KR20190031982A - A cognitive radio communication method - Google Patents

Abstract

A cognitive wireless communication method is disclosed. The cognitive wireless communication method comprises the steps of: receiving a signal transmitted from a licensed user by an unlicensed user; estimating a transmission power of the licensed user according to a reception power of the signal received by the unlicensed user, through an artificial neural network algorithm; determining a spectrum occupancy state of the licensed user according to the estimated transmission power, through a support vector machine; and using the spectrum by the unlicensed user when the licensed user is not occupying the spectrum.

Description

A cognitive radio communication method "

The present invention relates to cognitive radio communications, and relates to cognitive radio communications in which a licensee can flexibly use the spectrum of a licensed user depending on the spectrum occupancy status of the licensee.

Currently, the wireless spectrum allocation scheme uses a fixed allocation scheme that allocates spectrum access rights to users who have paid a fee to the government agency (I. Akyildiz, B. Lo, and R. Balakrishnan, "Cooperative Spectrum sensing in cognitive radio networks: A survey, "Physical Communication, vol 4, no. 1, pp. 40-62, 2011).

Recently, as the demand for a specific spectrum band has increased, a specific spectrum band has saturated, while a low-demand spectrum band has not been fully utilized, resulting in a decrease in the overall spectrum band efficiency.

A cognitive radio is a communication technology in which a secondary user (SU) can opportunistically use the spectrum band of a primary user (PU), and is not a conventional fixed spectrum allocation scheme It is a method to allocate spectrum flexibly and it is proposed as one of solutions to the problem of spectrum saturation.

In cognitive radio communications, license-exempt users should minimize interference in accessing the spectrum of the licensed user.

For this purpose, license-exempt users should be able to detect the spectrum occupation status of the licensee user, and the commonly used spectrum detection techniques are matched filter use, energy detection, feature detection (IF Akyildiz, WY Lee, MC Vuran, S. Mohanty, " A survey on spectrum management in cognitive radio networks, "IEEE Commun. Mag., Vol. 46, no.

First, the method of using the matched filter is a method of detecting a spectrum by designing a filter that maximizes the signal-to-noise (SNR) of the licensee. However, this method is highly accurate, but requires many precautions for the licensee.

The energy detection method is a method of determining that the license user is in a busy state if the received power of the license-exempt user is greater than or equal to a certain threshold value, and otherwise determining that the spectrum is not occupied (idle state). This method detects whether or not the spectrum user occupies the spectrum based on the received power, so that the performance is affected by the noise power.

The feature detection method uses a periodic characteristic of a license user signal. In order to do this, a license user must be observed for a long time and a large amount of calculation is required.

On the other hand, this method can erroneously determine the spectrum occupation state of the license user in the path loss and fading environment, and the energy detection based collaboration method has been proposed to overcome this.

Specifically, a machine learning algorithm suitable for deriving a threshold value in an energy detection method has been proposed (KM Thilina, KW Choi, N. Saquib, E. Hossain, "Machine learning techniques for cooperative spectrum sensing in cognitive radio networks (SVM) and K-average (SVM) in an environment where the license user has multiple transmission power, Clustering-based spectrum detection has been proposed (K. Zhang, J. Li, and F. Gao. "Machine learning techniques for spectrum sensing when the primary user has multiple transmit powers," in Proc. IEEE Int. (ICCS), Macau, China, Nov. 2014, pp. 137-141).

However, in the case of K-means clustering, complexity increases as more spectrum sensing nodes (i.e., license-exempt users) participate in cooperative detection, and complexity increases as the number of levels of transmission power of the license user increases. In addition, there are some inconsistencies with the current situation because the large power (63.1mW average) is used as the received power of the license - exempt user without considering the limited output power according to the electromagnetic wave regulation.

Accordingly, it is required to search for a method for effectively determining the spectrum occupation state of the license user based on the received power of the license-exempt users in the cognitive radio communication.

It is therefore an object of the present invention to provide a method and apparatus for predicting a transmission power of a license user using an artificial neural network algorithm to determine a spectrum occupation state of a license user, And a wireless communication method.

According to another aspect of the present invention, there is provided a cognitive radio communication method including receiving a signal transmitted from a licensee user by a license-exempt user, receiving, by an artificial neural network algorithm, a received power of a signal received by the license- Determining a spectrum occupancy state of the license user according to the predicted transmission power through a support vector machine; and if the license user does not occupy the spectrum , And the license-exempt user utilizes the spectrum.

Here, the determining may include: selecting a decision boundary corresponding to the predicted transmission power among a plurality of decision boundaries learned through the support vector machine, comparing the selected decision boundary with the received power, The spectrum occupation state can be determined.

In addition, the license-exempt user is located within a predetermined distance based on the license user, and the maximum distance that the license-exempt user can be placed based on the license user is determined by a level difference between a plurality of transmission powers of the license user, May be determined based on the difference between the wavelength of the signal to be transmitted and the received power to be targeted for each of the plurality of transmission powers of the licensee.

In this case, the maximum distance may be calculated based on Equation (1).

According to various embodiments of the present invention as described above, in order to determine the spectrum occupancy state of the license user, the transmission power of the license user is predicted using the artificial neural network algorithm, and the spectral occupancy state of the license user is estimated through the support vector machine It is possible to effectively determine the spectrum occupation state even if the license user has a large number of transmission power levels.

1 is a flowchart illustrating a cognitive radio communication method according to an embodiment of the present invention;
2 is a diagram for explaining a method for calculating a maximum distance at which a license-exempt user is arranged according to an embodiment of the present invention;
3 is a view for explaining a crystal boundary according to an embodiment of the present invention;
FIG. 4 is a diagram for explaining an arrangement state of a license-exempt user according to an embodiment of the present invention, and FIG.
5 is a diagram for explaining an electronic device for determining a spectral occupancy state of a licensed user according to an embodiment of the present invention.

Various embodiments of the invention will now be described with reference to the accompanying drawings. It should be understood, however, that the description herein is not intended to limit the invention to the particular embodiments, but includes various modifications, equivalents, and / or alternatives of the embodiments of the invention . In connection with the description of the drawings, like reference numerals may be used for similar components.

In the present invention, the expression "having," " having, "" comprising," Quot ;, and does not exclude the presence of additional features.

In the present invention, the expression "A or B," "at least one of A or / and B," or "one or more of A and / or B," etc. may include all possible combinations of the listed items . For example, "A or B," "at least one of A and B," or "at least one of A or B" includes (1) at least one A, (2) Or (3) at least one A and at least one B all together.

The terms "first," "second," "first," or "second," etc. used in the present invention may be used to denote various components, regardless of their order and / or importance, But is used to distinguish it from other components and does not limit the components.

(Or functionally or communicatively) coupled with / to "another component (eg, a second component), or a component (eg, a second component) Quot; connected to ", it is to be understood that any such element may be directly connected to the other element or may be connected through another element (e.g., a third element). On the other hand, when it is mentioned that a component (e.g., a first component) is "directly connected" or "directly connected" to another component (e.g., a second component) It can be understood that there is no other component (e.g., a third component) between other components.

The phrase " configured to be used "as used in the present invention means that, depending on the situation, for example," having the capacity to, To be designed to, "" adapted to, "" made to, "or" capable of ". The term " configured to (or set up) "may not necessarily mean" specifically designed to "in hardware. Instead, in some situations, the expression "configured to" may mean that the device can "do " with other devices or components. For example, a processor configured (or configured) to perform the phrases "A, B, and C" may be implemented by executing one or more software programs stored in a memory device or a dedicated processor (e.g., an embedded processor) , And a generic-purpose processor (e.g., a CPU or an application processor) capable of performing the corresponding operations.

1 is a flowchart illustrating a cognitive radio method according to an embodiment of the present invention.

First, a licensee user receives a signal transmitted by the licensee user (S110).

Here, the license-exempt user can be placed within a predetermined distance based on the license user.

At this time, the maximum distance that the license-exempt user can be allocated based on the license user is the difference in level between the plurality of transmission powers of the license user, the wavelength of the signal transmitted by the license user, As shown in FIG.

Specifically, the maximum distance d _max can be calculated based on the following equation (1).

는 잡음 전력이다.Here, d ₀ is a constant indicating a reference distance, and? P is a level difference between a plurality of transmission powers of the licensee user (for example, a license user has a transmission power level of 0, 100 mW, 200 mW, 300 mW, Where? Is a constant representing the channel environment,? Is the wavelength of the signal transmitted by the licensee user, n is the pathloss index, and? Lt; / RTI >

Is the noise power.

And k represents the difference between the target received power for each of a plurality of transmission powers of the license user, which can be set by the user.

For example, in FIG. 2, when a license user transmits signals with transmission powers of 0, 100 mW, 200 mW, 300 mW, and 400 mW, respectively, the three license-exiting users (SU1, SU2, SU3) (11, 12, 13, 14, 15).

That is, the received power 11 is a value obtained by multiplying the received power of the signal received by the three license-exemption users SU1, SU2, SU3 when the license user has not transmitted a signal (i.e., when the transmission power of the license user is 0) And the received power 12 represents the received power of the signal received by the three license-exempt users SU1, SU2 and SU3 when the license user transmits a signal with a transmit power of 100 mW, When the license user transmits a signal with a transmission power of 200 mW, the three license-exempt users (SU1, SU2, SU3) indicate the received power of the received signal, and the received power 14 indicates that the license user has a transmission power of 300 mW And the received power 15 indicates that the license user has transmitted a signal with a transmission power of 400 mW, and when the signal is transmitted, the three license-exempt users SU1, SU2, SU3 indicate the received power of the received signal. (SU1, SU2, SU3) of the number of license-exemption users.

At this time, k may mean a value targeted by the user for the difference (21, 22, 23, 24) between the received powers of the license-exempt users SU1, SU2, SU3 depending on the transmission power level of the license user.

The reason why the license-exemption user is arranged according to the distance as shown in Equation (1) is as follows.

Specifically, the distance between received powers (e.g., 21, 22, 23, 24 in FIG. 2) is determined depending on the location of the license-exempt user, which has a large influence on detecting the spectrum occupation state. Thus, the location at which the license-exempt user is located is determined based on Equation 1, taking into consideration the distances to other license-exempt users to ensure communication performance with other license-exempt users, while sufficiently spacing the distances between the received powers .

Referring back to FIG. 1, the transmission power of the license user is estimated based on the received power of the signal received by the license-exemption user through an artificial neural network algorithm (S120).

To do this, first, the received power of the signal received by the license-exempt user can be calculated.

Specifically, the received power of the signal received by the license-exemption user can be calculated based on the following equation (2).

인 가우시안 분포를 따르는 잡음 전력이다.Where P _SU is the received power of the signal received by the license-exempt user, P _US is the transmit power of the licensee user, ζ _h is the multipath page constant of channel h, P _n is the average of 0,

Is the noise power along the in-Gaussian distribution.

이다. 여기에서, d₀는 기준 거리를 나타내는 상수이고, d_{pu , su}는 비면허 유저와 면허 유저 간의 거리이고, λ는 면허 유저가 송신하는 신호의 파장이고, γ는 경로손실지수를 나타내는 상수이다(이는 전술한 수학식 1의 n과 동일한 상수이다).And,

to be. Where d ₀ is a constant representing the reference distance, d _{pu , su} is the distance between the license-issuer and the licensee user, λ is the wavelength of the signal transmitted by the licensee user and γ is a constant representing the pathloss index Which is the same constant as n in the above-mentioned equation (1)).

Then, it is possible to predict the transmission power of the license user according to the reception power of the license-exempt user through the artificial neural network algorithm.

Here, the artificial neural network algorithm is an algorithm for predicting the transmission power of the license user using the received power of the license-exempt user as an input, and may be learned through various sample data.

For example, the artificial neural network algorithm can be learned based on the received power of the license-exempt user calculated according to the transmission power of the license user as shown in Tables 1 to 3 below.

4.1396e-07 1.1942e-06 5.0723e-07 1.0311e-06 5.0002e-07 1.3216e-06 6.2526e-07 1.2546e-06 6.3755e-07 1.2014e-06 3.9032e-07 1.4684e-06 6.7607e-07 1.2215e-06 6.1324e-07 1.3309e-06 4.6182e-07 1.4648e-06 5.7025e-07 1.1264e-06 6.3661e-07 1.1668e-06 5.4946e-07 1.2218e-06 4.0896e-07 1.0479e-06 4.8145e-07 1.3949e-06

8.3292e-07 1.7228e-06 7.6327e-07 1.8679e-06 8.2496e-07 1.9399e-06 8.1522e-07 1.8716e-06 9.8098e-07 1.8934e-06 8.3654e-07 1.9591e-06 7.3705e-07 1.6010e-06 8.1489e-07 2.0730e-06 8.5308e-07 2.0704e-06 8.7484e-07 1.8984e-06 9.7075e-07 1.7721e-06 7.1915e-07 1.7777e-06 6.6195e-07 1.7974e-06 8.0574e-07 1.9768e-06

1.0570e-06 2.3782e-06 1.1661e-06 2.5007e-06 1.0642e-06 2.4846e-06 9.8860e-07 2.6134e-06 1.0169e-06 2.2409e-06 9.8607e-07 2.5380e-06 1.0023e-06 2.5787e-06 1.1758e-06 2.4422e-06 1.1505e-06 2.3905e-06 1.0184e-06 2.4923e-06 1.0092e-06 2.3888e-06 1.2885e-06 2.3743e-06 1.1261e-06 2.5439e-06 1.0421e-06 2.3970e-06

Table 1 shows the received power of the license-exempt user calculated when the transmission power of the license user is 200 mW. Table 2 shows the reception power of the license-exempt user calculated when the transmission power of the license user is 300 mW , And Table 3 shows the received power of the license-exempt user calculated when the transmission power of the license user is 400 mW.

It should be noted that the above-described Tables 1 to 3 are merely illustrative examples, and the artificial neural network algorithm can be learned based on various received powers calculated according to various transmission powers of the license user.

As a result, the transmission power of the license user can be predicted by using the reception power received by the license-exempt user as the input of the artificial neural network.

Then, the spectral occupancy state of the license user according to the predicted transmission power is determined through a support vector machine (SVM) (S130).

Specifically, a decision boundary corresponding to the predicted transmission power among a plurality of decision boundaries (i.e., threshold values) learned through the support vector machine is selected, and the received power is compared with the selected decision boundary, It is possible to determine the spectrum occupation state.

Here, the decision boundary is a boundary for distinguishing the idle state from the busy state and the spectrum state from the license user according to the received power of the license-exempt user, , And may be learned via a support vector machine. That is, the support vector machine is learned through various sample data indicating the spectrum occupation state of the license user according to the transmission power of the license user according to the transmission power of the license user, and the learned decision boundary is obtained for each transmission power of the license user have.

In this case, a decision boundary corresponding to the transmission power of the license user predicted through the artificial neural network algorithm among the learned decision boundaries can be selected for each transmission power of the license user.

For example, when a license user transmits a signal with a transmission power of 0, 100 mW, 200 mW, 300 mW, and 400 mW, respectively, transmission power levels of 0, 100 mW, 200 mW, 300 mW, There may be a learned decision boundary for.

In this case, when the transmission power of the license user predicted through the artificial neural network algorithm is 200 mW, transmission power of 200 mW among the learned boundaries for the transmission power levels of 0, 100 mW, 200 mW, 300 mW, You can select the learned decision boundary for the level.

Then, the selected decision boundary is compared with the received power received by the license-exemption user, and it is possible to judge whether or not the license user has a spectrum occupied.

Specifically, if the received power of the license-exempt user is greater than or equal to the selected decision boundary, the license user is judged to have occupied the spectrum, and if the received power of the license-exempt user is smaller than the selected decision boundary, the license user has not occupied the spectrum It can be judged.

For example, as shown in FIG. 3, if the received power of the signal received from the license user PU by the two license-unauthorized users SU1 and SU2 is equal to or greater than the decision boundaries 41, 42, 43, and 44, The receiving power of the signal received from the license user PU by the two license-exclusion users SU1 and SU2 is smaller than the decision boundaries 41, 42, 43 and 44 , It can be determined that the license user (PU) is idle without occupying the spectrum.

Here, the crystal boundaries 41, 42, 43, and 44 of FIGS. 3B to 3E may be learned crystal boundaries for transmission power levels of 100, 200, 300, and 400 mW, respectively.

Returning to FIG. 1, if the license user does not occupy the spectrum thereafter, the license may use the spectrum (S140). That is, the licensee can perform communication using the spectrum band in the sense that the license user does not occupy the spectrum.

As described above, in the present invention, to determine the spectrum occupation state of the license user, the transmission power of the license user is predicted using the artificial neural network algorithm, and the spectrum occupation state of the license user is determined through the support vector machine. Thus, even if the licensee user has a large number of transmission power levels, the spectral occupancy state can be effectively determined. In general, the spectral classification performance depends on the distribution of license-exempt users participating in spectrum detection. In the present invention, in order to maximize the communication performance with the license-exempt users while correcting the spectral classification performance, The licensee is placed according to the distance calculated by the licensee.

Accordingly, the present invention can maximize the communication performance with the license-exempt users while determining the spectrum occupation state of the license user based on the received power of the license-exempt user.

4 is a diagram for explaining an arrangement of license-exempt users according to an embodiment of the present invention.

For example, as shown in FIG. 4, it is assumed that a license user PU1 and a license-exempt user SU1 exist.

At this time, when a new user license (PU2) is found, and a new unlicensed user (SU2) within a circle of a radius relative to the licensed user (PU2) to d _max can be disposed.

Likewise, in the case, a new license seat (PU3) are found, and a new unlicensed user (SU3) within a circle of a radius relative to the licensed user (PU3) to d _max can be disposed. On the contrary, if the new license seat (PU4) are found, it can be a new unlicensed user (SU4) disposed within a circle of a radius relative to the licensed user (PU4) to d _max, new licensed user (PU5) is found case, a new license-exempt user (SU5) can be disposed within the circle of the radius relative to the licensed user (PU5) to d _max.

5 is a diagram for explaining an electronic device for determining a spectral occupancy state of a licensed user according to an embodiment of the present invention.

Figure 5, for convenience of description, the electronic apparatus 500 is described an example to detect the spectral occupancy of the license the user (PU2), and unlicensed within circle with a maximum distance d _max, based on the licensed user (PU3) It is assumed that the user SU3 is disposed. However, this is only an example, and it goes without saying that the number of license users and license-exempt users can be variously changed.

The electronic device 500 (e.g., a fusion center) for determining a spectrum occupancy state of a licensee user includes a communication unit 510, a memory 520, and a processor 530.

The communication unit 510 can perform communication with the license-exempt user.

In this case, the communication unit 510 can perform communication with the license-exiting user based on various types of communication methods. To this end, the communication unit 510 may include at least one of a wireless communication chip (not shown), a short range communication chip (not shown), and may include one or more modules for connecting the electronic device 500 to one or more networks You may.

The memory 520 may store various programs and data necessary for the operation of the electronic device 500. Memory 520 may also store instructions or data received from processor 530 or other components or generated by processor 530 or other components.

In this case, the memory 520 may be implemented as a nonvolatile memory, a volatile memory, a flash memory, a hard disk drive (HDD), or a solid state drive (SSD). The memory 520 is accessed by the processor 530 and read / write / modify / delete / update of data by the processor 530 can be performed.

In particular, memory 520 may store a module for executing the artificial neural network algorithm and a module for executing the support vector machine. In addition, the memory 520 may store information about the determined user's transmission power and the determined boundary, which are learned through the support vector machine.

Processor 530 controls the overall operation of electronic device 500. For example, the processor 530 may operate an operating system or an application program to control hardware or software components connected to the processor 530, and may perform various data processing and operations. The processor 530 may also load and process commands or data received from at least one of the other components into volatile memory and store the various data in non-volatile memory.

To this end, the processor 530 may be a general-purpose processor capable of performing corresponding operations by executing one or more software programs stored in a memory device or a dedicated processor (e.g., an embedded processor) (E.g., a CPU or an application processor).

First, the processor 530 can receive the received power calculated by the license-exempt user through the communication unit 510. [

When the license-exempt users SU1 to SU3 receive signals transmitted by the license user PU2, the license-exemption users SU1 to SU3 can calculate the received power of the received signals. Further, the license-exempt users SU1 to SU3 can transmit information on the calculated received power to the electronic device 500. [ In this case, the communication unit 510 can receive information on the received power transmitted by the license-exempt users SU1 to SU3.

Thereafter, the processor 530 predicts the transmission power of the license user according to the received power of the signal received by the license-exempt user through the artificial neural network algorithm.

Then, the processor 530, via the support vector machine, determines the spectrum occupation state of the license user according to the predicted transmission power.

Specifically, the processor 530 selects a decision boundary corresponding to the predicted transmission power among a plurality of decision boundaries learned through the support vector machine, compares the received power with the selected decision boundary, and obtains the spectrum occupation state of the license user It can be judged.

In this case, the processor 530 determines that the license user occupies the spectrum if the received power of the license-exempt user is greater than or equal to the selected decision boundary, and if the received power of the license-exempt user is smaller than the selected decision boundary, It can be judged that it is not in the state.

For example, it is assumed that the transmission power of the license user (PU3) predicted through the artificial neural network algorithm is 200 mW.

In this case, the processor 530 selects the learned decision boundary for the transmission power level of 200 mW, compares the learned decision boundary with the received power of the license-exemption users SU1 to SU3, The license user PU3 determines that the spectrum is occupied by the license user PU3 if the received power is greater than or equal to the selected decision boundary and the license user PU3 occupies the spectrum if the received power of the license- It can be judged that it is not in the state.

Then, the processor 530 can transmit information on the spectrum occupation state of the license user to the license-exiting user.

For example, the processor 530 may transmit to the license-exempt users SU1 through SU3 via the communication unit 510 that the license user PU3 occupies the spectrum or does not occupy the spectrum.

In this case, the license-exempt user may use the spectrum based on the information on the spectrum occupancy status of the license user received from the electronic device 500. [

For example, if the licensee user (SU1 to SU3) judges that the license user (PU3) does not occupy the spectrum based on the information on the spectrum occupancy status of the license user (PU3) received from the device (500) Communication can be performed using the spectrum of the user PU3.

Meanwhile, various embodiments of the present disclosure may be implemented in software, including instructions stored on machine-readable storage media, such as a machine (e.g., a computer) (E.g., electronic device A) in accordance with the disclosed embodiments, as an apparatus that is capable of calling instructions and operating in accordance with the called instructions. When the instructions are executed by a processor, Or other components under the control of the processor. The instructions may include code generated or executed by a compiler or interpreter. May be provided in the form of a non-transitory storage medium, where 'non-transient' means that the storage medium does not contain a signal It does not distinguish that data is stored semi-permanently or temporarily on the storage medium.

According to a temporary example, a method according to various embodiments disclosed in this disclosure may be provided in a computer program product. A computer program product can be traded between a seller and a buyer as a product. A computer program product may be distributed in the form of a machine readable storage medium (eg, compact disc read only memory (CD-ROM)) or distributed online through an application store (eg PlayStore ™). In the case of on-line distribution, at least a portion of the computer program product may be temporarily stored, or temporarily created, on a storage medium such as a manufacturer's server, a server of an application store, or a memory of a relay server.

Each of the components (e.g., modules or programs) according to various embodiments may be comprised of a single entity or a plurality of entities, and some of the subcomponents described above may be omitted, or other subcomponents May be further included in various embodiments. Alternatively or additionally, some components (e.g., modules or programs) may be integrated into a single entity to perform the same or similar functions performed by each respective component prior to integration. Operations performed by a module, program, or other component, in accordance with various embodiments, may be performed sequentially, in parallel, repetitively, or heuristically, or at least some operations may be performed in a different order, .

500: electronic device 510: communication part
520: Storage unit 530: Processor

Claims (4)

In a cognitive radio method,
Receiving a signal transmitted from a primary user by a licensee;
Estimating a transmission power of the license user according to a received power of a signal received by the license-exemption user through an artificial neural network algorithm;
Determining a spectral occupancy state of the license user according to the predicted transmission power through a support vector machine; And
And if the license user does not occupy the spectrum, the licensee uses the spectrum. The method according to claim 1,
Wherein the determining step comprises:
Selecting a decision boundary corresponding to the predicted transmission power among a plurality of decision boundaries learned through the support vector machine and comparing the selected decision boundary with the received power to determine a spectrum occupation state of the licensee user Wherein the wireless communication method comprises: The method according to claim 1,
Wherein the license-exempt user is disposed within a predetermined distance based on the license user,
Wherein the maximum distance that the license-exempt user can be placed on the basis of the license user is determined based on a level difference between a plurality of transmission powers of the licensee user, a wavelength of a signal transmitted by the licensee user, and a plurality of transmission powers of the licensee user, And the target received power is determined based on a difference between the target received powers. The method of claim 3,
_Wherein the maximum distance d _max is calculated on the basis of the following equation:

Images