JP6627496B2 - Management device, program to be executed by computer, and computer-readable recording medium recording the program - Google Patents

Description

  The present invention relates to a management device, a program to be executed by a computer, and a computer-readable recording medium storing the program.

  In a cognitive radio system or a frequency sharing mobile communication system, in a frequency band in which a primary user who is a licensed radio communication system exists, a communication band is secured by utilizing a time and spatial frequency space. , Improve communication capacity.

  However, since a radio station that performs communication using the frequency of the primary user uses the frequency of the primary user as a secondary user, there is a restriction that interference is not given to communication of the primary user. You. Therefore, the secondary user needs to appropriately set the transmission power and the like.

  Interference with the primary user is affected not only by the frequency channel, transmission power, and modulation method of the secondary user, but also by the positional relationship between the primary and secondary users, surrounding buildings, plants, weather, and the like. Receive. Therefore, it is difficult to accurately model the interference amount and derive theoretically appropriate parameters. In addition, since the positional relationship, weather, and the like change with time, it is necessary to set appropriate parameters in real time so as not to interfere with the primary user.

  Conventionally, the following techniques are known as techniques for setting parameters of a secondary user.

  Using a large number of sensors, a signal transmitted by a primary user is detected, a database of frequency availability at a certain point and time is used, and parameters of the secondary user are set using the database (Non-patent document). Reference 1).

  The secondary user performs a receiving operation before transmitting a signal, and performs transmission with confidence that the primary user does not use a frequency channel (Non-Patent Document 2). In other words, the secondary user performs carrier sense, and performs transmission when wireless communication is not performed as a result of the carrier sense.

  The secondary user calculates the expected value of the amount of interference to the primary user from the signal reception power from the primary user, and dynamically adjusts the transmission power by PID control to avoid interference (Non-Patent Document 1). Reference 3).

  The secondary user obtains an estimated value of the amount of interference with the primary user from the propagation model, and controls the transmission power so that the estimated value is equal to or less than a predetermined value (Non-Patent Document 4).

JP 2013-213609 A

FCC, “Second report and order and memorandum opinion and order,” Docket no. 08-260, Nov. 2008. A. Tsertou and D.I. Laurenson, “Revisiting the Hidden Terminal Problem in a CSMA / CA Wireless Network,” IEEE Trans. Mob. Comput., Vol. 7, no. 7. Pp 817-831, Jul. 2008. Motoki Matsui, et al. , "PID control type power adjustment algorithm considering radio interference in cognitive wireless networks (wireless)," IEICE technical report. NS, Network System 110.448 (2011): 217-222. Takeo Fujii, "Multi-channel multi-hop cognitive wireless network for avoiding interference due to power limitation," IEICE Transactions B91.11 (2008): 1369-1379. K. Karmarkar, "A New Polynomial-Time Algorithm for Linear Programming," Combinatorica, 4 (1984), 373-395. Glover, Fred (1986). "Future Paths for Integer Programming and Links to Artificial Intelligence". Computers and Operations Research 13 (5): 533-549.doi: 10.1016 / 0305-0548 (86) 90048-1. Cox, DR (1958). "The regression analysis of binary sequences (with discussion)". J Roy Stat Soc B 20: 215-242. Koby Crammer, Alex Kulesza and Mark Dredze, “Adaptive Regularization of Weight Vectors,” NIPS 2009.

  However, in the technique described in Non-Patent Document 1, when the primary user performs only reception, it is not possible to consider interference at the reception point. Also, if the transmission power is limited, it is difficult to find a usable frequency channel even if there is a usable frequency channel. Furthermore, radio wave propagation fluctuates due to the influence of surrounding buildings, but it is difficult for databases to follow such real-time changes.

  Also, in the technique described in Non-Patent Document 2, when the transmitting terminal of the primary user and the transmitting terminal of the secondary user are separated, and the receiving terminal of the primary user is close to the secondary user First, interference occurs at the receiving terminal of the primary user. This is a hidden terminal problem and cannot be avoided by avoiding interference by carrier sense.

  Furthermore, in the technology described in Non-Patent Document 3, it is assumed that the primary user is a transmitting / receiving terminal and the secondary user can receive the signal, but as described in the technology in Non-Patent Document 2, Regarding interference with the primary user's receiving terminal, it is not possible to receive a signal from the primary user's receiving terminal, so that interference cannot be avoided.

  Furthermore, in the technique described in Non-Patent Document 4, various parameters relating to a primary user, specifically, a propagation attenuation constant, transmission power, a maximum communication distance, a desired signal-to-noise ratio, and the like are acquired in advance. There is a need. When there are many primary users, it is difficult to acquire these parameters for all the primary users.

  According to an embodiment of the present invention, there is provided a management device capable of determining a communication parameter for a secondary user to perform wireless communication while avoiding interference in real time.

  Further, according to an embodiment of the present invention, there is provided a program for causing a computer to determine in real time a communication parameter for a secondary user to perform wireless communication while avoiding interference.

  According to an embodiment of the present invention, a management device includes a receiving unit, a generating unit, a machine learning unit, a communication parameter determining unit, and a transmitting unit. The receiving means receives from the primary user who is a licensed wireless communication system an interference notification including the position of the interfered terminal, which is the position of the terminal receiving the interference, and the presence or absence of interference. The generating means includes a label indicating the presence or absence of interference based on the communication history of the secondary user and the history of the interference notification, which is a wireless communication system that performs wireless communication using the frequency band of the primary user, and the position of the interfered terminal. And teacher data in which the teacher data and the communication parameters are associated with each other. The machine learning means converts the communication parameters in an n-dimensional space composed of n (n is an integer of 2 or more) including communication parameters based on the teacher data into a first class having no interference and a second class having interference with the second class. A decision boundary, which is a boundary for classifying into a class and a boundary existing at an equal distance from the first and second classes, is determined by machine learning. The communication parameter determining means sets a (nm) -dimensional hyperplane in which m parameters (m is an integer satisfying 1 ≦ m <n) that cannot be changed among the n parameters are fixed, and the setting is performed ( A point on the (nm) -dimensional hyperplane that maximizes or minimizes the gain function is searched using the distance between the point on the (nm) -dimensional hyperplane and the decision boundary as an index, and the searched (nm) From the points on the dimensional hyperplane, a communication parameter set that does not cause interference in the primary user is determined using the constraint conditions. The transmission means transmits the determined communication parameter set to the terminal group of the secondary user.

  According to the embodiment of the present invention, a program for causing a computer to execute is a program for causing a computer to determine a communication parameter set for performing wireless communication, and the receiving means is licensed. A first step of receiving, from a primary user who is a wireless communication system, an interference notification including the position of an interfered terminal, which is the position of a terminal that has received interference, and the presence or absence of interference; Based on the communication history of the secondary user and the history of interference notification, which is a wireless communication system that performs wireless communication using the user's frequency band, the label indicating the presence or absence of interference, the position of the interfered terminal, and the communication parameter are exchanged. And a second step of generating teacher data associated with the n. (N is an integer of 2 or more) parameters including communication parameters based on the teacher data. Boundary for classifying communication parameters into a first class having no interference and a second class having interference in an n-dimensional space composed of data and existing at the same distance from the first and second classes. A third step of determining a decision boundary, which is a boundary, by machine learning, and the communication parameter determining unit fixes m (m is an integer satisfying 1 ≦ m <n) unchangeable parameters out of the n parameters (Nm) -dimensional hyperplane is set, and the (nm) dimension that maximizes or minimizes the gain function using the distance between a point on the set (nm) -dimensional hyperplane and the decision boundary as an index. A fourth step of searching for a point on the hyperplane and determining a communication parameter set that does not cause interference in the primary user using constraints from the searched point on the (nm) -dimensional hyperplane; Transmission means, Is a program for executing a fifth step of transmitting the constant communication parameter set to a secondary user of the terminal groups on the computer.

  In the embodiment of the present invention, a communication parameter set that does not interfere with the terminal of the primary user using history information of communication parameters of the secondary user and history information of interference notification of the primary user Is determined, a communication parameter set that does not interfere with the terminal of the primary user can be determined in real time.

1 is a schematic diagram illustrating a communication device according to an embodiment of the present invention. FIG. 2 is a schematic diagram illustrating a configuration of a management device illustrated in FIG. 1. It is a conceptual diagram of a management table. It is a conceptual diagram of teacher data. It is a conceptual diagram of a decision boundary. FIG. 9 is a diagram for explaining a method for obtaining a decision boundary. FIG. 9 is a diagram for explaining a method for obtaining a decision boundary. FIG. 4 is a diagram for explaining a method of determining a communication parameter set using a determination boundary. It is a flowchart for demonstrating operation | movement of a terminal of a primary user, and a management apparatus. 10 is a flowchart illustrating a detailed operation of step S8 in FIG. 9. It is a figure which shows the parameter used for evaluation at the time of performing wireless communication using the communication parameter determined by the communication parameter set determination method by embodiment of this invention. It is a figure showing the relation between an actual measurement value and a prediction value.

  Embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding portions have the same reference characters allotted, and description thereof will not be repeated.

  FIG. 1 is a schematic diagram showing a communication device according to an embodiment of the present invention. Referring to FIG. 1, there are wireless communication systems WLC1 and WLC2. The wireless communication system WLC1 is a licensed wireless communication system and is called a “primary user”. The wireless communication system WLC1 includes a wireless station 1 and terminals 2 and 3. The wireless station 1 is a primary user's wireless station, and the terminals 2 and 3 are primary user terminals. Then, the wireless station 1 and the terminals 2 and 3 perform wireless communication with each other in the licensed frequency band.

  The wireless communication system WLC2 is a wireless communication system that performs wireless communication in a frequency band of a primary user, and is called a “secondary user”. The wireless communication system WLC2 includes a wireless station 11 and terminals 12 and 13. The wireless station 11 is a secondary user's wireless station, and the terminals 12 and 13 are secondary user's terminals. The wireless station 11 and the terminals 12 and 13 perform wireless communication with each other in the frequency band of the primary user.

  Each of the terminals 2 and 3 detects a received signal strength when receiving a signal from the terminal 12 or 13. Each of the terminals 2 and 3 determines that there is interference when the received signal strength is equal to or higher than 83 dBm, and determines that there is no interference when the received signal strength is smaller than 83 dBm.

  When each of the terminals 2 and 3 determines that there is interference, for example, it detects its own position using GPS (Global Positioning System) and uses an internal timer to determine the time when it is determined that there is interference. Is detected. Since the detected position is the position when the terminals 2 and 3 receive the interference, it is called “interfered terminal position”.

  Then, each of the terminals 2 and 3 generates an interference notification including the time, the presence / absence of interference, and the position of the interfered terminal, and transmits the generated interference notification to the wireless station 1.

  The wireless station 1 receives the interference notification from the terminals 2 and 3, and transmits the received interference notification to the wireless station 11.

  The management device 10 is located in the wireless station 11. Then, the management device 10 receives the interference notification from the wireless station 1. In addition, the management device 10 performs wireless communication with the terminals 12 and 13, and stores the time of the wireless communication and the history of the wireless communication (communication history of the secondary user) in association with each other.

  Then, the management device 10 extracts the time, the presence / absence of interference, and the position of the interfered terminal from the interference notification, and stores the extracted time, the presence / absence of interference, and the position of the interfered terminal in a management table described below in association with each other. At the same time, the stored time and the communication history of the secondary user are stored in a management table to be described later in association with each other.

  The management device 10 repeats the above operation, and sequentially stores the presence or absence of interference, the position of the interfered terminal, and the communication history of the secondary user in the management table in association with each time.

  The management device 10 creates teacher data by a method described later based on the management table, determines a communication parameter set that does not interfere with the primary user based on the created teacher data, and determines the communication parameter set. The communication parameter set is transmitted to the terminals 12 and 13.

  Upon receiving the communication parameter set from the management device 10, the terminals 12 and 13 perform wireless communication with the wireless station 11 using the received communication parameter set.

  FIG. 2 is a schematic diagram showing the configuration of the management device 10 shown in FIG. Referring to FIG. 2, management device 10 includes an information management unit 101, a machine learning unit 102, a parameter determination unit 103, and a parameter notification unit 104.

  The information management unit 101 has a built-in timer (not shown) and receives an interference notification from the wireless station 1.

  The information management unit 101 also manages communication parameters (communication parameters for secondary users) used for wireless communication with the terminals 12 and 13 in association with time. The communication parameters are, for example, transmission power, carrier frequency, bandwidth, modulation scheme, multiple access scheme, antenna sector, and the like.

  Then, the information management unit 101 stores the presence / absence of interference and the position of the interfered terminal included in the interference notification in the management table in association with the time, and associates the communication parameters in the secondary user with the time in the management table. To be stored. That is, the information management unit 101 manages the history of the interference notification (the history of the presence or absence of interference and the position of the interfered terminal) and the communication history of the secondary user (the history of the communication parameters of the secondary user) in chronological order. . A record including the history of the interference notification and the communication history of the secondary user in a time series is called a “management table”.

  When newly receiving the interference notification, the information management unit 101 updates the management table by adding the interference notification history and the communication history of the secondary user to the management table. Then, the information management unit 101 outputs the updated management table to the machine learning unit 102.

  The machine learning unit 102 receives the management table from the information management unit 101, creates teacher data based on the received management table by a method described later, performs machine learning based on the created teacher data, and performs communication. A decision boundary which is a boundary for classifying the parameters into a class CL1 having no interference and a class CL2 having interference and which is a boundary equidistant from both the classes CL1 and CL2 is determined.

  Then, the machine learning unit 102 outputs the determined decision boundary to the parameter determining unit 103.

  The parameter decision unit 103 receives the decision boundary from the machine learning unit 102 and uses the received decision boundary to determine a communication parameter set that does not cause interference in the primary user by a method described later. Then, parameter determination section 103 outputs the determined communication parameter set to parameter notification section 104.

  The parameter notification unit 104 receives the communication parameter set from the parameter determination unit 103, and transmits the received communication parameter set to the terminals 12, 13.

  FIG. 3 is a conceptual diagram of the management table. Referring to FIG. 3, management table Table-CTL includes a history of interference notification of the primary user and a communication history of the secondary user. The history of the interference notification of the primary user and the communication history of the secondary user are associated with each other.

  The history of the primary user's interference notification includes time 1, the presence or absence of interference, and the position of the interfered terminal. The time 1, the presence / absence of interference, and the position of the interfered terminal are associated with each other.

  The communication history of the secondary user includes time 2, transmission power, and center frequency. Time 2, transmission power and center frequency are associated with each other.

  It has been described above that the communication parameters of the secondary user include transmission power, carrier frequency, bandwidth, modulation scheme, multiple access scheme, antenna sector, and the like. However, in FIG. 3, only the transmission power and the center frequency are shown as communication parameters of the secondary user for ease of explanation. Actually, the communication history of the secondary user includes the above-described transmission power, carrier frequency, bandwidth, modulation scheme, multiple access scheme, antenna sector, and the like.

  In FIG. 3, at times t0, t3, and t4, there is no interference, and the transmission power and center frequency of the secondary user at that time are shown. Since there is no interference, "unknown" is stored in the position of the terminal to be interfered. Then, “no interference”, “unknown”, the transmission power and the center frequency of the secondary user are associated with times t0, t3, and t4.

  In FIG. 3, at times t1 and t2, there is interference, and the position of the interfered terminal and the transmission power and center frequency of the secondary user at that time are shown. Then, “interference present”, the position of the interfered terminal ((x1, y1), (x2, y2)), the transmission power of the secondary user and the center frequency of the secondary user are associated with the times t1 and t2. I have.

  As described above, the management table Table-CTL has a configuration in which the presence / absence of interference of the primary user, the position of the interfered terminal, and the communication parameters of the secondary user are associated with the time.

  When receiving the interference notification from the wireless station 1, the information management unit 101 extracts the time 1, the presence (or absence) of interference, and the position of the interfered terminal from the interference notification. Then, the information management unit 101 stores the time 1, the presence (or absence) of interference, and the position of the interfered terminal in the management table Table-CTL in association with each other, and stores the time 2, which is held by itself, and the secondary usage. The transmission power of the user and the center frequency of the secondary user are stored in the management table Table-CTL in association with each other.

  The information management unit 101 performs this operation every time an interference notification is received, and creates and stores a management table Table-CTL. Then, when receiving the new interference notification, the information management unit 101 performs the above operation to update the management table Table-CTL, stores the updated management table Table-CTL, and outputs the updated management table Table-CTL to the machine learning unit 102. .

  FIG. 4 is a conceptual diagram of the teacher data. Referring to FIG. 4, teacher data TCHR_D includes the presence or absence of interference, the position of the interfered terminal, the transmission power, and the center frequency. The presence / absence of interference, the position of the interfered terminal, the transmission power, and the center frequency are associated with each other.

  In FIG. 4, the transmission power and the center frequency are shown as communication parameters of the secondary user for ease of explanation, but in actuality, the transmission power and the center frequency are used as the communication parameters of the secondary user. , Carrier frequency, bandwidth, modulation scheme, multiple access scheme, antenna sector, and the like are stored in the teacher data TCHR_D.

  In the teacher data TCHR_D, the presence of interference is represented by “1”, and the absence of interference is represented by “−1”.

  In the teacher data TCHR_D, when there is no interference, the position of the interfered terminal is represented by (0, 0).

  The presence or absence of interference (-1 or 1), the position of the interfered terminal ((0, 0) or (x1, y1), etc.), the transmission power (7 dBm, 10 dBm, etc.) and the center frequency (2.42 GHz, etc.) They are associated with each other.

  Upon receiving the management table Table-CTL from the information management unit 101, the machine learning unit 102 refers to time 1 and time 2 of the received management table Table-CTL and refers to the interference Presence / absence, interfered terminal position, transmission power, and center frequency are extracted, and teacher data TCHR_D is created by associating the extracted presence / absence of interference, interfered terminal position, transmission power, and center frequency with each other. In this case, teacher data TCHR_D is created by indicating the presence of interference by "1", the absence of interference by "-1", and the position of the interfered terminal when there is no interference by (0, 0).

  As a result, the machine learning unit 102 creates the teacher data TCHR_D shown in FIG.

  When determining the transmission power of the secondary user in which no interference occurs in the primary user, learning is performed using the teacher data TCHR_D, and the communication parameters need to be classified into a class CL1 having no interference and a class CL2 having interference. There is.

  For example, a support vector machine (SVM: Support Vector Machine) is used as a machine learning device suitable for classifying communication parameters into two classes CL1 and CL2.

  The SVM is one of the pattern recognition models using supervised learning, and learns a boundary that boldly separates a learning sample (teacher data). In order to learn this boundary, an n-dimensional space composed of n (n is an integer of 2 or more) parameters is set.

  Then, as a result of learning using the learning sample (teacher data) arranged in the n-dimensional space, the boundary that boldly separates the learning sample (teacher data) from the sample (data) closest to the boundary is determined. The distance (margin) is determined to be maximum.

  A feature of SVM is to find a hyperplane with the largest margin among many candidate planes that can separate a plurality of points in n-dimensional space.

  The margin refers to a minimum value of a distance from the hyperplane to a plurality of points. If it is attempted to classify a plurality of points into two classes CL1 and CL2 while maximizing the margin, some points belonging to the class CL1 end up. The hyperplane must be positioned so that the minimum value of the distances to the points in the class CL2 and the minimum value of the distances to some points belonging to the class CL2 are equal.

  Such a hyperplane is a hyperplane having a maximum margin, and is referred to as a “decision boundary” in the embodiment of the present invention.

  FIG. 5 is a conceptual diagram of a decision boundary. FIG. 5 shows a two-dimensional space composed of two parameters for easy understanding.

  When determining the transmission power of the secondary user in which no interference occurs in the primary user, the presence or absence of interference and the transmission power are selected from the teacher data. Further, as a parameter that cannot be changed (cannot be changed), for example, a distance to an interference source is selected.

  Using the transmission power and the distance to the interference source as parameters, the presence of interference is represented by a black circle, and the absence of interference is represented by a white circle, as shown in FIG.

  Then, in the two-dimensional space, a decision boundary BD that separates a plurality of black circles and a plurality of white circles is determined by a method described below. In FIG. 5, since the n-dimensional space is a two-dimensional space, the above-described hyperplane having the maximum margin is a straight line.

  6 and 7 are diagrams for explaining a method for obtaining a decision boundary.

  The identification line for identifying the group without interference (the group with white circles) and the group with interference (the group with black circles) with the maximum margin is unknown, but it is unknown where the maximum margin is generated. It is assumed that there is data of the end of each group of the group of white circles without interference (the group of black circles with interference) and the negative example (the group of black circles with interference). This end data is called a support vector.

  Let two lines passing through the support vector be ax + by + c = 1 and ax + by + c = −1. Since the two lines are parallel, the coefficients are the same.

  The area indicated by ax + by + c ≦ 1 is the range where the data of the positive example (the group of white circles without interference) exists, and the area indicated by ax + by + c ≧ −1 is the data of the negative example (the group of black circles with interference). Is in the range where (see FIG. 6).

  Here, two inequalities, which are conditions necessary for drawing an identification line, have been obtained, but these are put together. That is, in the case of the positive example, the inequality expression is multiplied by "1" to obtain ax + by + c≤1, and in the case of the negative example, the inequality expression is multiplied by "-1" to obtain -1 (ax + by + c) ≥-1.

  Then, when a variable t (t = 1 or t = −1) representing a positive example or a negative example is used, t (ax + by + c) ≦ 1.

  The distance d of the perpendicular drawn from the support vector of each of the positive example and the negative example to the identification line is represented by the following equation.

その結果、教師データの全てについて、ｔ（ａｘ＋ｂｙ＋ｃ）≦１の条件を満たしながら、ｄの分母が最小になるａ，ｂ，ｃを求めることにより、最大マージンに基づく識別線の位置を決定できる（図７参照）。この最大マージンに基づく識別線の位置を決定することは、正例（干渉無しの白丸のグループ）と負例（干渉有りの黒丸のグループ）との両方から等距離の位置を決定することに相当する。

As a result, the position of the identification line based on the maximum margin can be determined by obtaining a, b, and c that minimize the denominator of d while satisfying the condition of t (ax + by + c) ≦ 1 for all the teacher data ( (See FIG. 7). Determining the position of the identification line based on this maximum margin is equivalent to determining the position at the same distance from both the positive example (the group of white circles without interference) and the negative example (the group of black circles with interference). I do.

When there are i (i = 1, 2, 3, 4,...) Teacher data, t ₁ (ax ₁ + by ₁ + c) ≦ 1, t ₂ (ax ₂ + by ₂ + c) ≦ 1,. Then, a, b, and c that give the minimum value of a ² + b ² are obtained by using i inequalities of t _i (ax _i + by _i + c) ≦ 1.

The problem of finding a parameter for determining an identification line based on the maximum margin is as follows. When two variables are x and y and N are teacher data, t _i (ax _i + by _i + c) ≦ 1 (i = 1 to N ) Is a constrained minimization problem for finding a, b, c that minimizes a ² + b ² .

A convenient way to solve such a constrained minimization problem is the Lagrange's undetermined multiplier (coefficient) method. The ^a 2 + ^{b 2} to minimize the ^{f (a, b) = a} 2 + b 2, constraint _{t i (ax i + by i} + c) ≦ 1 and _{g (x, y) = t} i (ax i + by i + c If) -1 ≦ 0, the following equation (2) can be converted to equation (3) by applying the Lagrange undetermined multiplier method.

式（２），（３）において、ｓ．ｔ．は、「ｓ．ｔ．の右側に記載された条件に従うこと」を意味する（以下、同じ）。

In equations (2) and (3), s. t. Means "to obey the conditions described on the right side of st" (hereinafter the same).

Equation (2) _{is, g 1 (a, b,} c) ≦ 0, g 2 (a, b, c) ≦ 0, ···, g n (a, b, c) ≦ 0 according f (a, This is the main problem because it means to find a and b that minimize b).

Equations (3a) and (3b) in equation (3) are the results of applying the Lagrange multiplier method. Equation (3b) represents a, b, c that minimizes L (a, b, c, μ ₁ ,..., Μ _n ). φ is a symbol indicating one function, similarly to f.

Equation (3c), (3d) is for showing the dual _{problem, φ (μ 1, ···,} μ n) μ 1 to maximize, ..., by obtaining the mu _n, the main problem It is guaranteed that the minimum values a, b, and c can also be obtained.

In the following, for the sake of simplicity, a main problem is created with two teacher data, and a function L is created by applying the Lagrange multiplier method. However, a ² + b ² is multiplied by 上 for convenience of calculation.

t _i (ax _i + by _i + c) ≦ 1 is transformed to t _i (ax _i + by _i + c) −1 ≦ 0. As a result, the following equation is obtained.

最大化すべき双対問題の関数を作るために、Ｌをａ，ｂ，ｃについて最小化する。従って、Ｌをａ，ｂ，ｃについて偏微分し、その結果を０と置く。そうすると、次式が得られる。

To create a dual problem function to be maximized, L is minimized for a, b, and c. Therefore, L is partially differentiated with respect to a, b, and c, and the result is set to 0. Then, the following equation is obtained.

そして、式（５）の上側の２つの式から次式が得られる。

Then, the following expression is obtained from the upper two expressions of Expression (5).

式（５），（６）において、ｔ_ｉは、ｉ番目の教師データの正負であり、ｘ_ｉ，ｙ_ｉは、ｉ番目の教師データのｘ座標およびｙ座標であるので、ｔ_ｉ，ｘ_ｉ，ｙ_ｉは、既知である。

Equation (5), in (6), _{t i} is the positive and negative i-th teacher _data, x i, _{y i} is because the x and y coordinates of the i-th teacher _{data, t} i, x _i and y _i are known.

Therefore, if μ _i that maximizes the function of the dual problem is known, a and b can be obtained from Expression (6).

  Then, in order to create a dual problem, when all the conditions (formula (5)) obtained by partial differentiation are substituted into L, the following formula is obtained.

その結果、双対問題は、次式にようになる。

As a result, the dual problem becomes

双対問題の関数φ（μ_１，μ_２）の最大化は、凸２次計画問題の解法である内点法（非特許文献５）等で求めることができる。解を求めるスピードを重視する場合、制約条件を考慮しながら、タブー探索法（非特許文献６）等の進化型計算を用いてもよい。

The function φ (μ ₁ , μ ₂ ) of the dual problem can be maximized by an interior point method (Non-Patent Document 5) or the like which is a solution of the convex quadratic programming problem. When emphasizing the speed of finding a solution, an evolutionary calculation such as a tabu search method (Non-Patent Document 6) may be used while considering the constraint conditions.

  The interior point method is an iterative iterative method for generating a point sequence converging on an optimal solution inside a constraint region of an optimization problem.

  The taboo search method is a type of local search method that searches for a certain solution in the vicinity of the currently obtained solution, and prevents the recently searched solution from being searched for a while as a taboo to prevent convergence to the local solution. This is to find the optimal solution by performing a wide area search of the solution space.

Accordingly, points that increase the function φ (μ ₁ , μ ₂ ) of the equation (8) are sequentially generated in the region of the constraint condition (the equation described on the right side of the _st ) of the equation (8), and the function φ (μ _1, μ ₂₎ optimal solution (μ _1, μ ₂₎ to maximize the seek.

When importance is placed on the speed of finding a solution, an optimal search for maximizing the function φ (μ ₁ , μ ₂ ) by performing a wide area search while preventing convergence to a local solution in consideration of the constraint condition of Expression (8). Find the solution (μ ₁ , μ ₂ ).

If μ ₁ and μ ₂ can be obtained, a and b can be obtained from equation (6) using known t ₁ , t ₂ , x ₁ , x ₂ , y ₁ , and y ₂ .

And, c is the x coordinate of a support vector and _{x S,} when the y-coordinate and _{y S,} obtained from _{ax S + by S + c =} 1 ( or _{ax S + by S + c =} -1).

  As a result, a decision boundary BD for identifying the classes CL1 and CL2 with the maximum margin is determined.

In the above, the case where the teacher data is two-dimensional represented by x and y has been described. However, when the teacher data has n dimensions, the x-axis is x ₁ , the y-axis is x ₂ , the z-axis is x ₃ , ..., the n-dimensional axis is represented by _xn .

Also, coefficients such as a and b are represented by w ₁ , w ₂ ,..., W _n corresponding to n dimensions.

Since (x ₁ , x ₂ ,..., X _n ) and (w ₁ , w ₂ ,..., W _n ) are vectors, X = (x ₁ , x ₂ ,. _n ), W = (w ₁ , w ₂ ,..., w _n ).

In ^addition, a ² + ^b 2 _is a _{^{w 1 2 + w 2 2 +}} ··· + w n 2. The square root of w ₁ ² + w ₂ ² +... + W _n ² is referred to as a (Euclidean) norm.

x _i, where are added over the same dimensions to each other, such as x _j is expressed by X ^{T X} in the inner product of vectors. X ^T means the transposed matrix of X.

  Further, when the number of the teacher data is n, “2” becomes “n”.

  Therefore, when the teacher data has n dimensions, the identification line (= decision boundary BD) having the maximum margin is determined using the following equations (9) to (16).

機械学習部１０２は、上述した方法によって、決定境界ＢＤを決定し、その決定した決定境界をパラメータ決定部１０３へ出力する。

The machine learning unit 102 determines the determination boundary BD by the above-described method, and outputs the determined determination boundary to the parameter determination unit 103.

  FIG. 8 is a diagram for explaining a method for determining a communication parameter set using the determination boundary BD.

  An (nm) -dimensional hyperplane in which m parameters (m is an integer satisfying 1 ≦ m <n) that cannot be changed, such as an interference occurrence position, are fixed for n parameters. At this time, a point on the (nm) -dimensional hyperplane represents a parameter set. This is called a parameter setting surface.

  An interference allowable threshold and a (n-1) -dimensional hyperplane (lowest communication condition plane) are set as thresholds for providing a parameter setting range. The specific determination method is as follows.

[Minimum communication conditions]
Smoothly connected parameter sets of the minimum value that satisfy the desired signal-to-noise ratio at the receiving station of the secondary user are defined as the minimum communication condition plane. If the parameter set is closer to the decision boundary than the minimum communication condition, it is guaranteed that the secondary user can communicate without giving any interference to the primary user.

[Interference tolerance threshold]
The interference tolerance threshold indicates allowable interference. The minimum distance between the observation point (point included in the teacher data) and the decision boundary BD is k times (k is a real number satisfying 0 <k <1) times the minimum value from the decision boundary BD. Let the value be the interference tolerance threshold. k is, for example, 0.5.

  Then, a point for maximizing or minimizing a certain gain function is searched on the parameter setting surface. At this time, the following constraints are set as the constraints.

  (1) The point for maximizing or minimizing the gain function exists on the side where no interference exists with respect to the decision boundary BD, and exists on the decision boundary BD side from the minimum communication condition plane.

  (2) Further, the distance d between the point where the gain function is maximized or minimized and the decision boundary BD is in a range satisfying the interference allowable threshold or more.

  For example, the following is assumed as the gain function.

(I) Channel capacity of terminal of secondary user Channel amount C is determined by C = Blog (1 + SINR). B is the bandwidth and depends on the frequency bandwidth used. SINR is a signal-to-noise ratio and is determined by transmission power, a frequency band, and the like.

(Ii) Communication distance of the terminal of the secondary user The communication distance is determined depending on the used frequency band, transmission power, and the like.

(Iii) Bit error rate Determined depending on the transmission power, modulation scheme, frequency bandwidth, and the like.

  With reference to FIG. 8, a method for determining a communication parameter set will be specifically described. In FIG. 8, the distance from the interference source is used as a parameter that cannot be changed. The distance from the interference source is determined as the distance between the position of the interfered terminal included in the teacher data and the position of the terminal of the secondary user that caused the interference.

  Then, the determined distance from the interference source is fixed, and the parameter setting plane is set in a two-dimensional space. After that, the above-mentioned minimum communication condition plane and interference allowable threshold are set in a two-dimensional space.

  In FIG. 8, since the teacher data is two-dimensional, the parameter setting plane, the minimum communication condition plane, and the interference allowable threshold are straight lines.

  Then, on the parameter setting plane (straight line), a point where the gain function becomes maximum or minimum is searched, and a point that satisfies the above-described constraints (1) and (2) is set as a candidate point from the searched points. Extract.

  When there are a plurality of candidate points, an arbitrary point among the plurality of candidate points is determined as a communication parameter set.

  On the other hand, when there is one candidate point, the one candidate point is determined as a communication parameter set.

  The parameter determining unit 103 determines a communication parameter set for the secondary user's terminal to perform wireless communication without causing interference to the primary user's terminal by using the determination boundary BD by the above-described method.

  Then, parameter determination section 103 outputs the determined communication parameter set to parameter notification section 104.

  The parameters that can be changed include, for example, the transmission power of the base station or terminal of the secondary user, the carrier frequency, the bandwidth, the modulation scheme, the multiple access scheme, the antenna sector, and the like.

  The parameters that cannot be changed include, for example, the position of the terminal of the primary user, the distance to the terminal of the primary user, the modulation method of the primary user, the multiple access method of the primary user, and the like. .

  When the teacher data has three or more dimensions, the decision boundary BD, the minimum communication condition plane, and the interference allowable threshold are planes.

  Therefore, the minimum communication conditions described in the claims consist of (n-1) -dimensional lines or planes.

  FIG. 9 is a flowchart for explaining operations of the primary user terminals 2 and 3 and the management device 10.

  Referring to FIG. 9, the terminal of the primary user (at least one of terminals 2 and 3) detects the interference (step S1) and transmits an interference notification to wireless station 1. Then, the wireless station 1 transmits the interference notification received from the terminal (at least one of the terminals 2 and 3) to the management device 10 (Step S2).

  On the other hand, when there is no interference, the information management unit 101 of the management apparatus 10 generates wireless communication between the wireless station 11 and the terminals 12 and 13 (step S3), and records the communication history between the secondary user and the terminals 12 and 13. Is stored (step S4).

  Then, the information management unit 101 receives the interference notification from the wireless station 1 (step S5), extracts the interference information and the communication history, and creates a management table Table-CTL (step S6). Then, the information management unit 101 outputs the management table Table-CTL to the machine learning unit 102.

  The machine learning unit 102 receives the management table Table-CTL from the information management unit 101, checks the interference information included in the received management table Table-CTL with the communication history and time, and creates teacher data by the above-described method. (Step S7).

  After that, the machine learning unit 102 inputs the teacher data to the machine learning device, and determines or updates the decision boundary BD using the SVM by the above-described method (step S8).

  Then, the machine learning unit 102 outputs the determined decision boundary BD to the parameter determining unit 103.

  Upon receiving the determination boundary BD, the parameter determination unit 103 determines a communication parameter set for preventing interference with the primary user terminal by using the received determination boundary BD by the above-described method (step S9). ), And outputs the determined communication parameter set to the parameter notification unit 104.

  Then, the parameter notification unit 104 transmits the communication parameter set to the terminals 12 and 13 of the secondary user (Step S10).

  The terminals 12 and 13 receive the communication parameter set, and update the parameters used for wireless communication using the received communication parameter set (step S11).

  Thus, a series of operations ends.

  Note that the decision boundary BD is determined or updated in step S8 because the flowchart shown in FIG. 9 is executed every time the primary user terminals 2 and 3 detect interference. This is because, when the processing is executed for the second time or later, the teacher data has also been updated, and the already determined decision boundary BD is updated using the updated teacher data.

  FIG. 10 is a flowchart for explaining the detailed operation of step S8 in FIG.

  Referring to FIG. 10, after step S7 in FIG. 9, machine learning unit 102 determines an objective function and a constraint condition shown in equation (10) (step S81).

  Then, as shown in Expression (11), the machine learning unit 102 converts the main problem into a dual problem by applying the Lagrange's undetermined multiplier method to the objective function and the constraint condition (step S82).

After that, the machine learning unit 102 solves the dual problem using Equations (12) to (16), and obtains μ _i that maximizes the function of the dual problem (Step S83).

Subsequently, the machine learning unit 102 obtains w ₁ , w ₂ ,..., W _n that gives the minimum value of the main problem using the teacher data and μ _i (step S84). More specifically, the machine learning unit 102, _{w 1, w} 2 known _t i indicated by the teacher _data, and x i1 _{~x in,} obtained a mu _i into Equation (14), - ..., seek _{w n.}

After step S84, the machine learning unit 102 determines that x ₁ coordinate, x ₂ coordinate,..., X _n coordinate of a certain support vector are x _1S , x _{2S ,.} Request _{1 x 1S + w 2 x 2S} + ··· + w n x nS + c = 1 ( or _{w 1 x 1S + w 2 x} 2S + ··· + w n x nS + c = -1) from the intercept c. Then, the machine learning unit 102 substitutes the obtained w ₁ , w ₂ ,..., W _n and c into w ₁ x ₁ + w ₂ x ₂ +... + W _n x _n + c = 0. A plane (or line) having the maximum margin (distance d) with both the positive and negative examples is determined, and the determined plane (or line) is determined as a determination boundary (step S85).

  Thereafter, a series of operations proceeds to step S9 in FIG.

  As described above, the management device 10 uses the history information of the communication parameters in the data transmission of the secondary user and the history information in which the presence or absence of the interference of the primary user in the data transmission is determined. A communication parameter set that does not cause interference to the terminal of the primary user is determined (see S6 to S9). No communication parameter set can be determined in real time.

  In addition, the management apparatus 10 receives from the primary user only an interference notification including the presence or absence of interference, the position of the interfered terminal, and the time when the interference is detected, and gives interference to the terminal of the primary user. Since no communication parameter set is determined, an appropriate communication parameter set can be set even if the setting of the primary user's wireless communication is unknown.

  Furthermore, since the flowcharts shown in FIGS. 9 and 10 are executed each time interference is detected in the terminal of the primary user, interference with the terminal of the primary user is caused in response to a real-time change in the interference situation. A communication parameter set that is not given can be set.

  Furthermore, if the learning of the decision boundary using the teacher data has been sufficiently performed, the communication parameter set can be determined with a small amount of calculation.

  Furthermore, the risk when interference occurs differs depending on the primary user, but the learned decision boundary is uniquely determined, so that even if the primary user changes, appropriate communication parameters can be set. . In other words, the above-described method for determining the communication parameter set can have robustness against the change of the primary user.

  Japanese Patent Laying-Open No. 2013-221609 (Patent Literature 1) discloses a technique of selecting an appropriate communication parameter according to a peripheral situation of a wireless communication device using a support vector machine (SVM).

  In this technique, a situation, a parameter, and a performance (communication performance) are stored in a history database in association with each other, and a communication parameter that provides good communication performance in a current peripheral situation is referred to by referring to the history database. To choose.

  The surrounding situation (context) is determined using a support vector machine (SVM).

  As described above, in Patent Literature 1, a support vector machine (SVM) is used to determine a surrounding situation (context). That is, in Patent Literature 1, a support vector machine (SVM) is used to classify a surrounding situation (context).

  On the other hand, in the embodiment of the present invention, a support vector machine (SVM) is used to classify communication parameters that cause interference and communication parameters that do not cause interference.

  Therefore, the object to be classified using the support vector machine (SVM) is completely different between the embodiment of the present invention and Patent Literature 1, and the patent document 1 classifies the communication parameters using the support vector machine (SVM). There is no suggestion to do.

  Also, in Patent Document 1, communication parameters between wireless communication devices that perform wireless communication with each other are selected. However, in the embodiment of the present invention, a secondary user is a primary user who is not a partner of wireless communication. Is to determine a communication parameter that does not cause interference.

  Therefore, the determination of the communication parameters in the embodiment of the present invention is completely different from the selection of the communication parameters in Patent Document 1.

  FIG. 11 is a diagram illustrating parameters used for evaluation when wireless communication is performed using the communication parameters determined by the communication parameter set determining method according to the embodiment of the present invention.

  The evaluation was performed as follows.

(I) The receiving terminal of the primary user and the transmitting terminal of the secondary user are arranged at a distance Ld apart from each other. (Ii) The transmitting terminal of the secondary user transmits a signal while changing the transmission power.
When creating the teacher data, the transmission power was randomly selected from the range of -10 dBm to 30 dBm.

(Iii) Calculate the reception power of the interference wave at the receiving terminal of the primary user from the theoretical formula of free space propagation
The interference condition was that there was interference when the received power at a position separated by the distance Ld was -83 dBm or more, and that there was no interference when the received power was less than -83 dBm.

  FIG. 12 is a diagram illustrating the relationship between the actually measured value and the predicted value. In FIG. 12, the predicted value is a value predicted using a support vector machine.

  Referring to FIG. 12, the number of coincidences between the actually measured value and the predicted value is 39, and the number of mismatches between the actually measured value and the predicted value is 4, when there is interference. Also, with no interference, the number of matches between the measured value and the predicted value is 37, and the number of mismatches between the measured value and the predicted value is eight.

  Since the number of test signal points is 90 (see FIG. 11), the correct answer rate when predicted using the support vector machine is (39 + 37) /90=0.844=84.4%.

  Therefore, the prediction accuracy by the support vector machine was found to be high.

  The average transmission power gain of the secondary user when transmitting a signal using the communication parameter set determined by the above-described method was 3.9 dB, and the transmission power of the secondary user could be improved.

  As described above, by determining the communication parameter set of the secondary user by the method for determining the communication parameter set according to the embodiment of the present invention, the average transmission power gain can be improved without causing interference to the primary user. I understood.

  Note that the average transmission power gain was obtained by obtaining the distance between the decision boundary BD and the sample point on the parameter setting surface as the transmission power gain, and averaging the obtained transmission power gains.

  In the embodiment of the present invention, the functions of the information management unit 101, the machine learning unit 102, the parameter determination unit 103, and the parameter notification unit 104 may be realized by a program PROG.

  In this case, the program PROG includes the above-described steps S3 to S10 (including steps S81 to S85).

  The management device 10 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The ROM stores the program PROG.

  When determining the communication parameter set, the CPU reads out the program PROG from the ROM and executes it. Then, the CPU stores the history information of the communication parameters of the secondary user in the RAM, and stores various calculation results when determining the communication parameter set in the RAM.

  Therefore, the program PROG is a program for causing a computer (CPU) to determine a communication parameter set for performing wireless communication.

  In the embodiment of the present invention, the program PROG may be provided by being recorded on a recording medium such as a CD or a DVD. The user sets a recording medium on which the program PROG is recorded in a computer, and the computer (CPU) reads out the program PROG from the recording medium and executes the program.

  Therefore, the recording medium on which the program PROG is recorded is a recording medium readable by a computer (CPU).

  In the above description, it has been described that the machine learning unit 102 determines the decision boundary BD using the support vector machine (SVM). However, in the embodiment of the present invention, the machine learning unit 102 is not limited to this. The decision boundary BD may be determined using regression (Non-Patent Document 7) or AROW (Adaptive Regulation of Weight Vectors) (Non-Patent Document 8). Any machine learning device may be used to determine the decision boundary BD.

  In the above description, the transmission power that does not interfere with the primary user terminals 2 and 3 has been described. However, the embodiment of the present invention is not limited to this, and it is not limited to this. , 3 may be determined. Generally, any communication parameter set that does not interfere with the primary user terminals 2 and 3 may be determined. You may.

  Further, in the above description, the parameter that cannot be changed is described as “distance from interference source”. However, in the embodiment of the present invention, the parameter is not limited to this and other parameters than “distance from interference source” May be a parameter that cannot be changed.

  Further, in the above description, it has been described that the management device 10 is installed in the wireless station 11. However, the embodiment of the present invention is not limited to this, and the management device 10 may be located at a different position from the wireless station 11. It may be. In this case, the management device 10 may receive the interference notification from the wireless station 11, may receive the interference notification from the wireless station 1, or may directly receive the interference notification from the terminals 2 and 3.

  Furthermore, although it has been described above that the management device 10 receives the interference notification from the wireless station 1, the embodiment of the present invention is not limited to this. May be received directly.

  Furthermore, in the above, when determining a communication parameter set that does not interfere with the terminals 2 and 3 of the primary user, a minimum communication condition plane and an interference allowable threshold are set, and a minimum communication condition plane is set on the parameter setting plane. It has been described that a point which is closer to the decision boundary BD and whose distance from the decision boundary BD is larger than the distance between the interference allowable threshold value and the decision boundary BD is selected as a candidate point of the communication parameter set. In the embodiment, the point belonging to the class CL1 having no interference with respect to the decision boundary BD on the parameter setting surface without setting the minimum communication condition surface and the interference allowable threshold is set as a candidate point of the communication parameter set. May be selected.

  Even if the secondary user terminals 12 and 13 perform wireless communication using the communication parameter set determined in this way, they do not interfere with the primary user terminals 2 and 3, so that Is achieved.

  Therefore, the management apparatus according to the embodiment of the present invention provides an interference notification from a primary user who is a licensed wireless communication system, including the position of an interfered terminal, which is the position of a terminal receiving interference, and the presence or absence of interference And a label indicating the presence or absence of interference based on the communication history of the secondary user, which is a wireless communication system that performs wireless communication using the frequency band of the primary user, and the history of interference notification. Generating means for generating teacher data in which an interference terminal position and a communication parameter are associated with each other; and an n-dimensional space including n (n is an integer of 2 or more) parameters including the communication parameter based on the teacher data. Determination of a boundary for classifying communication parameters into a first class having no interference and a second class having interference, and a boundary existing equidistant from the first and second classes. A machine learning means for determining a field by machine learning, and a (nm) -dimensional hyperplane in which m unchangeable (m is an integer satisfying 1 ≦ m <n) parameters out of n parameters are set. Then, using the distance between the set point on the (nm) -dimensional hyperplane and the decision boundary as an index, a point on the (nm) -dimensional hyperplane that maximizes or minimizes the gain function is searched for. Means for determining a communication parameter set that does not cause interference in the primary user from the points on the (nm) -dimensional hyperplane using constraints, And transmitting means for transmitting to the terminal group.

  Further, the program for causing a computer to execute according to the embodiment of the present invention is a program for causing a computer to determine a communication parameter set for performing wireless communication, wherein the receiving means is a licensed wireless program. A first step of receiving, from a primary user who is a communication system, an interference notification including a position of an interfered terminal, which is the position of a terminal that has received interference, and the presence or absence of interference; A label indicating the presence or absence of interference, the position of the interfered terminal, and the communication parameters are associated with each other based on the communication history of the secondary user and the history of interference notification, which is a wireless communication system that performs wireless communication using frequency bands A second step of generating teacher data, wherein the machine learning means generates n (n is an integer of 2 or more) parameters including communication parameters based on the teacher data. A boundary for classifying communication parameters into a first class having no interference and a second class having interference in an n-dimensional space consisting of: and a boundary existing at an equal distance from the first and second classes. A third step of determining the decision boundary by machine learning, and the communication parameter determination means fixes m (m is an integer satisfying 1 ≦ m <n) unchangeable parameters out of the n parameters (Nm) -dimensional hyperplane is set, and the (nm) -dimensional hyperplane that maximizes or minimizes the gain function using the distance between the set point on the (nm) -dimensional hyperplane and the decision boundary as an index. A fourth step of searching for a point on the plane, determining a communication parameter set that does not cause interference in the primary user from the searched point on the (nm) -dimensional hyperplane using constraint conditions, and transmitting Means determined A communication parameter set may be a program for executing a fifth step of transmitting to the secondary user terminal groups on the computer.

  Note that, in the embodiment of the present invention, the information management unit 101 that receives the interference notification constitutes a “receiving unit”.

  Further, in the embodiment of the present invention, the machine learning unit 102 that creates teacher data by the above-described method constitutes a “generating unit”.

  Further, in the embodiment of the present invention, the machine learning unit 102 that determines the decision boundary BD by the above-described method constitutes “machine learning means”.

  Furthermore, in the embodiment of the present invention, the parameter determining unit 103 that determines a communication parameter set that does not interfere with the primary user by the above-described method constitutes “parameter determining means”.

  Further, in the embodiment of the present invention, parameter notifying section 104 constitutes a “transmitting means”.

  The embodiments disclosed this time are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description of the embodiments, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  The present invention is applied to a management device, a program to be executed by a computer, and a computer-readable recording medium storing the program.

  1,11 radio station, 2,3,12,13 terminal, 10 management device, 101 information management unit, 102 machine learning unit, 103 parameter determination unit, 104 parameter notification unit.

Claims (11)

Receiving means for receiving, from a primary user who is a licensed wireless communication system, an interference notification including the position of the interfered terminal, which is the position of the interfered terminal, and the presence or absence of interference;
A label indicating the presence or absence of interference based on the communication history of the secondary user and the history of the interference notification, which is a wireless communication system that performs wireless communication using the frequency band of the primary user; Generating means for generating teacher data in which communication parameters are associated with each other;
Based on the teacher data, in the n-dimensional space including n (n is an integer of 2 or more) parameters including the communication parameters, the communication parameters are divided into a first class having no interference and a second class having interference. Machine learning means for determining, by machine learning, a decision boundary, which is a boundary for classifying into, and a boundary existing equidistant from the first and second classes;
A (nm) -dimensional hyperplane in which m parameters (m is an integer satisfying 1 ≦ m <n) that cannot be changed among the n parameters are fixed is set, and the set (nm) is set. A point on the (nm) -dimensional hyperplane that maximizes or minimizes the gain function is searched using the distance between a point on the two-dimensional hyperplane and the decision boundary as an index, and the searched (nm) -dimensional hyperplane is searched. Communication parameter determining means for determining a communication parameter set that does not cause interference in the primary user using a constraint condition from a point on a plane;
A transmission unit that transmits the determined communication parameter set to the terminal group of the secondary user.   2. The communication parameter set, wherein the communication parameter set is determined by using a minimum communication condition under which wireless communication of the secondary user is guaranteed and an interference tolerance threshold indicating allowable interference as the constraint condition. The management device according to item 1.   The communication parameter determining means is located among the searched (nm) -dimensional hyperplanes on the determination boundary side with respect to the minimum communication condition, and the distance from the determination boundary is the interference tolerance. The management device according to claim 2, wherein a point larger than a threshold value is determined as the communication parameter set.   The communication parameter determining means connects a parameter set satisfying a predetermined signal-to-noise ratio at the receiving station of the secondary user and sets the minimum communication condition consisting of an (n-1) -dimensional line or plane. The management device according to claim 2, wherein the communication parameter set is determined using the set minimum communication condition.   The communication parameter determining means may multiply a value obtained by multiplying a minimum value of a distance between the point on the n-dimensional space included in the teacher data and the determination boundary by k (k is a real number satisfying 0 <k <1). The management device according to any one of claims 2 to 4, wherein the management parameter set is set as an interference allowable threshold, and the communication parameter set is determined using the set interference allowable threshold. A program for causing a computer to determine a communication parameter set for performing wireless communication,
A first step of receiving, from a primary user who is a licensed wireless communication system, an interference notification including the position of an interfered terminal, which is the position of the terminal receiving interference, and the presence or absence of interference;
A generation unit configured to generate a label indicating presence or absence of interference based on a communication history of a secondary user, which is a wireless communication system performing wireless communication using the frequency band of the primary user, and a history of the interference notification; A second step of generating teacher data in which the interfering terminal position and the communication parameter are associated with each other;
The machine learning means interferes with the first class having no interference with the communication parameter in an n-dimensional space including n (n is an integer of 2 or more) parameters including the communication parameter, based on the teacher data. A third step of determining, by machine learning, a decision boundary that is a boundary for classification into the second class and that is a boundary existing equidistant from the first and second classes;
The communication parameter determining means sets a (nm) -dimensional hyperplane in which m parameters (m is an integer satisfying 1 ≦ m <n) that cannot be changed among the n parameters are fixed, and the setting is performed. A point on the (nm) -dimensional hyperplane that maximizes or minimizes the gain function is searched using the distance between the point on the (nm) -dimensional hyperplane and the decision boundary as an index, and the searched ( a fourth step of determining a communication parameter set that does not cause interference in the primary user using constraints from points on the (nm) -dimensional hyperplane;
A transmitting unit transmitting the determined communication parameter set to the terminal group of the secondary user.   In the fourth step, the communication parameter determination unit sets the communication parameter set as a restriction condition using a minimum communication condition for guaranteeing the wireless communication of the secondary user and an interference allowable threshold value indicating allowable interference. A program for causing a computer according to claim 6 to execute the following.   In the fourth step, the communication parameter determination unit is located on the determination boundary side of the minimum communication condition among the searched points on the (nm) -dimensional hyperplane, and A program for causing a computer to execute the computer according to claim 7, wherein a point at which a distance from the communication parameter set is greater than the interference allowable threshold is determined as the communication parameter set.   In the fourth step, the communication parameter determining means connects a parameter set satisfying a predetermined signal-to-noise ratio at the receiving station of the secondary user and comprises a (n-1) -dimensional line or surface. The program for causing a computer according to claim 7 or 8 to set a minimum communication condition and determine the communication parameter set using the set minimum communication condition.   In the fourth step, the communication parameter determination unit sets a minimum value of a distance between a point on the n-dimensional space included in the teacher data and the determination boundary to k (k is 0 <k <1). The computer according to any one of claims 7 to 9, wherein a value multiplied by (real number) is set as the interference allowable threshold, and the communication parameter set is determined using the set interference allowable threshold. Program for.   A computer-readable recording medium on which the program according to any one of claims 6 to 10 is recorded.

Images