KR102586761B1 - Intelligent dynamic real-time spectrum resource management system and intelligent dynamic real-time spectrum resource management method using data mining and case-based reasoning - Google Patents

Abstract

The present invention provides an intelligent dynamic real-time spectrum resource management method in an intelligent dynamic real-time spectrum resource management system, comprising the steps of a CR masternode collecting and pre-processing historical data as raw data; CR masternode creates a behavior pattern prediction model; A CR masternode analyzing input cases and generating result data; A CR user node determining whether to perform optimization; If it is decided to perform the optimization in the determining step, the CR user node generates transmission parameters that satisfy the criteria for the probability of interference with existing users in the channel; and the CR user node controlling transmission power based on transmission parameters. This prevents system performance from being degraded due to incorrect predictions when there are misclassification cases due to irregular patterns through an accurate learning algorithm for case-by-case data, reduces the complexity of intelligent dynamic real-time spectrum resource management, and enables rapid resource allocation. It can be made possible.

Description

Intelligent dynamic real-time spectrum resource management system and intelligent dynamic real-time spectrum resource management method using data mining and case-based reasoning REASONING}

The present invention relates to an intelligent dynamic real-time spectrum resource management system and an intelligent dynamic real-time spectrum resource management method using data mining and case-based reasoning. More specifically, it relates to an intelligent dynamic real-time spectrum resource management system using data mining and case-based reasoning that enables rapid resource allocation. It relates to a real-time spectrum resource management system and an intelligent dynamic real-time spectrum resource management method.

Recently, with the development of the Internet of Things (IoT), a paradigm shift in computing technology is rapidly occurring, and as a result, technologies that enable efficient processing of large amounts of data generated from various industrial Internet of Things devices are being developed. According to the International data Corporation, by 2025, IoT devices connected to the Internet will generate close to 79.4 zettabytes of data, and the number of such devices is expected to reach a whopping 44.1 billion. This will result in the smart devices, sensors, and cameras that make up the IoT. It is predicted that it will bring significant changes to major end terminals.

Meanwhile, these massive amounts of data are currently being processed in centralized data centers or external cloud systems, but these existing centralized computing systems provide various IoT services such as smart factories, smart farms, and self-driving cars that require real-time processing. There are limits to accepting it. Therefore, due to the risk of cloud server overload based on exponentially increasing data volume and network traffic, the computing transition from the existing centralized computing system to the network edge within the wireless access network is expected to become a new paradigm. Edge computing is a distributed computing system that brings computation and data storage closer to the data source. It processes computing services in a location close to the terminal used by the user, allowing users to receive faster and more stable services and perform flexible hybrid cloud computing. It has the advantage of

However, the limited frequency resources of edge computing cause frequent frequency handoffs in a dynamic spectrum environment that requires real-time mechanical device (MTD) operation, and interference with other homogeneous networks can cause severe performance degradation. Therefore, there is a need to dynamically optimize the distribution of frequency resources according to time-varying parameters such as channel state information.

To this end, the conventional intelligent dynamic spectrum resource management uses the learning engine to generate a backup channel list based on historical data representing the statistics of the existing local network for timely handoff to idle channels, and the inference engine uses the backup channel list. By performing a spectrum handoff operation to another idle channel, continuous communication was possible without disturbing existing users. At this time, if interference affects existing users in adjacent channels and locations, transmission parameters that can coexist between adjacent networks are dynamically reconfigured through an optimization engine that uses interference analysis and genetic algorithms.

However, decisions based on these optimization algorithms increase computational time and complexity, making rapid resource allocation difficult, and valuable solution information optimized according to the evolution of spectral states between space-time and frequency domains is wasted.

Therefore, there is a need to develop an intelligent dynamic real-time spectrum resource management method that can reduce the complexity of intelligent dynamic spectrum resource management and achieve rapid resource allocation.

Korean Patent Publication No. 10-2002-0030821

The present invention was created to solve the above problems, and the purpose of the present invention is to reduce system performance due to incorrect predictions when there are misclassification cases due to irregular patterns by using an accurate learning algorithm for case-specific data. The purpose is to provide an intelligent dynamic real-time spectrum resource management system and an intelligent dynamic real-time spectrum resource management method using data mining and case-based reasoning to prevent the complexity of intelligent dynamic real-time spectrum resource management and enable rapid resource allocation.

In order to achieve the above object, an intelligent dynamic real-time spectrum in an intelligent dynamic real-time spectrum resource management system including a CR masternode, at least one CR user node, and a case database constituting a CR network according to an embodiment of the present invention. The resource management method includes the steps of the CR master node collecting and pre-processing history data including sensing data sensing the channel state of the CR network network from the CR user node as raw data; The CR masternode learns behavioral patterns of users of the CR network using preprocessed data and creates a behavior pattern prediction model; The CR masternode analyzing input cases based on the generated behavior pattern prediction model to generate result data; The CR user node determining whether to perform optimization based on the result data; If it is decided to perform the optimization in the determining step, the CR user node generates transmission parameters that satisfy a standard for interference probability with existing users in the channel; and the user node controlling transmission power based on the transmission parameters.

In addition, the step of generating the result data includes verifying the similarity of the case in which the CR masternode is input by comparing the predicted value predicted through the behavior pattern prediction model with the actual value; The CR masternode generating result data based on the similarity verification result; And it may include the step of transmitting the resulting data generated by the CR master node to the CR user node.

In addition, the result data includes at least one of a previous solution, which is a solution for previous cases stored in the case database, and a backup channel list, which is a list of idle channels that are not used by users at a specific time and location in the CR network, , In the step of generating the result data based on the similarity verification result, if the predicted value and the actual value are the same, the CR master node generates the result data to include both the previous solution and the backup channel list, If the predicted value and the actual value are not the same, the result data may be generated to include only the backup channel list.

In addition, the step of determining whether to perform the optimization may include, upon receiving result data including the solution, performing solution verification, which is a verifier for the solution received by the CR user node; If the CR user node satisfies the interference probability standard as a result of the solution verification, determining whether to perform the optimization or not and maintaining the received solution; And if the CR user node does not satisfy the interference probability standard as a result of the solution verification, it may include determining whether to perform the optimization.

Additionally, in the step of performing the solution verification, the solution verification may be performed using interference analysis based on a Monte Carlo algorithm.

Additionally, the step of determining whether to perform the optimization may include determining whether to perform the optimization when result data including only the backup channel list is received.

In addition, the step of allowing the CR user node to update the case database by matching the generated transmission parameters with a solution for the input case may be further included.

Additionally, in the preprocessing step, the raw data may be preprocessed using at least one of data cleaning, data reduction, outlier detection, and data scaling.

Meanwhile, in an intelligent dynamic real-time spectrum resource management system including a CR master node, at least one CR user node, and a case database constituting a CR network according to another embodiment of the present invention to achieve the above object, the CR The master node includes a preprocessor that collects and preprocesses history data including sensing data sensing the channel state of the CR network network from the CR user node as raw data; a learning unit that learns behavioral patterns of users of the CR network using preprocessed data and generates a behavioral pattern prediction model; and a verification unit that generates result data by analyzing input cases based on the generated behavior pattern prediction model, wherein the CR user node includes: an interference analysis unit that determines whether to perform optimization based on the result data; and an optimization unit that, upon determining to perform the optimization, generates transmission parameters that satisfy standards for the probability of interference with existing users in the channel, and controls transmission power based on the transmission parameters.

In addition, the verification unit verifies the similarity of the input case by comparing the predicted value predicted through the behavior pattern prediction model with the actual value, and generates the result data based on the similarity verification result, and the generated result data is the It can be delivered to the CR user node.

In addition, the result data includes at least one of a previous solution, which is a solution for previous cases stored in the case database, and a backup channel list, which is a list of idle channels that are not used by users at a specific time and location in the CR network, , the verification unit generates the result data so that the CR master node includes both the previous solution and the backup channel list if the predicted value and the actual value are the same, and if the predicted value and the actual value are not the same, the backup The resulting data can be generated to include only the channel list.

In addition, when the interference analysis unit receives result data including the solution, it performs solution verification, which is a verifier for the solution received by the CR user node. If the solution verification result satisfies the interference probability standard, it determines whether the optimization is performed. is determined not to be performed so that the received solution is maintained, and if the solution verification result does not satisfy the interference probability standard, it can be decided whether to perform the optimization.

Additionally, the interference analysis unit may perform the solution verification using interference analysis based on a Monte Carlo algorithm.

Additionally, when the interference analysis unit receives result data including only the backup channel list, it can determine whether to perform the optimization.

Additionally, the CR user node can match the generated transmission parameters with solutions for the input cases and update them in the case database.

According to one aspect of the present invention described above, by providing an intelligent dynamic real-time spectrum resource management system and an intelligent dynamic real-time spectrum resource management method using data mining and case-based reasoning, irregular patterns are generated through an accurate learning algorithm for case-specific data. In cases of misclassification, it can prevent system performance from being degraded due to incorrect predictions, reduce the complexity of intelligent dynamic real-time spectrum resource management, and enable rapid resource allocation.

1 is a diagram illustrating the performance of spectrum resource management using a conventional optimization engine;
Figure 2 is a diagram illustrating spectrum resource management performance through an intelligent dynamic real-time spectrum resource management system according to an embodiment of the present invention;
Figure 3 is a configuration diagram of an intelligent dynamic real-time spectrum resource management system according to an embodiment of the present invention;
Figure 4 is a block diagram showing the specific configuration of the CR masternode shown in Figure 1;
Figure 5 is a block diagram showing the specific configuration of the CR user node shown in Figure 1;
Figure 6 is a diagram showing an example of pre-processing in the pre-processing unit shown in Figure 4;
Figure 7 is a diagram showing the process of learning and reasoning in the learning unit shown in Figure 4;
Figure 8 is a diagram showing an intelligent dynamic real-time spectrum resource management method according to another embodiment of the present invention, and
Figures 9 and 10 are diagrams specifically showing the intelligent dynamic real-time spectrum resource management process shown in Figure 8.

The detailed description of the present invention described below refers to the accompanying drawings, which show by way of example specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that the various embodiments of the present invention are different from one another but are not necessarily mutually exclusive. For example, specific shapes, structures and characteristics described herein may be implemented in one embodiment without departing from the spirit and scope of the invention. Additionally, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the invention. Accordingly, the detailed description that follows is not intended to be taken in a limiting sense, and the scope of the invention is limited only by the appended claims, together with all equivalents to what those claims assert, if properly described. Similar reference numbers in the drawings refer to identical or similar functions across various aspects.

The components according to the present invention are components defined by functional division rather than physical division, and can be defined by the functions each performs. Each component may be implemented as hardware or program code and processing units that perform each function, and the functions of two or more components may be included and implemented in one component. Therefore, the names given to the components in the following embodiments are not intended to physically distinguish each component, but are given to suggest the representative function performed by each component, and the names of the components refer to the present invention. It should be noted that the technical idea is not limited.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the drawings.

First, FIG. 1 is a diagram illustrating the performance of spectrum resource management using a conventional optimization engine, and FIG. 2 is a diagram illustrating the performance of spectrum resource management through an intelligent dynamic real-time spectrum resource management system according to an embodiment of the present invention. , and Figure 3 is a configuration diagram of an intelligent dynamic real-time spectrum resource management system according to an embodiment of the present invention.

As shown in Figure 1, spectrum resource management based on the conventional optimization algorithm increases calculation time and complexity, making it difficult to quickly allocate resources within a limited time, and wastes valuable solution information optimized according to the evolution of each spectrum state in the space-time and frequency domains. The problem is that you have no choice but to do it.

Accordingly, the intelligent dynamic real-time spectrum resource management system (hereinafter referred to as the system) of the present invention is designed to reduce the complexity of spectrum resource management using an existing optimization engine and achieve rapid resource allocation, as shown in FIG. 2. For this purpose, this system uses case-based reasoning. If there are misclassified cases due to irregular patterns, incorrect predictions may cause system performance to deteriorate, so the need for an accurate learning algorithm for case-specific data is emphasized. do.

Accordingly, this system can use data mining along with case-based reasoning.

And the system according to this embodiment includes at least one CR user node (CR User, 200) and a CR master node (CR Master, 100), forming a CR (Cognitive Radio) network.

Here, the CR network is IEEE 802.22. A base station using the TV White Space (TVWS) band according to the wireless local area network (WRAN) standard may be an infrastructure-based CR network that controls IoT devices in suburban areas where broadband access is difficult to support, but is not limited to this.

At this time, the CR masternode 100, which is each base station constituting a single wireless network, independently manages frequency resources within the CR network through database contact for self-coexistence between cells of the same type existing in adjacent channels and locations. The CR user node 200 is a terminal under the management of the CR master node 100 and can opportunistically use idle frequency resources within the CR network.

For this purpose, the sensing data sensed by each CR user node 200 of the channel status in the CR network is generated by the CR master, which regularly uses data mining to analyze the activity patterns of existing users from the history data stored in the database (DB). It is transmitted (ηdata) to the database through the node 100. Additionally, when making inferences using previous cases classified through data mining, the CR Masternode 100 determines (②retrieve) whether to provide a solution for cases similar to this case through similarity verification. In addition, the CR Masternode 100 preprocesses history data from the database (DB) (data cleaning, data reduction, data scaling) and performs machine learning based on the preprocessed data to determine user pattern patterns. The pattern can be extracted and delivered to the database (DB) for storage.

Also, in the idle channel inference process (③) using the backup channel list, the CR user node 200 operates an optimization engine according to the similarity verification result (⑤ optimization engine) when interference affects existing users existing in adjacent channels or locations. Alternatively, the solution provided by the CR Masternode (100) is reused (④ reuse (solved case)). Additionally, the CR user node 200 can verify the solution for determining the operation of the optimization engine through interference analysis based on a Monte Carlo algorithm. Additionally, the optimized solution is updated (⑥ retain(optimized case)) as a case database in the database (DB) for the next inference.

In addition, case-based reasoning performed in the system of the present invention is performed by classifying the roles of the CR master node 100 and CR user node 200, as shown, and is performed using a case database created through data mining for each case. Similarity verification, solution reuse and verification, solution maintenance, and update operations can be performed.

Below, with reference to FIGS. 4 to 7 , the specific configuration of each CR master node 100 and CR user node 200 will be described.

FIG. 4 is a block diagram showing a specific configuration of the CR master node 100 shown in FIG. 1, FIG. 5 is a block diagram showing a specific configuration of the CR user node 200 shown in FIG. 1, and FIG. 6 is a drawing. 4 is a diagram showing an example of preprocessing in the preprocessing unit 121, and FIG. 7 is a diagram illustrating a learning and inference process in the learning unit 123 shown in FIG. 4.

In addition, the CR master node 100 and CR user node 200 that constitute the system may have software (application) installed and executed to perform an intelligent dynamic real-time spectrum resource management method.

CR masternode 100 includes a communication unit 110, a control unit 120, and a storage unit 130.

The communication unit 110 communicates with at least one CR user node 200 in the CR network, receives channel state sensing data from the CR user node 200, and sends the resulting data including the backup channel list to the CR user node 200. ) is sent to the side.

In addition, the communication unit 110 communicates with a database (DB) linked to the CR network to transmit channel state sensing data, solutions, and optimization cases from the CR user node 200 to the database (DB), and communicates with previous cases for a certain period of time. The collected sensing data can be received from a database (DB) as history data.

In particular, the control unit 120 of the CR masternode 100 according to this embodiment creates a backup channel list through characterization, frequency determination, and classification based on history data representing statistical information of the CR network, which is a local network, to provide CR user information. It is possible to facilitate timely transition of the node 200 to an idle channel.

In addition, the control unit 120 performs data preprocessing and data using an artificial neural network based on information sensed from the CR user node 200 for each preset inference cycle unit to improve prediction performance through accurate pattern learning of the sensed data. Mining can be performed.

For this purpose, the control unit 120 may include a preprocessing unit 121, a learning unit 123, and a verification unit 125.

The preprocessor 121 may collect and preprocess history data including sensing data obtained by sensing the channel state of the CR network network from the CR user node 200 as raw data. The data preprocessed in this way is used by the learning unit 123 for machine learning.

Data preprocessing performed in the preprocessing unit 121 is a technology to improve the quality of raw data. As shown in FIG. 6, data cleaning, data reduction, and outlier identification using historical data, which is raw data, are performed. Raw data can be preprocessed using at least one of outlier detection and data scaling.

At this time, the preprocessing unit 121 of the present invention performs a data preprocessing process consisting of data cleaning, data reduction, outlier detection, and data scaling steps to make data preprocessing more desirable. It can be done.

Data cleaning aims to construct new features based on linear combinations of existing variables. A representative linear feature extraction method is a statistical feature extraction method that calculates statistics such as the average or median value of raw data within a specific time range. However, if time series fluctuations are severe, data loss may occur, so the preprocessor 121 of the present invention can perform data reduction from statistical data to categorical data.

Categorical data refers to the frequency of occurrence of values within the range when the range in which statistical data values exist is divided into several intervals of equal size according to user definition. This method can minimize information loss and data size. When analyzing large-scale data, the amount of calculation can be effectively reduced.

More specifically, the preprocessor 121 calculates the occupancy probability representing the statistical characteristics of the data sensed within a total of M time slots per channel. Can be calculated based on Equation 1 below.

[Equation 1]

here means the jth channel, may mean the timeslot at time i.

At this time, the section width is based on the number of sections L specified within the categorical data set range (min, max). The divided value can be defined as Equation 2 below.

[Equation 2]

Therefore The kth median of these statistics Based on If it is in the range, it is defined as 1; otherwise, it is defined as 0. At this time The frequency of appearance per section is It can mean the result value counted up to. This can be defined as Equation 3 below.

[Equation 3]

here It can be.

Meanwhile, repetitive and irregular data due to various transmission errors, sensor malfunctions, or external interference during the data collection process are highly likely to generate outliers during data preprocessing. This excessively increases the variance in estimating the data population mean, causing a decrease in data analysis accuracy.

In order to solve this problem, the pre-processing unit 121 of the present invention calculates statistical data from raw data that is difficult to distinguish between outliers and noise through data reduction and outlier identification, and represents the characteristics of each channel. At this time, when the frequency of occurrence is calculated by dividing into several sections within the range where statistical data values exist, the data in the section with the fewest observed values is identified as an outlier and converted to 0 as shown in FIG. 6.

Conventionally, a method of encoding data scaling into binary numbers is used, which facilitates pattern classification by data characteristics and enables efficient search. However, in order to classify categorical data into binary numbers of 0 and 1, a high-resolution categorical data set that can be distinguished from neighboring data sets with different characteristics is required.

Accordingly, in order to solve this problem, the preprocessor 121 according to this embodiment rescales the range of each component in the categorical data set to a range between 0 and 1 using Min-Max Normalization to provide overall unity to the data. can be granted. The general formula for normalization is shown in Equation 4 below.

[Equation 4]

Here, max and min may refer to the size of the categorical data set, and x' may refer to the normalized value. The preprocessor 121 of the present invention can minimize the amount of calculation by preprocessing data in the above order.

Meanwhile, the learning unit 123 can learn the behavior patterns of CR network users using preprocessed data and create a behavior pattern prediction model.

Specifically, the learning unit 123 classifies the preprocessed history data by case and determines whether to deliver optimized solution information to the CR user node 200 according to the inference results obtained using the classified case database.

Artificial neural networks are used in a wide range of areas such as communication, signal processing, and intelligent control. In particular, in CRN (Cognitive Radio Network), artificial neural networks learn behavioral patterns of existing users from data sets to provide high prediction accuracy for existing user patterns in the future. have Accordingly, the learning unit 123 of the present invention can use an artificial neural network configured as shown in FIG. 7 to learn and infer features of historical data preprocessed according to categorical data size and data scaling techniques.

Referring to FIG. 7, the learning unit 123 of the present invention can use an ANN structure to achieve high prediction accuracy for future existing user patterns by learning behavioral patterns of existing users from a data set. The ANN structure consists of an input layer, a hidden layer, and an output layer. And the input layer plays the role of delivering content image information composed of N neurons to the hidden layer. The hidden layer checks whether the content image received from the input layer contains a characteristic pattern and transmits the content cost to the output layer. Assuming that a feature pattern is defined for each hidden layer as shown in FIG. 7, the content cost calculation for the content image included in the feature pattern can be determined by the thickness (weight) of the arrow connecting the hidden layers. The output layer considers the content cost information calculated in the hidden layer and outputs certainty as a value between 0 and 1. Finally, the final content image can be determined by comparing the certainty factors of each output layer from 1 to 9.

This learning unit 123 can define the variables necessary to explain the relationship between neurons constituting the ANN and organize the relational expressions for the content cost and certainty factor of the hidden layer and output layer in the previously constructed ANN. First, the ith input layer neuron weight on and threshold Output for each hidden layer given Is It can be defined as, which can be calculated based on Equation 5 below.

[Equation 5]

here It can be.

At this time is a sigmoid function continuously expressed at intervals between 0 and 1, and the input linear sum a of the jth hidden layer can be defined as Equation 6 below.

where the threshold is to block noise signals coming from transmission sources other than the transmission signals of the input layer connected by the thick arrow shown in FIG. 7.

Through this, the learning unit 123 determines the content cost calculation result output by the sigmoid function as an output certainty factor having a value between 0 and 1 for the content image output layer. In other words, the preprocessed learning data can be classified into classes according to the average occupancy probability from 0.1 to 0.9. For example, if the certainty coefficient of output layer 1 is very close to 1, we can be sure that the average occupancy probability of the input image is 0.1.

And the learning unit 123 may define an objective function for weight and threshold optimization. More specifically, the squared error of the output layer for each input data defined above using linear regression. The optimal weight and threshold for minimizing the sum of can be derived based on Equation 6 below.

[Equation 6]

In the above regression equation, and is a variable for entering the predicted value of the data, and on the right side is an explanatory variable, on the left side can be defined as a predictor variable. At this time, the objective variable given to the kth output layer is and predicted value square error between Can be calculated based on Equation 7 below.

[Equation 7]

Here, the parameter p(= ) and q(= ) Objective function to optimize is calculated from the entire T input data. The optimal parameters can be derived when the sum of is minimized.

[Equation 8]

=

here It can be.

Meanwhile, the verification unit 125 may generate result data by analyzing input cases based on the behavior pattern prediction model generated by the learning unit 123.

This verification unit 125 verifies the similarity of the input case by comparing the predicted value predicted through a behavior pattern prediction model with the actual value, generates result data based on the similarity verification result, and the generated result data is sent to the CR user. It can be delivered to the node 200.

Here, the result data may include at least one of a previous solution, which is a solution for a previous case stored in the case database, and a backup channel list, which is a list of idle channels not used by users at a specific time and location in the CR network.

Accordingly, in order to verify this similarity, the verification unit 125 uses the objective function when a new case is input. Similarity verification is performed using the optimal number of hidden layers and threshold that minimizes .

If the predicted value and the actual value are the same, the verification unit 125 may generate result data so that the CR master node 100 includes both the previous solution and the backup channel list.

And, if the predicted value and the actual value are not the same, that is, if they are not the same, the verification unit 125 may generate result data to include only the backup channel list.

Meanwhile, the storage unit 130 is provided to store information transmitted, received, or created by the CR master node 100, and software (application) for performing an intelligent dynamic real-time spectrum resource management method can be stored.

Meanwhile, the CR user node 200 must not cause interference to existing users when sharing spectrum in an opportunistic manner by identifying idle channels defined as unused spectrum holes or white spaces at a specific time and location. .

If an existing user is detected during the process of inferring an idle channel, the CR user node 200 must pause its operation and switch to another idle channel to continue communication. This switching method is defined as spectrum handoff, and in the present invention, inference engine operation can be performed through a reactive handoff technique based on a backup channel list.

At this time, if the CR user node 200 using the idle channel with the maximum transmission power causes interference to existing users existing in adjacent channels and locations, the CR user node 200 reuses the solution or operates the optimization engine according to the similarity verification results. It can be done. For this purpose, the CR user node 200 includes a communication unit 210, a control unit 220, and a storage unit 230.

When the control unit 220 senses the spectrum for a certain period and generates sensing data, which is channel state information, the communication unit 210 transmits the sensing data to the CR master node 100 and transfers the sensing data from the CR master node 100 to the backup channel. Receive result data containing a list.

Meanwhile, the control unit 220 performs spectrum handoff operations on other idle channels to enable continuous communication without disturbing existing users.

More specifically, the control unit 220 includes an interference analysis unit 221 composed of an inference engine and can determine whether to perform optimization based on the result data.

When the interference analysis unit 221 receives result data including a solution from the CR master node 100, it performs solution verification, which is a verifier for the received solution, and optimizes the solution if the solution verification result satisfies the interference probability standard. By determining whether to perform, it is possible to maintain the solution received from the CR masternode 100 without performing separate optimization through the optimization unit 223.

In addition, if the solution verification result does not satisfy the interference probability standard, that is, it is confirmed that the interference analysis unit 221 does not satisfy the interference probability standard, the interference analysis unit 221 may decide whether to perform optimization and allow the optimization unit 223 to perform the optimization.

This interference analysis unit 221 can perform solution verification using interference analysis based on a Monte Carlo algorithm.

More specifically, if the predicted value and the actual value are the same, the interference analysis unit 221 reuses the solution received from the CR masternode 100 and can use interference analysis based on a Monte Carlo algorithm to verify the solution. To this end, the interference analysis unit 221 acquires randomly generated radio wave transmission-related samples based on input parameters applied to a specific interference environment scenario in interference analysis based on a Monte Carlo algorithm, and uses the obtained samples to Determine the probability of interference. In this interference environment scenario, a propagation loss model can be defined according to the interference parameters (antenna height, transmission power, center frequency, etc.) and specific distance, location, and terrain characteristics. And this interference environment refers to existing users in use at a specific location and time in the licensed band, the interfered link including the interfered transmitter and the interfered receiver, and the adjacent channel and location based on CR technology in the licensed band. Any communication that temporarily uses an idle channel that is not used by existing users at a specific time may include an interfering link including an interfered receiver and an interfering transmitter (CR user node).

In addition, the interference analysis unit 221 calculates the received power that the interfered receiver receives from the interfered transmitter and the received power that the interfered receiver receives from the interfered transmitter, and determines the probability of interference based on this.

Additionally, when the interference analysis unit 221 receives result data including only the backup channel list, it can determine whether to perform optimization and allow the optimization unit 223 to perform the optimization.

In addition, solution verification can determine solution maintenance or operation of the optimization unit 223 according to the 5% interference probability standard applied in field tests. If the interference probability exceeds the 5% standard or more during the solution verification process, the interference analysis unit 221 determines that the received solution information is not appropriate, and the interference probability 223 determines that the interference probability standard within 5% is met. Transmission parameters are derived. Conversely, if the interference probability standard of less than 5% is met, the received past solution is maintained. These specific interference probability standards are illustrative for convenience of explanation and are not limited thereto.

If the predicted value and the actual value are different, the interference analysis unit 221 determines that there is no similar case in the past, calculates the probability of interference in the interference analysis unit 221, and then, when interference occurs, the existing user and the existing user are processed through the optimization unit 223. Optimal transmission parameters that can coexist can be derived.

And when the interference analysis unit 221 decides to perform optimization, the control unit 220 generates transmission parameters that satisfy the standards for the probability of interference with existing users in the channel, and the optimization unit 223 controls the transmission power based on the transmission parameters. ) further includes.

And the control unit 220 reuses the solution for each time slot until the entire inference cycle or matches the transmission parameters (parameters) generated through the operation of the optimization unit 223 with the solution for the new input case, which leads to the next inference. For this reason, it can be updated in the case database.

Meanwhile, the storage unit 230 is provided to store information transmitted, received, or generated by the CR user node 200, and software (application) for performing an intelligent dynamic real-time spectrum resource management method can be stored.

Through this, the system of the present invention facilitates classification by characteristics of history data sensed from a large amount of IoT, and reuses solutions for past similar cases when inferring using a case database containing past optimized solution information to optimize the existing optimization algorithm. It can reduce the amount of calculation and complexity of decision-making based on and enable efficient real-time resource allocation.

Meanwhile, Figure 8 is a diagram illustrating an intelligent dynamic real-time spectrum resource management method according to another embodiment of the present invention, and Figures 9 and 10 specifically illustrate the intelligent dynamic real-time spectrum resource management process shown in Figure 8. It is a drawing.

Since the intelligent dynamic real-time spectrum resource management method according to an embodiment of the present invention is carried out on substantially the same configuration as the system shown in FIGS. 2 to 7, the same reference numerals refer to the same components as the systems shown in FIGS. 2 to 7. will be given, and repeated explanations will be omitted.

The intelligent dynamic real-time spectrum resource management method according to this embodiment is an intelligent dynamic real-time spectrum resource management method in a system including at least one CR user node 200 and CR master node 100 constituting a CR network. , raw data is collected and preprocessed in the CR masternode 100 (S110), a behavior pattern prediction model is created based on the data preprocessed in the CR masternode 100 (S120), and a case is created based on the behavior pattern prediction model. is analyzed (S130), the resulting data is transmitted from the CR master node (100) to the CR user node (200) (S140), and the CR user node (200) determines whether to perform optimization (S150). When determined (S150-Yes), the CR user node 200 creates transmission parameters (S160), and the CR user node 200 controls transmission power based on the generated transmission parameters (S170) and maintains transmission power. Control (S180) to do so.

If it is determined that optimization is not performed (S150-No), the CR user node 200 controls the transmission power to be maintained as the existing solution based on the existing solution (S180).

In the step (S110) of collecting and preprocessing raw data in the CR master node 100, history data including sensing data sensing the channel status of the CR network network in the CR user node 200 is collected as raw data and preprocessed. You can.

Figure 9 is a flow of data mining using history data. Referring to Figure 9, in the step (S120) of generating a behavior pattern prediction model based on the data preprocessed in the CR masternode 100, the preprocessed data Using this, you can learn the behavior patterns of CR network users and create a behavior pattern prediction model.

And in the step (S130) where the CR masternode 100 analyzes the case based on the behavior pattern prediction model, result data can be generated by analyzing the input case based on the generated behavior pattern prediction model.

The step 130 of analyzing these cases and generating result data is a step of verifying the similarity of the cases entered by the CR master node 100 by comparing the predicted values predicted through a behavior pattern prediction model with the actual values. CR master It may include the step of the node 100 generating result data based on the similarity verification result and the step of the CR master node 100 transmitting the generated result data to the CR user node 200.

Here, the result data includes at least one of a previous solution, which is a solution for a previous case stored in the case database, and a backup channel list, which is a list of idle channels not used by users at a specific time and location in the CR network.

And if the predicted value predicted through the behavior pattern prediction model for the input case is the same as the actual value, the CR Masternode (100) generates result data to include all previous solutions and backup channel lists, and the predicted value and the actual value are different. If they are the same, the resulting data can be generated to include only the backup channel list.

Additionally, in the step (S150) where the CR user node 200 determines whether to perform optimization, it determines whether to perform optimization based on the result data.

In the step of determining whether to perform such optimization (S150), upon receiving result data including a solution, the CR user node 200 performs solution verification, which is a verification of the received solution. If the interference probability standard is satisfied as a result of the solution verification, it is decided whether to perform optimization or not, and the received solution is maintained. If the CR user node 200 does not meet the interference probability standard as a result of the solution verification, the CR user node 200 determines whether to perform optimization. It may include a step of deciding.

In this step of performing solution verification, solution verification can be performed using interference analysis based on Monte Carlo algorithm.

And in the step of determining whether to perform optimization (S150), when the CR user node 200 receives result data including only the backup channel list from the CR master node 100, a step of determining whether to perform optimization is further performed. It may also be included.

When optimization performance is determined (S150-Yes), in the step (S160) where the CR user node 200 generates transmission parameters, the CR user node 200 creates transmission parameters that satisfy the criteria for the probability of interference with existing users in the channel. can be created.

Thereafter, the CR user node 200 may perform a step (S170) of controlling transmission power based on the generated transmission parameters.

And the CR user node 200 controls to maintain transmission power (S180).

If it is determined that optimization is not performed (S150-No), the CR user node 200 controls the transmission power to be maintained as the existing solution based on the existing solution (S180).

In addition, the intelligent dynamic real-time spectrum resource management method may further include a step of allowing the CR user node 200 to update the case database by matching the generated transmission parameters with a solution for the input case.

As shown in Figure 10, after learning is completed in the CR master node 100, the spectrum resource management method using case-based reasoning is performed by classifying the roles of the CR master node 100 and CR user node 200. , Similarity verification, solution reuse and verification, solution maintenance, and update operations can be performed using the case database created through case-specific data mining.

The intelligent dynamic real-time spectrum resource management method of the present invention can be implemented in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., singly or in combination.

The program instructions recorded on the computer-readable recording medium may be those specifically designed and configured for the present invention, or may be known and usable by those skilled in the computer software field.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floptical disks. media), and hardware devices specifically configured to store and perform program instructions, such as ROM, RAM, flash memory, etc.

Examples of program instructions include not only machine language code such as that created by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the invention and vice versa.

Although various embodiments of the present invention have been shown and described above, the present invention is not limited to the specific embodiments described above, and may be used in the technical field to which the invention pertains without departing from the gist of the invention as claimed in the claims. Of course, various modifications can be made by those skilled in the art, and these modifications should not be understood individually from the technical idea or perspective of the present invention.

100: CR master node 200: CR user node

Claims (15)

As an intelligent dynamic real-time spectrum resource management method in an intelligent dynamic real-time spectrum resource management system including a CR masternode, at least one CR user node, and a case database constituting a CR network,
The CR master node collects and pre-processes history data including sensing data obtained by sensing the channel state of the CR network network from the CR user node as raw data;
The CR masternode learns behavioral patterns of users of the CR network using preprocessed data and creates a behavior pattern prediction model;
The CR masternode analyzing input cases based on the generated behavior pattern prediction model to generate result data;
The CR user node determining whether to perform optimization based on the result data;
If it is decided to perform the optimization in the determining step, the CR user node generates transmission parameters that satisfy a standard for interference probability with existing users in the channel; and
Comprising the user node controlling transmission power based on the transmission parameters,
The step of generating the result data is,
Verifying the similarity of cases entered by the CR masternode by comparing predicted values predicted through the behavior pattern prediction model with actual values;
The CR masternode generating result data based on the similarity verification result; and
It includes the step of transmitting the resulting data generated by the CR master node to the CR user node,
The result data is,
Contains at least one of a previous solution, which is a solution for previous cases stored in the case database, and a backup channel list, which is a list of idle channels not used by users at a specific time and location in the CR network,
In the step of generating the result data based on the similarity verification result,
If the predicted value and the actual value are the same, the CR master node generates the result data to include both the previous solution and the backup channel list, and if the predicted value and the actual value are not the same, only the backup channel list is included. An intelligent dynamic real-time spectrum resource management method characterized by generating the above result data. delete delete According to paragraph 1,
The step of determining whether to perform the optimization is,
Upon receiving result data including the solution, the CR user node performs solution verification, which is a verification of the received solution;
If the CR user node satisfies the interference probability standard as a result of the solution verification, determining whether to perform the optimization or not and maintaining the received solution; and
An intelligent dynamic real-time spectrum resource management method comprising the step of determining whether to perform the optimization if the CR user node does not satisfy the interference probability standard as a result of the solution verification. According to paragraph 4,
In the step of performing the solution verification,
An intelligent dynamic real-time spectrum resource management method characterized in that the solution verification is performed using interference analysis based on a Monte Carlo algorithm. According to paragraph 1,
The step of determining whether to perform the optimization is,
An intelligent dynamic real-time spectrum resource management method comprising the step of determining whether to perform the optimization when receiving result data including only the backup channel list. According to paragraph 1,
An intelligent dynamic real-time spectrum resource management method further comprising allowing the CR user node to update the case database by matching the generated transmission parameters with a solution for the input case. According to paragraph 1,
In the preprocessing step,
An intelligent dynamic real-time spectrum resource management method characterized by preprocessing the raw data using at least one of data cleaning, data reduction, outlier detection, and data scaling. It is an intelligent dynamic real-time spectrum resource management system that includes a CR master node, at least one CR user node, and a case database that constitute the CR network,
The CR masternode is,
a pre-processing unit that collects and pre-processes history data including sensing data obtained by sensing the channel state of the CR network network from the CR user node as raw data;
a learning unit that learns behavioral patterns of users of the CR network using preprocessed data and generates a behavioral pattern prediction model; and
And a verification unit that analyzes input cases based on the generated behavior pattern prediction model and generates result data,
The CR user node is,
an interference analysis unit that determines whether to perform optimization based on the result data; and
When it is decided to perform the optimization, it generates transmission parameters that satisfy the standards for the probability of interference with existing users in the channel, and includes an optimization unit that controls transmission power based on the transmission parameters,
The verification department,
The similarity of the input case is verified by comparing the predicted value predicted through the behavior pattern prediction model with the actual value, the result data is generated based on the similarity verification result, and the generated result data is delivered to the CR user node. And
The result data is,
Contains at least one of a previous solution, which is a solution for previous cases stored in the case database, and a backup channel list, which is a list of idle channels not used by users at a specific time and location in the CR network,
The verification department,
If the predicted value and the actual value are the same, the CR master node generates the result data to include both the previous solution and the backup channel list, and if the predicted value and the actual value are not the same, only the backup channel list is included. An intelligent dynamic real-time spectrum resource management system characterized by generating the above result data. delete delete According to clause 9,
The interference analysis unit,
Upon receiving result data containing the solution, the CR user node performs solution verification, which is a verifier for the received solution,
If the solution verification result satisfies the interference probability standard, the optimization is decided not to be performed and the received solution is maintained.
An intelligent dynamic real-time spectrum resource management system, characterized in that if the solution verification result does not satisfy the interference probability standard, it is decided whether to perform the optimization. According to clause 12,
The interference analysis unit,
An intelligent dynamic real-time spectrum resource management system characterized in that the solution verification is performed using interference analysis based on a Monte Carlo algorithm. According to clause 9,
The interference analysis unit,
An intelligent dynamic real-time spectrum resource management system characterized in that, upon receiving result data including only the backup channel list, it is determined whether to perform the optimization. According to clause 9,
The CR user node is,
An intelligent dynamic real-time spectrum resource management system characterized in that the generated transmission parameters are matched with solutions for the input cases and updated in the case database.