresponding to the relevant base station corresponding to each representative base station in the first batch of base station set classes is counted, and the first number of each base station as a relevant base station is counted;

[0111] When the first number is greater than the preset number threshold, if the base station corresponding to the first number is other remaining base stations, the other remaining base stations are removed from the relevant base stations, and the other remaining base stations are returned;

[0112] If the base station corresponding to the first number is other representative base stations, the first batch of base station set classes corresponding to the other representative base stations are not merged with the first batch of base station set classes corresponding to other representative base stations;

[0113] When the first number is less than or equal to the preset number threshold, if the base station corresponding to the first number is other remaining base stations, the other remaining base stations are added to the first batch of base station set classes corresponding to the representative base station with the largest base station similarity, to form a new first batch of base station set classes;

[0114] If the base station corresponding to the first number is other representative base stations, the first batch of base station set classes corresponding to other representative base stations are merged into the first batch of base station set classes corresponding to other representative base stations with the largest base station similarity, to form a new first batch of base station set classes;

[0115] S40, S20 is re-executed until the other remaining base stations are completely added, whether the category number corresponding to the first batch of base station set classes at this time reaches the preset category threshold is judged, if yes, the first batch of base station set classes at this time is the final base station aggregation class, if not, S20 is re-executed until the category number reaches the preset category threshold.

[0116] S300, according to the formed plurality of base station aggregation classes, determining the adjacent other base station aggregation classes corresponding to a certain base station aggregation class, and determining the first outermost base stations corresponding to the base station aggregation class and the second outermost base stations adjacent to the first outermost base stations corresponding to the base station aggregation class and the adjacent other base station aggregation classes;

[0117] S400, according to the base station influence degree corresponding to each base station, calculating the station-influence-degree difference value between the first outermost base stations corresponding to the base station aggregation class and the station-outside-influence-degree difference value between the first outermost base stations corresponding to the base station aggregation class and the second outermost base stations corresponding to the adjacent other base station aggregation classes;

[0118] The calculation of the station-influence-degree difference value is as follows:

[0119]

[0120] In the formula, is the station-influence-degree difference value, is the historical average base station influence degree corresponding to the jth first outermost base station, is the current base station influence degree corresponding to the jth first outermost base station, is the total number of first outermost base stations corresponding to a certain base station aggregation class, is the total number of base stations corresponding to a certain base station aggregation class, is the normalization coefficient of the station-influence-degree difference, is the historical average base station influence degree of the first outermost base station (i.e., the average influence degree of the base station on plants in the past period of time).

[0121] The calculation of the station-outside-influence-degree difference value is as follows:

[0122]

[0123] In the formula, is the station-outside-influence-degree difference value between a certain base station aggregation class and a certain other base station aggregation class adjacent thereto, is the total number of second outermost base stations corresponding to the adjacent other base station aggregation class;

[0124] S500, judging whether the station-influence-degree difference value and the station-outside-influence-degree difference value both satisfy the corresponding preset difference threshold value, if yes, judging that each base station aggregation class is a feasible division at this time, outputting the corresponding plurality of base station aggregation classes, otherwise, re-S200;

[0125] S600, according to the output of the plurality of base station aggregation classes, randomly selecting a base station corresponding to the base station influence degree from each base station aggregation class, judging the base station influence level corresponding to each base station aggregation class based on the preset influence degree division threshold, the base station influence level includes first level, second level and third level;

[0126] S700, according to the first level corresponding to the base station aggregation class and the third level corresponding to the base station aggregation class, based on the preset device adjustment table, the base station device corresponding to each base station in the first level corresponding to the base station aggregation class is selected and added to each base station in the third level corresponding to the base station aggregation class, and the corresponding base station device adjustment scheme is output.

[0127] In this embodiment, a display module is also included for visualizing the base station device adjustment scheme and synchronously sending it to the other intelligent terminal closest to the device to be adjusted corresponding to the base station device adjustment scheme, and informing the operator corresponding to the intelligent terminal to make corresponding device adjustment. The base station device adjustment scheme is presented in an intuitive graphical interface, and for each base station to be adjusted, different colors are used to identify its current state and target state after adjustment. For example, the base station device whose transmission power needs to be adjusted is represented by a red flashing icon, and the current transmission power value and the target power value after adjustment are listed in detail beside it. For the base station whose antenna direction needs to be changed, the angle change before and after the antenna adjustment is shown by a dynamic arrow, and the specific reason for the adjustment is explained in words below, such as signal shielding due to surrounding plant growth, which requires adjusting the antenna direction to optimize signal coverage.

[0128] First, the latitude and longitude coordinate information of the device to be adjusted is obtained, and the position information of the surrounding intelligent terminals is compared through the built-in map algorithm to accurately calculate the distance between each intelligent terminal and the device to be adjusted. Then, the intelligent terminals closest to the device to be adjusted are selected in order from near to far. For example, when there are multiple intelligent terminals within a certain range around the base station, the display module quickly calculates and determines the three intelligent terminals closest to the base station device to be adjusted as the information receiving target.

[0129] The embodiment also discloses an intelligent terminal automatic survey data analysis and processing method based on deep learning, which uses the intelligent terminal automatic survey data analysis and processing system based on deep learning.

[0130] The above-mentioned are only embodiments of the present application, and the common knowledge of the specific structure and characteristics in the scheme is described too much, the ordinary skilled in the art knows all the ordinary technical knowledge in the field of the present application before the application date or the priority date, can know all the prior art in the field, and has the ability to apply the conventional experimental means before the date, the ordinary skilled in the art can perfect and implement the scheme under the enlightenment given by the present application combined with their own ability, some typical known structure or known method should not become the obstacle for the ordinary skilled in the art to implement the present application. It should be pointed out that for those skilled in the art, without departing from the structure of the present application, a number of modifications and improvements can also be made, which should also be considered as the protection scope of the present application, which will not affect the effect and practicality of the patent. The protection scope of the present application should be subject to the content of its claims, and the specific implementation mode and the like in the specification can be used to explain the content of the claims.

Claims

1. A deep learning-based intelligent terminal automated survey data analysis and processing system, characterized in that: The service end, an intelligent terminal for surveying around each base station site, and a monitoring device for collecting base station basic information corresponding to each base station are included. The service end is configured to receive the base station information corresponding to each base station collected by the monitoring device. The base station information includes base station basic information and base station surrounding plant growth video information. The intelligent terminal includes: A data collection module configured to collect surrounding environment data within a preset range installed around each base station, including base station surrounding meteorological data, temperature and humidity data, soil pH data, light intensity data, and gas concentration data. A data retrieval module configured to retrieve base station basic information and base station surrounding plant growth video information corresponding to each base station from the service end. A plant identification module configured to identify the plant category and current plant growth state of the main plant corresponding to the base station surrounding area based on the retrieved base station surrounding plant growth video information, and retrieve plant standard growth requirement information corresponding to the plant category in the current plant growth state from the service end; the plant standard growth requirement information includes the temperature range, humidity range, light intensity range, soil pH range, gas concentration range, and water requirement parameters required by the plant category in the current growth state. A processing module configured to determine the base station influence degree of the main plant corresponding to each base station based on the surrounding environment data, plant standard growth requirement information, base station basic information, and base station surrounding plant growth video information corresponding to each base station, based on a preset base station influence degree determination strategy. An analysis module configured to adjust and optimize the base station equipment corresponding to each base station based on a preset base station equipment adjustment strategy, according to the base station influence degree of the main plant corresponding to each base station, and output the corresponding base station equipment adjustment scheme. 2.The deep learning based intelligent terminal automated survey data analysis and processing system of claim 1, wherein: The preset base station influence degree determination strategy is: Based on the obtained base station surrounding meteorological data, temperature and humidity data, soil pH data, light intensity data, gas concentration data, and plant standard growth requirement information, the element influence factor corresponding to each data is calculated based on a preset element influence factor calculation formula. The preset element influence factor calculation formula is: In the formula, is an element influence factor corresponding to data i, n is the total number of data corresponding to the surrounding environment data, is a current data value corresponding to data i, is a standard minimum data value corresponding to data i, is a standard maximum data value corresponding to data i, , is a first weight value and a second weight value corresponding to data i; Based on the plant standard growth requirement information and the base station surrounding plant growth video information, the plant morphological feature data corresponding to the main plant is identified based on a preset plant identification model, and the element influence factor corresponding to each plant morphological feature data is calculated; the plant morphological feature data includes plant height, leaf number, and leaf color. The plant identification model uses a convolutional neural network model including an input layer, multiple convolutional layers, a pooling layer, a fully connected layer, and an output layer; ResNet50 or VGG16 is used as a basic network, and a custom classification layer is added at the end; after collecting and labeling plant image data, the model is input for iterative training, a cross-entropy loss function is used, training parameters are set, and the model converges until the model converges. According to the element influence factor corresponding to each data and the base station basic information, a base station influence degree of the main plant corresponding to the base station is calculated based on a preset base station influence degree calculation formula; The preset base station influence degree calculation formula is: In the formula, a base station influence degree of a main plant corresponding to the base station, a corresponding weighting coefficient, an element influence factor corresponding to a plant height, an element influence factor corresponding to a leaf number, an element influence factor corresponding to a leaf color, , respectively a main plant sensitive coefficient to a base station transmit power and a sensitive coefficient to an antenna height, a current base station transmit power corresponding to the base station, a maximum transmit power corresponding to a same type base station, a current antenna height corresponding to the base station, a maximum antenna height corresponding to a same type base station. 3.The deep learning based intelligent terminal automated survey data analysis and processing system of claim 2, wherein: The preset base station device adjustment strategy is: S100, the latitude and longitude coordinate information and the altitude information corresponding to each base station are called from the server; S200, according to the base station influence degree of the main plant corresponding to each base station, the meteorological data around the base station, the latitude and longitude coordinate information and the altitude information, each base station is clustered and divided based on a preset base station clustering strategy, and a plurality of base station aggregation classes are formed; S300, according to the plurality of base station aggregation classes formed, the adjacent other base station aggregation classes corresponding to a certain base station aggregation class are determined, and each first outermost base station corresponding to the base station aggregation class and each second outermost base station adjacent to the first outermost base station corresponding to the base station aggregation class and the adjacent other base station aggregation classes are determined; S400, according to the base station influence degree corresponding to each base station, the station influence degree difference value between each first peripheral base station corresponding to the base station aggregation class and the station influence degree difference value between each first outermost base station corresponding to the base station aggregation class and each second peripheral base station corresponding to the adjacent other base station aggregation class are calculated; The calculation of the station influence degree difference value is as follows: In the formula, is an in-station influence degree difference value, is a historical average base station influence degree corresponding to the jth first peripheral base station, is a current base station influence degree corresponding to the jth first peripheral base station, is a total number of first peripheral base stations corresponding to a certain base station aggregation class, is a total number of base stations corresponding to a certain base station aggregation class; is a normalization coefficient of the in-station influence degree difference, is a historical average base station influence degree of the first peripheral base station; The calculation of the station influence degree difference value is as follows: In the formula, is a station-out influence difference value between a certain base station aggregation class and a certain other base station aggregation class adjacent thereto, is a second total number of peripheral base stations corresponding to the other base station aggregation class adjacent thereto. S500, it is judged whether the station influence degree difference value and the station influence degree difference value meet the corresponding preset difference threshold value, if yes, it is judged that each base station aggregation class is feasible at this time, and the corresponding plurality of base station aggregation classes are output, otherwise, S200 is reselected; S600, according to the plurality of base station aggregation classes output, the base station influence degree corresponding to a base station selected at random from each base station aggregation class is selected, and the base station influence level corresponding to each base station aggregation class is judged based on a preset influence degree division threshold value, the base station influence level includes a first level, a second level and a third level; S700, according to the base station aggregation class corresponding to the first level and the base station aggregation class corresponding to the third level, the base station device corresponding to each base station in the base station aggregation class corresponding to the first level is selected and added to each base station in the base station aggregation class corresponding to the third level based on a preset device adjustment table, and a corresponding base station device adjustment scheme is output. 4.The deep learning based intelligent terminal automated survey data analysis and processing system of claim 3, wherein: The preset base station clustering strategy is: S10, according to the base station influence degree of the main plant corresponding to each base station, the meteorological data around the base station, the latitude and longitude coordinate information and the altitude information, the base station similarity between each base station is calculated, and the base station with the smallest base station similarity corresponding to each base station is selected and integrated as a first batch of base station set class; The base station of the first batch of base station set class is greater than or equal to 2; S20, randomly selecting one base station from the first batch of base station set classes as the representative base station corresponding to the corresponding first batch of base station set classes, calling the base station similarity between the representative base station and other remaining base stations, and the base station similarity between the representative base station and other representative base stations, and arranging them in descending order of base station similarity, and selecting R base stations as the relevant base stations corresponding to the representative base station from the first one; S30, counting the number of each base station corresponding to the relevant base stations corresponding to each representative base station in the first batch of base station set classes, and counting the first number of each base station as a relevant base station; When the first number is greater than the preset number threshold, if the base station corresponding to the first number is other remaining base stations, the other remaining base stations are excluded from the relevant base stations, and the other remaining base stations are returned; If the base station corresponding to the first number is other representative base stations, the first batch of base station set classes corresponding to the other representative base stations are not merged with the first batch of base station set classes corresponding to other representative base stations; When the first number is less than or equal to the preset number threshold, if the base station corresponding to the first number is other remaining base stations, the other remaining base stations are added to the first batch of base station set classes corresponding to the representative base station with the largest base station similarity, forming a new first batch of base station set classes; If the base station corresponding to the first number is other representative base stations, the first batch of base station set classes corresponding to other representative base stations is merged into the first batch of base station set classes corresponding to other representative base stations with the largest base station similarity, forming a new first batch of base station set classes; S40, re-executing S20 until the other remaining base stations are completely added, judging whether the number of classes corresponding to the corresponding first batch of base station set classes reaches the preset class threshold, if yes, the corresponding first batch of base station set classes at this time is the corresponding final base station aggregation class, if not, re-executing S20 until the number of classes reaches the preset class threshold. 5.The deep learning based intelligent terminal automated survey data analysis and processing system of claim 4, wherein: The calculation formula for calculating the base station similarity between each base station is: In the formula, is a base station similarity between the base station g and the base station l, , , is a corresponding weighting coefficient, are historical average temperature data, historical average humidity data and historical average wind speed data corresponding to the base station g respectively; , , is a corresponding weight value, are longitude and latitude coordinate information corresponding to the base station g respectively, are altitude information corresponding to the base station g and the base station l respectively. 6.The deep learning based intelligent terminal automated survey data analysis and processing system of claim 5, wherein: The value of R is related to the number of other remaining base stations and representative base stations at this time and the number of classes corresponding to the first batch of base station set classes at this time; The specific association formula is: In the formula, Rtis the value of R corresponding to the tth clustering cycle, Ntis the number of other remaining base stations and the number of representative base stations corresponding to the tth clustering cycle, Ktis the number of categories corresponding to the first batch of base station set classes at this time corresponding to the tth clustering cycle, αtis the attenuation coefficient corresponding to the tth clustering cycle, and the size of the attenuation coefficient is negatively related to the number of clustering cycles.

7. The method for automatic analysis and processing of survey data based on intelligent terminal of deep learning, characterized in that: The deep learning-based intelligent terminal automatic survey data analysis and processing system of any one of claims 1 to 6 is used.

Images