KR102445733B1 - Method, server and system for optimizing system network in tactical environment - Google Patents

Abstract

According to the present invention, a system network optimization method through DNN-based mobile base station cell power control in a tactical environment including a plurality of mobile base stations and a plurality of terminals comprises the steps of: extracting learning data and test data for learning and testing of a DNN; and determining a DNN structure for setting optimal cell power with respect to each of the plurality of mobile base stations based on the extracted learning data and test data.

Description

System network optimization method, server and system in tactical environment

The present invention relates to a system network optimization method, server and system through DNN-based mobile base station cell power control in a tactical environment including a plurality of mobile base stations and a plurality of terminals. That is, the present invention is a method for optimizing a system network in a tactical environment that can minimize frequency interference between mobile base stations by determining and applying the optimal cell power of each base station based on DNN in a platoon-level or lower wireless combat network, a server and It's about the system.

1 is a diagram showing an example of operation of a mobile base station. 2 is a view for explaining the operation form of the existing tactical mobile base station.

Commercial base stations are installed at intervals of about 500m, and interference between base stations can be minimized by using the cell plan tool. Unlike the commercial operation method, the tactical mobile base station provides a service in real time by connecting to the terminal while the base station is activated. In such a tactical environment, frequency interference between base stations inevitably occurs if cell power between base stations is not adjusted in a situation in which a plurality of base stations are concentrated. In order to solve this problem, it is necessary to minimize interference and maximize the channel capacity of the system by adjusting the cell power of the base station in real time according to the location of each base station and the terminal.

On the other hand, the currently operated base station determines the optimal base station cell power by collecting and analyzing surrounding environment information by using a terminal module in the tactical mobile base station vehicle after the unit is moved. In the prior art, after performing sensing of a frequency and signal strength of a neighboring base station, FA selection and output are set, and the base station is turned on. The execution time for this series of processes is usually about 2 minutes, and at this time, the base station service becomes impossible. This is an unsuitable method in a tactical environment that requires real-time operational operation.

The problem to be solved by the present invention is a system network optimization method in a tactical environment that can minimize frequency interference between mobile base stations by determining and applying the optimal cell power of each base station based on DNN in a platoon-level or lower wireless combat network , servers and systems.

However, the problems to be solved by the present invention are not limited to the problems described above, and other problems may exist.

The method for optimizing a system network through DNN-based mobile base station cell power control in a tactical environment including a plurality of mobile base stations and a plurality of terminals according to the first aspect of the present invention for solving the above problems is the learning and testing of the DNN extracting learning data and test data for and determining a DNN structure for setting optimal cell power for each of the plurality of mobile base stations, based on the extracted learning data and test data.

In some embodiments of the present invention, the extracting of the training data and test data for the learning and testing of the DNN may include: randomly arranging the plurality of mobile base stations and the plurality of terminals; calculating RSSI and SINR for the plurality of terminals as terminal information; determining a cell power of a mobile base station capable of maximizing a system channel capacity based on the calculated terminal information and cell power set in the mobile base station; and extracting the learning data and the test data based on the determined cell power and terminal information.

In some embodiments of the present invention, the determining of the cell power of the mobile base station includes: calculating a downlink channel capacity for a plurality of terminals based on the terminal information; and determining the cell power of the mobile base station capable of maximizing the downlink channel capacity for each terminal.

In some embodiments of the present invention, the step of extracting the training data and the test data based on the determined cell power output value and the terminal information includes the SINR of each of the plurality of terminals and a plurality of interference values received from adjacent mobile base stations. It is possible to extract training data and test data using an input parameter of the DNN and cell power of the mobile station as an output parameter.

In some embodiments of the present invention, the determining of the DNN structure for setting the optimal cell power for each of the plurality of mobile base stations includes: performing learning of the plurality of DNN structures based on the learning data; and selecting a DNN structure that minimizes a mean squared error (MSE) from among the plurality of learned DNN structures based on the test data.

According to some embodiments of the present invention, the mobile base station periodically collects terminal information including RSSI and SINR and provides it to a mobile command vehicle; and deriving cell power for each mobile base station based on the determined DNN structure in the mobile command vehicle.

In addition, the system network optimization server in a tactical environment including a plurality of mobile base stations and a plurality of terminals according to the second aspect of the present invention is a tactical environment system through cell power adjustment for the plurality of mobile base stations based on the determined DNN structure. As the program for optimizing the memory is stored and the program stored in the memory is executed, when the mobile base station receives the terminal information including the RSSI and SINR collected by the mobile station according to a preset period, the DNN structure determined in advance is learned in advance. and a processor for deriving a cell power for each mobile base station based on the.

In addition, the DNN learning system for system network optimization through DNN-based mobile base station cell power control in a tactical environment including a plurality of mobile base stations and a plurality of terminals according to the third aspect of the present invention is provided to the plurality of mobile base stations. In order to optimize the tactical environment system by controlling the cell power for the Accordingly, a DNN structure for extracting learning data and test data for learning and testing of the DNN and setting an optimal cell power for each of the plurality of mobile base stations based on the extracted learning data and test data It includes a processor for determining.

In some embodiments of the present invention, the processor randomly arranges the plurality of mobile base stations and the plurality of terminals, calculates RSSI and SINR for the plurality of terminals as terminal information, and calculates the terminal information and the mobile After determining the cell power of the mobile base station capable of maximizing the system channel capacity based on the cell power set in the base station, the learning data and the test data may be extracted based on the determined cell power and terminal information.

In some embodiments of the present invention, the processor may calculate the downlink channel capacity for each of a plurality of terminals based on the terminal information, and determine the cell power of the mobile base station capable of maximizing the downlink channel capacity for each terminal. have.

In some embodiments of the present invention, the processor uses the SINR of each of the plurality of terminals and the plurality of interference values received from the adjacent mobile base station as an input parameter of the DNN, and training data and test using the cell power of the mobile base station as an output parameter. data can be extracted.

In some embodiments of the present invention, the processor performs learning for a plurality of DNN structures based on the learning data, and minimizes a mean squared error (MSE) among the plurality of learned DNN structures based on the test data. DNN structure can be selected.

A computer program according to another aspect of the present invention for solving the above problems is combined with a computer that is hardware to execute a system network optimization method through DNN-based mobile base station cell power control in the tactical environment, and a computer readable record stored on the medium.

Other specific details of the invention are included in the detailed description and drawings.

One embodiment of the present invention described above is to improve the operational problems of the current TICN (Tactical Information Communication Network) powered mobile base station, it is possible to improve the operability and usability. In particular, system optimization is possible by dramatically reducing processing time by applying DNN in a tactical environment that operates in real time. This is a core technology element that will be reflected in future TICN performance improvement or TICN 2.0.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

The accompanying drawings below are provided to help understanding of the present embodiment, and provide embodiments together with detailed description. However, the technical features of the present embodiment are not limited to specific drawings, and features disclosed in each drawing may be combined with each other to constitute a new embodiment.
1 is a diagram showing an example of operation of a mobile base station.
2 is a view for explaining the operation form of the existing tactical mobile base station.
3 is a diagram illustrating an example of a small unit level operation method applied to an embodiment of the present invention.
4 is a flowchart of a system optimization method in a tactical environment according to an embodiment of the present invention.
5 is a diagram for explaining in more detail a method for optimizing a system in a tactical environment according to an embodiment of the present invention.
6 is a diagram for explaining the proposed DNN structure in an embodiment of the present invention.
7 is a diagram for explaining the contents of calculating the MSE for each base station.
8 is a block diagram of a system network optimization system and server in a tactical environment according to an embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully inform those skilled in the art of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. In this specification, the singular also includes the plural, unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Therefore, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless clearly specifically defined.

Hereinafter, a method for optimizing a system network in a tactical environment according to the present invention will be described in detail with reference to the accompanying drawings.

3 is a diagram illustrating an example of a small unit level operation method applied to an embodiment of the present invention.

Referring to FIG. 3, a platoon is usually composed of four squads, and there are a main squad including a platoon leader, and squads 1, 2 and 3. As shown in Fig. 3, it is assumed that the mobile command vehicle is the main unit, and the manpack-type base station is the 1st, 2nd, and 3rd units. On the other hand, the server 200 according to an embodiment of the present invention may be operated by being mounted on a mobile command vehicle.

4 is a flowchart of a system optimization method in a tactical environment according to an embodiment of the present invention. 5 is a diagram for explaining in more detail a method for optimizing a system in a tactical environment according to an embodiment of the present invention.

The method for optimizing a system network in a tactical environment according to an embodiment of the present invention extracts learning data and test data for learning and testing a deep neural network (DNN) (S110).

Specifically, in order to extract the learning data and the test data, first, a plurality of mobile base stations and a plurality of terminals are randomly arranged (S111). As an example, 4 EAs of mobile base stations and 12 EAs of terminals may be randomly arranged based on MATLAB.

Next, the cell power of the current mobile base station is set (S112), and RSSI and SINR for a plurality of terminals are calculated as terminal information (S113).

Next, the system channel capacity is calculated based on the calculated terminal information and the cell power set in the mobile station (S114). At this time, the system channel capacity is calculated through the process of calculating the downlink channel capacity for each terminal based on the terminal information (Equation 1) and determining the cell power of the mobile base station capable of maximizing the downlink channel capacity for each terminal. can be calculated (Equation 2).

[Equation 1] Base station power to maximize DL (Downlink) channel capacity

[Equation 2] DL channel capacity per terminal

Meanwhile, the detailed system parameters applied for the MATLAB simulation are shown in Table 1.

division The details note system parameters Pathloss Model Okumura-hata Large Scale Fading 10dB Freq. 2.○GHz BW 10 MHz thermal noise -174dBm/Hz number of base stations 4EA number of terminals 12EA mobile base station maximum output 43dBm Output range 1-43dBm *Output change unit: 1dB operating height 4m terminal maximum output 23dBm operating height 1.5m resource allocation 1 MHz

Next, it is checked whether the calculated system channel capacity is the maximum (S115), and when the corresponding condition is satisfied, the cell power of the mobile station is determined under the condition of maximizing the system channel capacity (S116). If the calculated system channel capacity is not the maximum, steps after S112 are repeated and performed. After extracting the training data and test data based on the cell power and terminal information determined through MATLAB simulation in this way, the next extracted Based on the training data and the test data, a DNN structure for setting the optimal cell power for each of the plurality of mobile base stations is determined (S120).

6 is a diagram for explaining the proposed DNN structure in an embodiment of the present invention. 7 is a diagram for explaining the contents of calculating the MSE for each base station.

As shown in FIG. 6 , the DNN structure proposed in an embodiment of the present invention consists of a fully connected layer, and ReLU is applied as an activation function.

Each terminal can know the SINR and a plurality of interference values (Interference, for example, RSSI received from three adjacent base stations) received from the adjacent mobile base station. The number of is 48. The output parameter is the cell power of each mobile station.

First, learning is performed on a plurality of DNN structures based on the training data (S121). In this learning process, learning is performed by setting terminal information, which is an input parameter of the training data, as an input parameter of each DNN structure, and setting the cell power of the mobile station as an output parameter.

Next, based on the test data, a DNN structure that minimizes the mean squared error (MSE) from among the plurality of learned DNN structures is selected ( S122 ). In step S122, the terminal information, which is an input parameter of the test data, is set as an input parameter of the learned DNN structure, cell power of the mobile station is set as an output parameter, and MSEs for a plurality of DNN structures are respectively calculated. Then, as a result of the calculation, the DNN structure with the smallest MSE is selected and determined.

Table 2 below shows the results of the MSE cost function based on the number of layers and the number of neural networks. In this example, the 96-96-48-48 DNN structure with the smallest cost function was finally selected.

N of neurons per layer MSE BS1 BS2 BS3 BS4 BS5 96-48 10.280 12.836 10.205 11.862 11.296 144-96 9.23 10.130 9.146 9.251 9.440 24-24-12-12 14.380 13.251 11.707 12.404 12.936 96-96-48-48 7.611 8.583 7.118 7.769 7.770 144-144-96-96 7.671 8.581 7.118 7.719 7.772 24-24-24-24-12-12-12-12-12 7.789 8.600 7.203 7.762 7.839 48-48-48-48-24-24-24-24 7.692 8.623 7.173 7.731 7.805 96-96-96-96-48-48-48-48 7.612 8.573 7.122 7.770 7.772 144-144-144-144-96-96-96-96 7.612 8.588 7.139 7.803 7.788

Table 3 below shows the results of performance comparison based on the selected DNN structure (96-96-48-48). At this time, as a performance comparison method, k=5 was applied as a k-fold cross validation method for comparative analysis.

division Testing Data Set Avg.System Channel Capacity Performance Ratio Avg. Processing Time Base station output calculation model based on Full Searching K=1 74.38 Mbps 100% 34.09s K=2 74.79 Mbps 100% 33.93s K=3 71.88 Mbps 100% 34.24s K=4 74.68 Mbps 100% 34.94s K=5 72.56 Mbps 100% 34.22s The proposed DNN model K=1 73.28 Mbps 98.52% 0.32s K=2 73.66 Mbps 98.49% 0.38s K=3 70.83 Mbps 98.54% 0.31s K=4 72.95 Mbps 97.92% 0.35s K=5 71.15 Mbps 98.06% 0.32s Base station fixed power (43dBm) K=1 68.31 Mbps 91.83% N/A K=2 69.27 Mbps 92.62% N/A K=3 64.72 Mbps 90.04% N/A K=4 67.43 Mbps 90.29% N/A K=5 65.96 Mbps 90.86% N/A

Referring to Table 3 above, the full searching-based method is an iteration method that finds the optimal cell power in MATLAB, and can find the optimal system channel capacity. The proposed DNN structure is a value calculated based on the output value of the mobile base station derived from the 96-96-48-48 basis. Finally, the base station fixed power is the standard for maximizing the base station cell power in a randomly moving environment. calculated. The system channel capacity in the proposed DNN structure is reduced on average by 1.7%, but the processing time is confirmed to be more than 100 times faster. The full searching base requires a lot of processing time to find by changing all the outputs of the base station. Complexity can be expressed as N _k . N is the number of base stations, and k is the output control range (1-43 dBm). It is determined that the difference in processing time is larger if the number of terminals actually operated is large.

When such a learned DNN structure is determined, it is applied to the actual tactical environment. The mobile base station of each squad collects terminal information (SINR/Interference) at a preset period (for example, within 1 second) and provides it to the headquarters platoon vehicle, a mobile command vehicle, and the headquarters platoon vehicle utilizes a small unit level server 200 By setting input parameters of the determined DNN structure, the cell power for each mobile base station is derived. Thereafter, the cell power value can be provided to each mobile base station in the headquarters platoon vehicle to optimize the system network.

Meanwhile, in the above description, steps S110 to S122 may be further divided into additional steps or combined into fewer steps according to an embodiment of the present invention. In addition, some steps may be omitted if necessary, and the order between the steps may be changed. In addition, the contents of the method for optimizing the system network through the DNN-based mobile base station cell power adjustment in the tactical environment of FIGS. 1 to 7 may be applied to the contents of FIG. 8 even if other contents are omitted.

8 is a block diagram of a system network optimization system 100 and a server 200 in a tactical environment according to an embodiment of the present invention. At this time, the system 100 in the present invention extracts learning data and test data, learns a DNN structure based on this, and determines a DNN structure capable of optimizing the system network among the learned DNN structures to determine the server 200 provided with The server 200 is provided in the mobile command vehicle, receives the DNN structure from the system 100 and mounts it, and performs a role of optimizing the system network based on terminal information.

The system network optimization system 100 in a tactical environment according to an embodiment of the present invention includes a memory 110 and a processor 120 .

A program for optimizing the tactical environment system by adjusting cell power for a plurality of mobile base stations based on the DNN is stored in the memory 110 , and the processor 120 executes the program stored in the memory 110 .

The processor 120 first extracts learning data and test data for learning and testing of the DNN, and based on the extracted learning data and test data, a DNN structure for setting an optimal cell power for each of a plurality of mobile base stations to decide

The system network optimization server 200 in a tactical environment according to an embodiment of the present invention includes a memory 210 and a processor 220 .

A program for optimizing the tactical environment system by adjusting cell power for the plurality of mobile base stations based on the determined DNN structure is stored in the memory 210 , and the processor 220 executes the program stored in the memory 210 .

When the processor 220 receives the terminal information including the RSSI and SINR collected by the mobile base station according to a preset period, the processor 220 derives cell power for each mobile base station based on the determined DNN structure and provides it to each mobile base station. System network optimization can be performed.

The method for optimizing a system network through DNN-based mobile base station cell power control in a tactical environment according to an embodiment of the present invention described above is implemented as a program (or application) to be executed in combination with a computer, which is hardware, and stored in a medium. can be saved.

The above-mentioned program, in order for the computer to read the program and execute the methods implemented as a program, C, C++, JAVA, Ruby, which the processor (CPU) of the computer can read through the device interface of the computer; It may include code coded in a computer language such as machine language. Such code may include functional code related to a function defining functions necessary for executing the methods, etc. can do. In addition, the code may further include additional information necessary for the processor of the computer to execute the functions or code related to memory reference for which location (address address) in the internal or external memory of the computer should be referenced. have. In addition, when the processor of the computer needs to communicate with any other computer or server located remotely in order to execute the functions, the code uses the communication module of the computer to determine how to communicate with any other computer or server remotely. It may further include a communication-related code for whether to communicate and what information or media to transmit and receive during communication.

The storage medium is not a medium that stores data for a short moment, such as a register, a cache, a memory, etc., but a medium that stores data semi-permanently and can be read by a device. Specifically, examples of the storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the program may be stored in various recording media on various servers accessible by the computer or in various recording media on the computer of the user. In addition, the medium may be distributed in a computer system connected by a network, and computer-readable codes may be stored in a distributed manner.

The foregoing description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and likewise components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100, 200: system network optimization system in tactical environment, server
110, 210: memory
120, 220: processor

Claims (12)

In a system network optimization method through DNN-based mobile base station cell power adjustment in a tactical environment including a plurality of mobile base stations and a plurality of terminals,
extracting learning data and test data for learning and testing the DNN; and
Based on the extracted learning data and test data, comprising the step of determining a DNN structure for setting the optimal cell power for each of the plurality of mobile base stations,
The step of extracting learning data and test data for learning and testing of the DNN is,
randomly disposing the plurality of mobile base stations and a plurality of terminals;
calculating RSSI and SINR for the plurality of terminals as terminal information;
determining a cell power of a mobile base station capable of maximizing a system channel capacity based on the calculated terminal information and cell power set in the mobile base station; and
Extracting the learning data and the test data based on the determined cell power and terminal information,
The step of determining the cell power of the mobile base station,
calculating a downlink channel capacity for a plurality of terminals based on the terminal information; and
Determining the cell power of the mobile base station capable of maximizing the downlink channel capacity for each terminal,
The step of extracting the learning data and the test data based on the determined cell power output value and the terminal information,
Extracting training data and test data using the SINR of each of the plurality of terminals and a plurality of interference values received from the adjacent mobile base station as an input parameter of the DNN and the cell power of the mobile base station as an output parameter,
Determining a DNN structure for setting the optimal cell power for each of the plurality of mobile base stations comprises:
performing learning on a plurality of DNN structures based on the learning data; and
Selecting a DNN structure that minimizes MSE (Mean Squared Error) from among the plurality of learned DNN structures based on the test data,
collecting, by the mobile station, terminal information including RSSI and SINR and periodically providing it to a mobile command vehicle; and
Further comprising the step of deriving a cell power for each mobile base station based on the determined DNN structure in the mobile command vehicle,
A system network optimization method through DNN-based mobile base station cell power control in a tactical environment.
delete delete delete delete delete delete In the DNN learning system for system network optimization through DNN-based mobile base station cell power adjustment in a tactical environment including a plurality of mobile base stations and a plurality of terminals,
a memory in which a program for determining a DNN structure for setting an optimal cell power and learning a DNN structure to optimize a tactical environment system through cell power adjustment for the plurality of mobile base stations;
As the program stored in the memory is executed, learning data and test data for learning and testing of the DNN are extracted, and based on the extracted learning data and test data, an optimal solution for each of the plurality of mobile base stations is obtained. A processor for determining a DNN structure for setting cell power,
The processor randomly arranges the plurality of mobile base stations and the plurality of terminals, calculates RSSI and SINR for the plurality of terminals as terminal information, and based on the calculated terminal information and the cell power set in the mobile station , after determining the cell power of the mobile base station capable of maximizing the system channel capacity, extracting the learning data and the test data based on the determined cell power and terminal information,
The processor calculates the downlink channel capacity for each terminal based on the terminal information, and determines the cell power of the mobile base station capable of maximizing the downlink channel capacity for each terminal,
The processor extracts training data and test data using the SINR of each of the plurality of terminals and a plurality of interference values received from the adjacent mobile base station as an input parameter of the DNN and the cell power of the mobile base station as an output parameter,
The processor performs learning on a plurality of DNN structures based on the learning data, and selects a DNN structure that minimizes MSE (Mean Squared Error) among the learned plurality of DNN structures based on the test data. ,
DNN learning system for system network optimization in tactical environment.
delete delete delete delete

Images