KR101741361B1 - Mtehod for optimal transceiver design in interference alignment based cognitive radio network - Google Patents

Abstract

Disclosed is a method for designing an optimal transceiver in a cognitive radio network based on interference alignment. The method for designing the optimal transceiver in the cognitive radio network including multiple secondary users and a primary user in which a transmitter and a receiver based on multiple antennas have a pair structure includes: a step of determining either a first interference alignment technique of aligning interference influencing the primary user (PU) from the multiple secondary users (SU) participating in the cognitive radio network and the interference between the multiple secondary users (SU) in accordance with a use purpose or a second interference alignment technique of aligning the interference influencing the primary user from the secondary users excepting for the interference between the multiple secondary users, as a technique for the interference alignment; and a step of generating a precoding matrix to remove the interference to the primary user based on the determined interference alignment technique.

Description

TECHNICAL FIELD [0001] The present invention relates to a method of designing an optimal transceiver apparatus for an interference-

Embodiments of the present invention relate to techniques for designing optimal transmission and reception devices for efficient channel use using interference alignment techniques in a cognitive radio network.

This study was conducted as a result of the research project of the future ICT Research Center of the Creation Science Department and the Information and Communication Technology Promotion Center (IITP-2016-H8501-16-1019).

The patent assignment number is 1711035210.

As the development of mobile communication services and the demand for wireless spectrum explosively increase, improving the efficiency of spectrum has recently become a major problem in wireless communication research. Cognitive Radio (CR) technology exists as a solution to the lack of spectrum. Cognitive radio technology is a wireless technology that operates by measuring the propagation environment and setting operation parameters of the wireless device in accordance with the measured propagation environment. For example, the technology that maximizes the transmission capacity of a wireless device according to channel characteristics, minimizes interference between devices, promotes mutual operation between different types of systems, or uses unused frequencies for a time not used by a primary user, It belongs to the category of technology.

Underlay scheme spectrum sharing technology is a typical technology in cognitive wireless networks. In the case of underlay scheme, it is a major issue to control interference with a primary user (PU). First, There are various methods for controlling the interference of the user. However, the trade off problem of improving the performance of a secondary user (SU) while guaranteeing quality of service (QoS) to the user is still a problem to be solved to be.

The Interference Alignment (IA) technique is a technique for controlling interference using a precoding matrix and a decoding matrix to remove interference in a signal subspace. With interference alignment, the signal to be received within the signal space can be recovered without interference. Recently, research is being conducted to apply the performance enhancement of wireless network system to cognitive radio due to interference sorting. One of the major problems in interference-based cognitive wireless networks is solving the low signal-to-interference noise ratio (SINR) of the user (PU).

H. Zhou , T. Ratnarajah , and YC Liang , "On secondary network interference alignment in cognitive radio, " in Proc . IEEE DySPAN, Aachen , Germany, May 2011, pp. 637-641. Describes a technique for improving the transmission rate of a cognitive wireless user by applying a method of using a precoding matrix and power allocation to a cognitive wireless network in which a first user and a plurality of cognizant wireless users coexist. Non-Patent Document [2] F. Rezaei and A. Tadaion , "Interference alignment in cognitive radio networks," IET Commun ., Vol. 8, no. 10, pp. 1769-1777, Jul . In 2014 , we first describe a technique for performing cooperative communication between users and quickly performing interference alignment in a cognitive radio network using a water filling method and an interference cancellation technique. However, in the case of non-patent documents [1] and [2], a problem solving study on the performance improvement of the user is additionally demanded.

Accordingly, it is an object of the present invention to solve a trade off that improves the performance of a secondary user (SU) while ensuring quality of service (QoS) for a priority user (PU) in a cognitive wireless network based on interference- An interference alignment technique is needed to improve the performance (e.g., SINR) of the PU.

[1] H. Zhou, T. Ratnarajah, and Y. C. Liang, "On secondary network interference in cognitive radio," Proc. IEEE DySPAN, Aachen, Germany, May 2011, pp. 637-641. [2] F. Rezaei and A. Tadaion, "Interference alignment in cognitive radio networks," IET Commun., Vol. 8, no. 10, pp. 1769-1777, Jul. 2014. [3] K. Gomadam, V. R. Cadambe, and S. A. Jafar, "A Distributed Numerical Approach to Interference Alignment and Applications to Wireless Interference Networks," IEEE Trans. Inf. Theory, vol. 57, no. 6, pp. 3309-3322, Jun. 2011. [4] H. Huang and V.K.N. Lau, "Partial interference alignment for K-user MIMO interference cahnnels," IEEE Trans. Signal Process., Vol. 59, no. 10, pp. 4900-4908, Aug. 2011. [5] H. Gao et al., "New uplink opportunistic interference alignment: An active alignment approach," In Proc. IEEE WCNC 2013, pp. 3099-3104, Shanghai, China, Apr. 2013. [6] H. Men, N. Zhao, M. Jin and J. M. Kim, "Optimal Transceiver Design for Interference Alignment Based Cognitive Radio Networks," IEEE Communications Letters, vol. 19, no. 8, pp. 1442-1445, Aug. 2015.

The present invention relates to a technique for designing an Optimal Transceiver Design (OTD) in an interference-based cognitive radio network, wherein a cognitive radio user (SU) is used to sort interference to a user (PU) (PIA) is used in generating the precoding matrix of the UE and the perceptual radio user (SU) takes priority over the interference of the user (PU) on the perceived wireless user (SU) To a technique for aligning interference in consideration of interference to a user (PU).

The precoding matrix of the priority user (PU) maximizing the performance of the priority user (PU) is generated in the transmission apparatus corresponding to the user. To a technique for generating a decoding matrix that maximizes a received SINR in order to maximize performance.

A method of interference arrangement in a transmitter in a cognitive radio network including a primary user having a pair structure and a plurality of cognitive radio users (Secondary User) A first interference sorting technique for sorting interferences between a plurality of cognitive radio users (SU) participating in the perceptual wireless network and interference between a plurality of cognizant wireless users, And a second interference alignment scheme for aligning interference on a preference user from the perceived wireless users, excluding interference between the plurality of perceived wireless users, as a technique for interference alignment, and determining the determined interference Generating a precoding matrix for eliminating interference to the preferred user based on an alignment scheme; . ≪ / RTI >

According to an aspect, generating the precoding matrix comprises: calculating an interference covariance of a signal received from a different cognizant wireless user from a signal of the preferential user according to the first interference alignment technique; Performing an eigenvalue decomposition on an interferometric covariance matrix including the eigenvalue decomposition, and generating the precoding matrix based on the eigenvalue vectors obtained through the eigenvalue decomposition.

According to another aspect, the precoding matrix generated according to the first interference alignment technique may be a precoding matrix that reduces or minimizes interference to the other cognizant wireless users.

According to another aspect, the step of generating the precoding matrix further comprises the steps of: removing a signal received from another cognitive radio user according to the second interference alignment technique; Performing eigenvalue decomposition on the generated interference covariance matrix, and generating the precoding matrix including the eigenvalue vectors obtained through the eigenvalue decomposition.

According to another aspect, the precoding matrix generated according to the second interference alignment technique may be a precoding matrix that reduces or minimizes interference affecting the preferred user.

According to another aspect of the present invention, when the receiving device is any one of the plurality of cognizant wireless users, the signal received from the transmitting device corresponding to the preferred user and the plurality of cognitive radio A decoding matrix may be generated in which a signal to interference noise ratio (SINR) of the perceived wireless user is maximized based on a signal received from a transmitting apparatus corresponding to another cognizant wireless user except for any one of the cognizant wireless users have. In this case, the step of generating the precoding matrix may generate the precoding matrix according to the decoding matrix.

The present invention creates a precoding matrix and a decoding matrix using a suitable technique according to the purpose of use by using two different interference sorting algorithms using partial interference alignment (PIA) The receiving apparatus of the first embodiment can sort the interference received from the cognitive radio user (SU) and maximize the performance of the priority user (PU) by using the precoding matrix of the user (PU).

In addition, the performance of the perceived wireless user (SU) can be improved by generating a decoding matrix of the perceived wireless user (SU) at the receiving device of the perceived wireless user (SU).

FIG. 1 is a diagram illustrating a configuration of an interference-based cognitive radio network according to an exemplary embodiment of the present invention. Referring to FIG.
2 is a flow diagram illustrating an operation for generating a precoding matrix for interference alignment in a cognitive radio network, in accordance with an embodiment of the present invention.
3 is a block diagram showing an internal configuration of a transmitting apparatus for generating a precoding matrix.
4 is a diagram illustrating a perceptual wireless network that performs interference alignment based on a PIA-SU algorithm, in one embodiment of the present invention.
5 is a diagram illustrating a perceptual wireless network that performs interference alignment based on a PIA-NSU algorithm, in one embodiment of the present invention.
FIG. 6 is a diagram illustrating performance of a priority user and a cognizant wireless user in performing interference alignment based on the PIA-SU and PIA-NSU algorithms in an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The embodiments are directed to interference cancellation techniques for reducing or eliminating interference to a priority user (PU) in a cognitive radio network in which a transmitter and a receiver are based on a Multi Input Multi Output (MIMO). In particular, in a transmitting apparatus of a cognitive radio user (SU) participating in a cognitive radio network, a precoding matrix is formed in one of two different ways using partial interference alignment (PIA) To a technique for generating a signal. For example, without considering the interference to the first interference alignment technique (PIA-SU) and the cognitive radio user (SU) that generates the precoding matrix considering the interference to the user PU and other cognizant wireless users Any one of the second interference alignment technique (PIA-NSU) that generates a precoding matrix considering only the interference to the user (PU) can be used according to the purpose of use.

In addition, the embodiments may be configured such that in a receiving apparatus of a cognitive radio user (SU) forming a pair with a transmitting apparatus of the cognitive radio user (SU), the SINR of a signal received from the transmitting apparatus of the cognizant wireless user is maximized To a technique for generating a decoding matrix.

Also, the embodiments of the present invention may be configured such that a transmitting apparatus of a user (PU) first generates a precoding matrix for maximizing the performance of a priority user (PU) and generates a precoding matrix for a priority user To a technique for minimizing interference from a transmitting device of at least one cognizant wireless user (SU) participating in a perceptual wireless network at a receiving device of a wireless local area network (PU).

In the following embodiments, the precoding matrix generated in the transmitting apparatus of the cognitive radio user (SU) is a precoding matrix of the 'cognitive radio user', the decoding matrix generated in the receiving apparatus of the cognizant wireless user (SU) The decoding matrix generated in the receiving apparatus of the priority user (PU) is a decoding matrix generated by the priority user (PU) Quot; decoding matrix " of < / RTI >

In this specification, the perceived wireless user (SU) refers to a transmitting apparatus (i.e., transmitting node) and a receiving apparatus (i.e., receiving node) corresponding to a pair of cognizant wireless users. (I.e., a transmitting node) and a receiving apparatus (i.e., a receiving node) corresponding to a priority user constituting a pair. In other words, a receiving priority user node can participate in a cognitive wireless network in a pair relationship with a receiving wireless user node having a pair relationship with a transmitting wireless user node and a transmitting priority user node.

FIG. 1 is a diagram illustrating a configuration of an interference-based cognitive radio network according to an exemplary embodiment of the present invention. Referring to FIG.

Referring to FIG. 1, the perceived wireless network 100 may include a plurality of transmission devices 110 and a plurality of reception devices 120. The transmitting devices 110 include a user (PU 111) and a cognizant wireless users 112 and the receiving devices 120 may include a user 121 and a cognizant wireless users 122 have. At this time, the priority user 111, which is a transmitting node, and the priority user 121, which is a receiving node, correspond to each other, and the cognitive radio users 112, , That is, a pair. Although FIG. 1 shows a case where there are a plurality of cognizant wireless users 112 and 122, this corresponds to the embodiment, and the cognizant wireless users 112 and 122 may be one or more than one.

Hereinafter, with reference to FIG. 2 to FIG. 5, a case will be described on the assumption that there is one pair of PU participating in the perceived wireless network and a plurality of cognizant wireless users having a pair structure .

FIG. 2 is a flowchart illustrating an operation of generating a precoding matrix for interference alignment in a cognitive radio network according to an embodiment of the present invention. FIG. 3 illustrates an internal configuration of a transmitting apparatus for generating a precoding matrix Block diagram.

Specifically, FIGS. 2 and 3 illustrate how a receiving device (i.e., a receiving node) of a priority user (PU) receives a partial interference alignment (PIA) from a transmitting device And generating a precoding matrix for eliminating the interference. That is, the precoding matrix is for interference sorting in the receiving apparatus of the user PU, considering only the interference of the perceived wireless users SUs to the user PU, Interference to wireless users (SUs) may not be considered.

At this time, the precoding matrix may be generated based on any one of two different techniques (PIA-SU, PIA-NSU) using partial interference alignment (PIA). In consideration of the interference that a cognitive radio user (SU) affects a priority user (PU), the PIA considering interference (SU) between the cognitive radio users (SUs) (PIA-NSU) is a scheme for considering interference that affects a user (PU). In consideration of interference, interference between perceived wireless users (SUs) is not considered, May represent a technique for generating a matrix.

2, each of the steps 210 and 220 is performed by a determination unit 310, a precoding matrix generation unit 320, and a transmission unit 330, which are components of the transmission apparatus 301 shown in FIG. . 3, the transmitting apparatus 301 operates as a transmitting apparatus of a perceived wireless user. However, the transmitting apparatus 301 may operate as a transmitting apparatus of a user (PU).

In step 210, the determination unit 310 may determine an interference alignment technique for sorting the interference to the user (PU) first based on the purpose of use. At this time, the determination unit 310 may determine any one of the first interference alignment technique (PIA-SU) and the second interference alignment technique (PIA-NSU) as a technique for interference alignment, depending on the purpose of use.

For example, when the purpose of use is enhancement of the transmission rate of the user (PU), the determination unit 310 may determine the PIA-NSU algorithm by the interference alignment technique. When the purpose of use is QoS guarantee of the user (PU), the determination unit 310 can determine the PIA-SU algorithm as an interference sorting technique. In this case, the full IA may be determined as a technique for interference alignment according to the purpose of use, in addition to the PIA-SU and PIA-NSU algorithms.

[ Non-Patent Document 4] H. Huang and VKN . Lau , "Partial interference alignment for K-user MIMO interference cahnnels ," IEEE Trans. Signal Process., Vol. 59, no. 10, pp. 4900-4908, Aug. 2011. [5] H. Gao et al., "New uplink opportunistic interference alignment: An active alignment approach," In Proc . IEEE WCNC 2013, pp. 3099-3104, Shanghai, China, Apr. 2013. [6] H. Men, N. Zhao , M. Jin and JM Kim, "Optimal Transceiver Design for Interference Alignment Based Cognitive Radio Networks," IEEE Communications Letters, vol. 19, no. 8, pp. 1442-1445, Aug. In 2015 , an embodiment of PIA-SU and PIA-NSU can be confirmed, and in the non-patent document [3], an embodiment of a perfect interference sorting technique can be confirmed. The PIA-NSU algorithm can be used if the user PU only constructs a mutually exclusive network with the cognizant wireless user SU, if the purpose of use is to achieve the protection objective for the user PU first. And, if the purpose of use is to consider interference between aware wireless users (SU), then the PIA-SU algorithm can be used. In case of constructing a cooperative network between the first user (PU) and the cognitive radio user (SU), a full interference scheme (full IA) can be used.

Here, the PIA-SU algorithm indicates that the interference sorting is performed considering the interference between the first user and the plurality of cognitive radio uses from a plurality of cognitive radio users participating in the cognitive radio network, and the PIA-NSU algorithm Only interference from a plurality of cognitive radio users participating in the cognitive radio network to the first user may be considered and interference alignment may be performed without consideration of the interference between the plurality of cognizant wireless users. A detailed description of the PIA-SU algorithm and the PIA-NSU algorithm will be described later with reference to FIG. 4 and FIG.

In operation 220, the precoding matrix generator 320 may generate a precoding matrix according to the determined interference alignment technique.

For example, if the PIA-SU algorithm is determined to be an interference alignment scheme, the precoding matrix generator 320 first calculates the interference covariance of the signal received from the other user (PU) An interference covariance matrix including covariance can be generated. The precoding matrix generator 320 may generate a precoding matrix including eigenvalue vectors obtained by performing eigenvalue decomposition on the interference covariance matrix. That is, a signal received from a transmitter of a user (PU) is classified into interference and a signal received from a transceiver of other cognitive radio users is arranged in an interference space, and a signal of a cognitive radio user, A precoding matrix for aligning in space can be generated. Herein, another cognitive radio user may refer to a cognitive radio user other than the cognitive radio user who generates a precoding matrix according to the determined interference alignment scheme among a plurality of cognitive radio users participating in the cognitive radio network. For example, in FIG. 1, when generating the precoding matrix at the perceived wireless user SU1, the other cognitive wireless users may be SU2 through SUK.

As such, the precoding matrix generated based on the PIA-SU algorithm can be used to reduce or minimize interference to other cognitive wireless users, as well as considering interference between cognizant wireless users.

In another example, when the PIA-NSU algorithm is determined to be an interference alignment technique, the precoding matrix generator 320 may use the signals received from the user PU, except for signals received from other cognitive radio users, A matrix can be generated. The precoding matrix generator 320 may generate a precoding matrix based on the eigenvalue vectors obtained by performing eigenvalue decomposition on the interference covariance matrix. As described above, since the precoding matrix generated according to the PIA-NSU algorithm considers only the interference to the priority usage (PU), it can be used to minimize the interference to the user (PU).

As the precoding matrix is generated using partial interference alignment (PIA), such as the PIA-SU or PIA-NSU algorithm described above, the interference to the user (PU) can be aligned first. For example, the transmitting unit 330 of the transmitting apparatus 310 may precode and transmit a data stream to be transmitted based on the generated precoding matrix.

At this time, in the present invention, in order to improve or maximize the performance of the perceived wireless user, a decoding matrix for maximizing the interference ratio between received signals may be generated for a cognitive wireless user, and based on the generated decoding matrix, .

For example, the receiving device 122-1 of the cognitive radio user first receives signals from the transmitting device (PU) 111 of the user and transmitting devices 112-2 and 112-3 of the plurality of cognitive radio users (SINR) of a signal received from a transmitting apparatus 112-1 forming a pair with the receiving apparatus 112-1 based on a signal received from the transmitting apparatus 112-1.

Hereinafter, when the MIMO-CR interference network operates as a cognitive radio transmission apparatus in a (K + 1) -user MIMO-CR interference network in which a PU and a K- A precoding matrix is generated according to the PIA-SU and PIA-NSU algorithms in the transmitter, and a decoding matrix for maximizing the SINR of the perceived wireless user is generated in the receiver considering the case of operating as a cognitive radio receiver An operation of generating a precoding matrix for maximizing a data rate of a priority user (PU) in a transmitting apparatus in consideration of a case of operating as a transmitting apparatus of a user, and a case of operating as a receiving apparatus of a user first And generating a decoding matrix at the receiving apparatus will be described in detail with reference to mathematical expressions.

행렬

는 j번째 송신 장치에서 i번째 수신장치로 가는 채널 계수를 나타낼 수 있다. 여기서,

이고, 채널계수

의 모든 성분들은 독립항등분포를 따르며, 확률분표는

를 따를 수 있다. 이때, i번째 수신장치에서의 수신신호는 아래의 수학식 1과 같이 표현될 수 있다. The 0th user in the MIMO-CR interference network means a priority user (PU), and the remaining K users can mean a cognitive radio user (SU). At this time, a transmitting apparatus operating as a user or a cognizant wireless user may have M antennas, and the receiving apparatus may have N antennas. Consisting of N rows and M columns of complex numbers

procession

May represent a channel coefficient from the j < th > transmitter to the i < th > receiver. here,

And the channel coefficient

All components of the equation follow the independent distribution, and the probability distribution is

. At this time, the received signal at the i-th receiving apparatus can be expressed by Equation (1) below.

는 i번째 사용자(즉, i번째 송신장치)의 송신신호, 즉, 송신심볼(symbol)을 나타내고,

는 j번째 사용자(즉, j번째 송신장치)의 송신신호를 나타내고,

은

을 따르는 CSCG(Circularly Symmetric Complex Gaussian) 잡음 벡터를 나타내고,

는 j번째 송신 장치에서 i번째 수신장치로의 채널 계수를 나타내고,

는 i번째 송신 장치에서 i번째 수신장치로의 채널 계수를 나타낼 수 있다. 수학식 1에 따르면, 사용자들 간 간섭(예컨대, 우선사용자에게 영향을 주는 복수의 인지무선 사용자들에 의한 간섭 및 복수의 인지무선 사용자들 간의 간섭)으로 인해 우선사용자(PU)와 인지무선 사용자(SU)의 전송 성능이 감소될 수 있다. 이에 따라, 사용자들 간의 간섭을 제거하기 위해 간섭 정렬 알고리즘(예컨대, PIA-SU, PIA-NSU)이 인지무선 네트워크에 적용될 수 있다. i번째 사용자의 d개의 전송 스트림을 아래의 수학식 2를 통해 수신될 수 있다. 즉, 간섭 정렬 알고리즘이 적용된 i번째 수신장치에서의 수신신호는 아래의 수학식 2와 같이 표현될 수 있다.In Equation (1)

Denotes a transmission signal, i.e., a transmission symbol, of the i-th user (i.e., the i-th transmission apparatus)

Denotes a transmission signal of the j-th user (i.e., the j-th transmission apparatus)

silver

(Circularly Symmetric Complex Gaussian) noise vector,

Represents the channel coefficient from the j < th > transmitter to the i < th > receiver,

May represent the channel coefficient from the i < th > transmitter to the i < th > receiver. According to Equation (1), interference between users (e.g., interference by a plurality of cognizant wireless users affecting a user first and interference between a plurality of cognizant wireless users) SU) can be reduced. Accordingly, interference cancellation algorithms (e.g., PIA-SU, PIA-NSU) can be applied to the perceived wireless network to eliminate interference between users. The d transport streams of the i-th user may be received via Equation (2) below. That is, the received signal at the i-th receiving apparatus to which the interference sorting algorithm is applied can be expressed by Equation (2) below.

는 i번째 사용자(즉, i번째 송신장치)의 송신신호, 즉, 송신심볼(symbol)을 나타내고,

는 j번째 사용자(즉, j번째 송신장치)의 송신신호를 나타내고,

은

을 따르는 CSCG(Circularly Symmetric Complex Gaussian) 잡음 벡터를 나타내고,

는 j번째 송신 장치에서 i번째 수신장치로의 채널 계수를 나타내고,

는 i번째 송신 장치에서 i번째 수신장치로의 채널 계수를 나타낼 수 있다. 수학식 1에 따르면, 사용자들 간 간섭(예컨대, 우선사용자에게 영향을 주는 복수의 인지무선 사용자들에 의한 간섭 및 복수의 인지무선 사용자들 간의 간섭)으로 인해 우선사용자(PU)와 인지무선 사용자(SU)의 전송 성능이 감소될 수 있다. 이에 따라, 사용자들 간의 간섭을 제거하기 위해 간섭 정렬 알고리즘(예컨대, PIA-SU, PIA-NSU)이 인지무선 네트워크에 적용될 수 있다. i번째 사용자의 d개의 전송 스트림을 아래의 수학식 2를 통해 수신될 수 있다. 즉, 간섭 정렬 알고리즘이 적용된 i번째 수신장치에서의 수신신호는 아래의 수학식 2와 같이 표현될 수 있다.In Equation (1)

Denotes a transmission signal, i.e., a transmission symbol, of the i-th user (i.e., the i-th transmission apparatus)

Denotes a transmission signal of the j-th user (i.e., the j-th transmission apparatus)

silver

(Circularly Symmetric Complex Gaussian) noise vector,

Represents the channel coefficient from the j < th > transmitter to the i < th > receiver,

May represent the channel coefficient from the i < th > transmitter to the i < th > receiver. According to Equation (1), interference between users (e.g., interference by a plurality of cognizant wireless users affecting a user first and interference between a plurality of cognizant wireless users) SU) can be reduced. Accordingly, interference cancellation algorithms (e.g., PIA-SU, PIA-NSU) can be applied to the perceived wireless network to eliminate interference between users. The d transport streams of the i-th user may be received via Equation (2) below. That is, the received signal at the i-th receiving apparatus to which the interference sorting algorithm is applied can be expressed by Equation (2) below.

는 i번째 송신장치에서 전송한 송신신호를 나타내고, 송신전력 P는

로 표현될 수 있다.

는 i번째 송신장치의 프리코딩 행렬을 나타내고,

는 상기 i번째 송신장치와 페어(pair)를 이루는 i번째 수신장치의 디코딩 행렬을 나타낼 수 있다. 이때, 사용자들 간의 간섭을 제거하기 위해 프리코딩 행렬

와 디코딩 행렬

는 아래의 수학식 3 및 수학식 4를 만족해야 한다. In Equation (2)

Represents the transmission signal transmitted from the i < th > transmission apparatus, and the transmission power P is

. ≪ / RTI >

Represents the precoding matrix of the i < th > transmitter,

May represent a decoding matrix of an i-th receiving apparatus that is a pair with the i-th transmitting apparatus. At this time, in order to eliminate interference between users, a precoding matrix

And a decoding matrix

Must satisfy the following equations (3) and (4).

In Equation (4), d represents the number of transport streams.

According to Equations (3) and (4) above, the constraint for d can be expressed as Equation (5) below.

In Equation (5), M represents the number of antennas of the transmitting apparatus, N represents the number of antennas of the receiving apparatus, K represents the number of recognized wireless users (SU), and d represents the number of transport streams.

Equation (5) represents a condition that must be satisfied in order to enable interference alignment using a full-IA technique. First, a load can be given to a user PU. If full-IA techniques are used, the interference to the user (PU) may be eliminated first, but the data rate may be lower than that of a single user multi-antenna system of the same conditions. That is, the transmission rate of the user is limited as the PU removes the interference between all the cognizant wireless users. Accordingly, the user PU needs to first generate a precoding matrix for improving the transmission rate of the user. For example, a technique for generating a precoding matrix that improves a user's transmission rate may include a PIA-SU algorithm that considers interference between cognitive radio users (SUs) and a PIA-SU algorithm that does not consider interference between cognitive radio users (SUs) The NSU algorithm can be used. At this time, the interference between the acknowledged wireless users (SUs) can be firstly aligned in the user (PU). Accordingly, the precoding matrix of the user PU may first be generated to maximize the throughput of the user, and a decoding matrix may be generated to improve the performance of the perceived wireless users.

4 is a diagram illustrating a perceptual wireless network that performs interference alignment based on a PIA-SU algorithm, in one embodiment of the present invention.

4, the cognitive radio network 400 includes a transmitting device 401 and a pairing receiving device 405 of a single priority user (PU), a transmitting device 402, 403, 404 and a receiving device 406, 407, 408, which form a pair.

First, the PIA-SU algorithm can satisfy the condition of Equation (6) below in order to remove the interference between the cognitive radio users.

,

,

)은 우선사용자의 수신장치에서 동일한 서브공간(subspace), 즉, 간섭공간에 정렬될 수 있다. 예컨대, 상기 우선사용자(405)에게 미치는 간섭들이 정렬되는 서브공간은 우선사용자의 송신장치(401)로부터 수신한 신호가 정렬되는 서브공간과 직교(orthogonal)할 수 있다.That is, according to equation (6) and FIG. 4 above, the interference (s) from the cognizant wireless users 402,403, 404 to the first user 405

,

,

May first be aligned in the same subspace, i. E., An interference space, at the receiving device of the user. For example, the subspace in which the interference to the preferred user 405 is aligned may first be orthogonal to the subspace in which the signal received from the user's transmitter 401 is aligned.

,

), 즉, 인지무선 사용자들 간의 간섭은 동일한 서브공간에 정렬될 수 있다. 동일한 방법으로, 인지무선 사용자의 수신장치(407)에 간섭을 미치는 인지무선 사용자들 간의 간섭이 동일한 서브공간에 정렬되고, 인지무선 사용자의 수신장치(408)에 간섭을 미치는 인지무선 사용자들 간의 간섭이 동일한 서브공간에 정렬될 수 있다.All interferences between the perceived wireless users 402, 403, and 404 may then be aligned in the same subspace in the receiving device of the perceived wireless user. For example, the signals received at the receiving device 406 of the cognizant wireless user may be transmitted to the receiving device 406 of the cognizant wireless user and transmitted from the transmitting devices 403 and 404 of the cognitive wireless user, The signals received at the receiving device 406 (

,

), I.e., the interference between cognizant wireless users may be aligned in the same subspace. In the same way, the interference between the cognitive wireless users that are interfering with the receiving wireless device 407 of the cognizant wireless user is aligned in the same subspace and the interference between the cognizant wireless users that are interfering with the receiving wireless device 408 of the cognizant wireless user Can be aligned in the same subspace.

At this time, a modified recursive interference sorting algorithm may be used to satisfy Equation (6) above. In a recursive interference sorting algorithm, a forward direction algorithm is obtained based on Equation (7) below, and a reverse direction algorithm can be obtained based on Equation (8) below. [3] K. Gomadam, VR Cadambe , and SA Jafar , "A Distributed Numerical Approach to Interference Alignment and Applications to Wireless Interference Networks," IEEE Trans. Inf . Theory, vol. 57, no. 6, pp. 3309-3322, Jun . 2011. By modifying the recursive interference sorting algorithm according to Equation (7) and Equation (8) below, the covariance matrix in the forward direction and the covariance matrix in the reverse direction can be respectively calculated.

In Equations (7) and (8), P denotes transmit power, d denotes the number of transport streams, H denotes a channel coefficient, V denotes a precoding matrix, and Q denotes an interference covariance matrix.

According to Equation (7), the interference covariance of each of the signals received from other cognizant wireless users 403,404, including the user 401 first, is calculated, and an interference covariance matrix containing each calculated interference covariance is generated . At this time, eigenvalue decomposition is performed on the interference covariance matrix, so that a decoding matrix can be generated.

In Equation (8), the backward algorithm can consider only interference leakage at the receiving device of each recognized wireless user. Eigenvalue decomposition is performed on the interference covariance matrix according to Equation (8), so that a decoding matrix for the reverse direction can be generated. In this case, the forward algorithm and the backward algorithm in the recursive interference sorting algorithm, that is, the calculation according to Equation (7) and Equation (8) above, can be repeatedly performed alternately until the predetermined performance converges. Through the iterative operation of the forward and backward algorithms, a precoding matrix of the wireless perceptual user and a decoding matrix of the preferential user for eliminating the interference between the wireless perceptual users can be generated.

In the PIA-SU algorithm, the number of transmittable streams for each user may be equal to M, N, K, and the number of transmittable data streams of the PIA-SU algorithm may be greater than the full interference sorting algorithm (full IA) . In the case of the PIA-SU algorithm, only the interference of the signals from a plurality of cognitive wireless users to the users is taken into consideration. First, since the interference of the signals by the users to each cognitive wireless user is not considered, A larger number of data streams can be transmitted.

)와 전체 변수의 개수(

)는 아래의 수학식 9와 같이 표현될 수 있다.The number of total equations satisfying the condition of Equation (6) (

) And the total number of variables (

) Can be expressed by Equation (9) below.

)와 전체 변수의 개수(

)보다 크지 않아야 한다. 이에 따라, 수학식 6에서 식의 개수가 변수의 개수보다 크지 않아야 수학식 6의 조건이 만족될 수 있다. 그러면, 각 사용자의 최대 전송스트림의 수 d는 아래의 수학식 10과 같이 표현될 수 있다. 즉, PIA-SU 알고리즘을 이용하는 경우, 수학식 10에 기초하여 전송가능한 최대 전송 데이터 스트림의 수가 계산될 수 있다.According to Bezout's theory in equation (9), the condition that a polynomial has a solution is the number of total equations

) And the total number of variables (

). Accordingly, the condition of Equation (6) can be satisfied only when the number of equations in Equation (6) is not larger than the number of variables. Then, the maximum number of transport streams d of each user can be expressed by Equation (10) below. That is, when using the PIA-SU algorithm, the maximum number of transmittable data streams that can be transmitted can be calculated based on Equation (10).

In Equation (5), M represents the number of antennas of the transmitting apparatus, N represents the number of antennas of the receiving apparatus, K represents the number of recognized wireless users (SU), and d represents the number of transport streams.

5 is a diagram illustrating a perceptual wireless network that performs interference alignment based on a PIA-NSU algorithm, in one embodiment of the present invention.

5, the cognitive radio network 500 includes a transmitting device 501 and a pair of receiving devices 505 of a single priority user (PU), a transmitting device of each of a plurality of cognitive radio users 502, 503, 504, and a receiving device 506, 507, 508, which form a pair.

The PIA-NSU algorithm may first be conditional on that all interference is completely removed from the user's receiving device 505, and the condition can be expressed as Equation (11) below.

According to Equation 11 and FIG. 5, in the case of the PIA-NSU algorithm, all of the interferences are first sorted in the user 505, but all of the interferences can be aligned without considering the interference between the acknowledged wireless users. At this time, a method similar to PIA-SU can be applied to satisfy the condition of Equation (11). In the case of the forward algorithm, first, the interference covariance matrix of the user can be generated as shown in Equation (12) below, considering only interference leakage in the receiver 505 of the user.

In the case of the reverse algorithm in the PIA-NSU, the interference covariance matrix of the cognitive radio user can be expressed by Equation (13) below.

At this time, in the forward and reverse directions in the PIA-NSU algorithm, the corresponding decoding matrix can be obtained through eigenvalue decomposition of Equation (12) and Equation (13). Accordingly, the precoding matrix of the perceived wireless user and the preference user's decoding matrix can be obtained based on the forward and backward algorithms.

Then, the maximum number of transportable streams d in the PIA-NSU algorithm can be expressed by Equation (14) below.

According to Equations (14) and (10), it can be confirmed that the number (d) of transport streams when using the PIA-NSU algorithm is larger than the case of using the PIA-SU algorithm. In the PIA-SU and PIA-NSU algorithms, the receiving apparatus of the user may have only channel state information (CSI) from the transmitting apparatus corresponding to each of the plurality of cognitive radio users to the receiving apparatus of the user. In the case of full interference (full IA), channel information between a transmitting apparatus corresponding to each of a plurality of cognizant wireless users and a receiving apparatus of a priority user must first be known in both the transmitting apparatus and the receiving apparatus of the user. Therefore, when transmitting and receiving data using a PIA-SU and a PIA-NSU algorithm in a cognitive radio network, the overhead for a user can be significantly reduced compared to when using a full IA scheme In addition, the number of transmission data streams (d) that can be transmitted by the user first may also increase over the full IA scheme. In this case, the above-mentioned d may be larger than when the PIA-NSU algorithm is used, but the PIA-NSU algorithm does not consider the interference between the cognitive wireless users, -SU. ≪ / RTI >

In case of receiving low SNR and intermediate SNR in an interference-based cognitive wireless network, it may be difficult to satisfy the performance requirement of the user first. Accordingly, the user ' s precoding matrix may be first generated according to a decoding matrix to be orthogonal to all interferences from the cognizant wireless users to maximize the data rate.

First, the transmission capacity R of the user can be expressed by the following equation (15).

은 수신기의 잡음분산을 나타낼 수 있다.In Equation 15, R denotes a transmission capacity of a user first, I denotes an identity matrix, P denotes a transmission power, d denotes the number of transport streams, U denotes a decoding matrix, V denotes a precoding matrix, H denotes a channel coefficient,

May represent the noise variance of the receiver.

는 아래의 수학식 16의 조건을 만족해야 하며, 수학식 16에 대한 최적해는 아래의 수학식 17과 같이 표현될 수 있다.First, in order to maximize the data transfer rate of the user,

Must satisfy the condition of the following equation (16), and the optimal solution of the equation (16) can be expressed as the following equation (17).

는 특이값 분해를 통해 획득되는 좌측특이벡터 연산을 나타낼 수 있다. 수학식 17에 기초하여 우선사용자의 프리코딩 행렬

이 생성될 수 있다.In Equation 17,

Can represent a left singular vector operation obtained through singular value decomposition. Based on Equation (17), the user's precoding matrix

Can be generated.

의 특이값 분해(Singular Value Decomposition, SVD)는 아래의 수학식 18과 같이 표현될 수 있다.Proof:

The singular value decomposition (SVD) can be expressed by Equation (18) below.

각각은 특이값 분해를 통해 획득되는 값일 수 있다.In Equation 18,

Each may be a value obtained through singular value decomposition.

와 디코딩 행렬

에 기초하여 특이값 분해(SVD)를 수행함으로써, 우선사용자의 데이터 전송률을 최대화하는 우선사용자의 프리코딩 행렬이 생성될 수 있다.As shown in Equation (18), the channel coefficient between the transmitting apparatus of the user and the receiving apparatus of the preferred user

And a decoding matrix

By performing singular value decomposition (SVD) based on the precoding matrix, the precoding matrix of the preferred user that maximizes the data rate of the user first can be generated.

Then, the exponential part of the log of Equation (15) can be expressed as Equation (19) below.

From Equation (19), an optimal solution to Equation (16) can be obtained, and an optimal solution can be expressed as Equation (20) below.

는

의 특이값 분해를 통해 얻어지는 좌측 특이값 행렬을 나타낼 수 있다. 이처럼 위의 수학식 18 내지 수학식 20의 증명 과정을 통해, 위의 수학식 16의 최적해는 수학식 17과 같이 정의될 수 있으며, 우선사용자의 최적 프리코딩 행렬

은 우선사용자의 데이터 전송률을 최대화하도록 구성되며

와 동일한 신호부공간에 놓여질 수 있다.In equation (20)

The

The left singular value matrix obtained by the singular value decomposition of the left singular value matrix. Through the proof process of Equations (18) to (20), the optimal solution of Equation (16) can be defined as Equation (17)

Is first configured to maximize the user ' s data rate

Lt; RTI ID = 0.0 > a < / RTI >

By precoding and transmitting the transmission signal using the precoding matrix generated using Equation (17) above, the data rate of the user can be maximized first. However, since the interference to the cognizant wireless user is not considered, The QoS of the user can be attenuated. Accordingly, the decoding matrix of the perceived wireless user may be designed to improve the transmission performance of the perceived wireless user.

 Since the precoding matrix of the priority user generated according to Equation (17) above and the interference in the receiving apparatus of the perceived wireless user in the PIA-SU and PIA-NSU algorithms are not completely aligned, the received signal interference noise ratio (SINR ) May be used. The problem of generating a decoding matrix maximizing the SINR can be expressed as Equation (21).

는 아래의 수학식 22와 같이 표현될 수 있다.In Equation 21,

Can be expressed by the following equation (22).

에 대해 수학식 21의 최적 디코딩 행렬은 아래의 수학식 23에 따라 획득된 프리코딩 벡터

에 기초하여 생성될 수 있다.A given precoding matrix

The optimal decoding matrix of Equation (21) is obtained by using the precoding vector < RTI ID = 0.0 >

. ≪ / RTI >

By performing decoding on the received signal at the receiving device of the perceived wireless user based on the decoding matrix according to equation (23), the performance of the perceived wireless user can be improved. At this time, the QoS of the wireless user may not be guaranteed when the transmission power of the user is relatively strong. It is understood that improving the transmission power of the cognizant wireless user does not affect the performance of the user first because the interference to the first user from the cognizant wireless user is eliminated. Accordingly, if necessary, the transmission power can be increased to improve the performance of the wireless user. Even if the transmission power of the perceived wireless user is increased, the performance of the user can be maintained as before the transmission power increase.

In the proposed OTD scheme, the precoding matrix of the cognitive wireless user can be designed to cooperate to align the interference in the user's receiving apparatus. First, the precoding matrix of the user is PIA- SU and PIA-NSU algorithms are used to remove the interference, thereby increasing the data rate of the first user, but attenuating the performance of the cognitive radio user. Accordingly, in order to improve the performance of the cognitive radio user to be attenuated by precoding using the PIA-SU and PIA-NSU algorithms to transmit data, a decoding matrix maximizing the SINR of the cognitive radio user is transmitted to the receiving device ≪ / RTI >

및 인지무선 사용자의 프리코딩 행렬

이 생성될 수 있다. 그리고, 우선사용자의 디코딩 행렬

과 위의 수학식 17에 기초하여 우선사용자의 프리코딩 행렬

이 생성될 수 있다. 이어, 수학식 23에 기초하여 인지무선 사용자의 SINR을 최대화하는 디코딩 행렬

,

이 생성될 수 있으며, 간섭정렬 기반의 인지무선 네트워크에서 최적의 송수신기를 설계하는 알고리즘(OTD)은 아래의 표 1과 같을 수 있다.In summary, based on the PIA_SU algorithm or the PIA-NSU algorithm, the user's decoding matrix

And a precoding matrix of the cognitive radio user

Can be generated. Then, first, the user's decoding matrix

On the basis of Equation (17), the user's precoding matrix

Can be generated. Then, based on Equation (23), a decoding matrix for maximizing the SINR of the perceived wireless user

,

(OTD) for designing an optimal transceiver in an interference-based cognitive radio network may be as shown in Table 1 below.

이 1인 플랫 레일레히 블록 페이딩(flat Rayleigh block fading) 채널인 경우를 가정할 수 있다. 이때, 각 사용자의 전송 데이터 스트림의 수(d)는 위의 수학식 4, 수학식 10 및 수학식 14에 기초하여 계산될 수 있다. 예컨대, 완전간섭정렬을 이용한 경우의 전송 스트림의 수(d)는 3, PIA-SU 알고리즘을 이용한 경우의 전송 스트림의 수(d)는 3, PIA-NSU 알고리즘을 이용한 경우의 전송 스트림의 수(d)는 4로 계산될 수 있다.The transmission rates when the interference alignment is performed according to the PIA-SU and PIA-NSU algorithms based on Table 1 above may be as shown in FIG. In FIG. 6, the simulation assumes an interference-based cognitive radio network consisting of three cognizant wireless users and one priority user. The cognitive wireless user's transmitter and receiver, and firstly the user's transmitter and receiver, Antenna, and the transmission channel has a noise variance

It can be assumed that the channel is a flat Rayleigh block fading channel. At this time, the number (d) of transmission data streams of each user can be calculated based on Equation (4), Equation (10) and Equation (14). For example, the number (d) of transport streams when using the full interference alignment is 3, the number of transport streams (d) when using the PIA-SU algorithm is 3, the number of transport streams using the PIA- d) can be calculated as 4.

In Figure 6, graph 601 shows the performance of the user PU first, and graph 602 shows the performance of the perceived wireless user (SU).

6, it can be seen that the QoS of the user PU is significantly increased in the case of using the PIA-SU 620 and the PIA-NSU 610 rather than using the full interference 630 have. It can be confirmed that the PIA-NSU 610 further increases the transmission performance of the user PU prior to the PIA-SU 620. It can be seen that the PIA-SU 650 increases the transmission performance of the perceived wireless user more than the PIA-NSU 640. The results of the simulation according to FIG. 6 in the full IA, PIA-SU, and PIA-NSU algorithms can be summarized as shown in Table 2 below.

algorithm Overhead Complexity The rate of the PU The SU rate Full IA Prize Prize Ha Prize PIA-SU medium medium medium medium PIA-NSU Ha Ha Prize Ha

According to Table 2, PIA-SU. POA-NSU and full IA may be selectively used to generate a precoding matrix suitable for the purpose of use. For example, if the purpose of use is not concerned with the performance of the perceived wireless user and only the protection objective for the user (PU) is considered, the precoding matrix may be generated using the PIA-NSU. When the purpose of use is to consider the interference between the perceived wireless users, that is, when the performance of the perceived wireless users should be guaranteed to a certain level, a precoding matrix may be generated based on the PIA-SU. And, if a collaborative network is used between a user and a cognizant wireless user, a precoding matrix can be generated based on a perfect interference sorting technique.

As such, the present invention can reduce the overhead and complexity of the system by creating precoding matrices using partial interference alignment (PIA) to align the interference, as well as reduce the overhead and system complexity of the PIA-SU and PIA-NSU May be used to maximize the performance of the user and improve the performance of the perceived wireless user (SU).

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (6)

A method of interference arrangement in a transmitter in a cognitive radio network including a primary user having a pair structure and a plurality of cognitive radio users (Secondary User) based on a transmitting apparatus and a receiving apparatus based on a multiple antenna,
A first interference alignment technique for aligning interferences between a plurality of cognitive radio users (SU) participating in the perceptual wireless network and a first user (PU) and interferences between a plurality of perceived wireless users; A second interference sorting technique for sorting the interference on the first user from the perceived wireless users except for the interference among the wireless users, determining the interference sorting technique according to the purpose of use; And
Generating a precoding matrix for eliminating interference to the preferred user based on the determined interference alignment technique
Lt; / RTI >
The method of claim 1, wherein the first interference alignment technique further comprises: receiving from the plurality of cognizant wireless users (SU) a priority user (PU), excluding interference on each of the plurality of cognizant wireless users (SU) from the priority user And a technique for sorting the interference among a plurality of cognizant wireless users,
Wherein generating the precoding matrix comprises:
Calculating an interference covariance of a signal received from a different cognizant wireless user from a signal of the preferred user according to the first interference alignment technique if the determined interference alignment technique is the first interference alignment technique;
Performing eigenvalue decomposition on an interference covariance matrix including the calculated interference covariance;
Generating the precoding matrix based on the eigenvalue vectors obtained through the eigenvalue decomposition;
If the determined interference alignment technique is the second interference alignment technique, an interference covariance matrix is generated using only the signal received from the priority user, except for signals received from other cognitive radio users according to the second interference alignment technique ;
Performing eigenvalue decomposition on the generated interference covariance matrix; And
Generating the precoding matrix including eigenvalue vectors obtained through the eigenvalue decomposition
Wherein the interference is received by the receiver. delete The method according to claim 1,
Wherein the precoding matrix generated according to the first interference alignment technique is a precoding matrix for reducing or minimizing interference to the other cognizant wireless users
Gt; interference < / RTI > in a cognitive radio network. delete The method according to claim 1,
Wherein the precoding matrix generated according to the second interference alignment technique is a precoding matrix for reducing or minimizing interference affecting the preferred user
Gt; interference < / RTI > in a cognitive radio network. The method according to claim 1,
Wherein when the receiving device is any one of the plurality of cognizant wireless users, the receiving device transmits a signal received from the transmitting device corresponding to the preferred user and a signal received from any one of the plurality of cognizant wireless users A decoding matrix is generated in which a signal to interference noise ratio (SINR) of the perceived wireless user is maximized based on a signal received from a transmitting apparatus corresponding to another cognizant wireless user,
Wherein generating the precoding matrix comprises generating the precoding matrix according to the decoding matrix.

Images