KR101199569B1 - Opportunistic interference aligned user selection in multiuser mimo interference channels - Google Patents

Abstract

A method for opportunistic interference alignment in a MIMO interference channel having a plurality of users, the method comprising: at least one transmitter each broadcasting a random beam; wherein the plurality of users each correlates the interference channels; Feeding back a transmitter of the apparatus; The at least one transmitter discloses a method of opportunistic interference alignment, comprising selecting a user whose interference signals are best aligned among the plurality of users. The method of opportunistic interference alignment according to the present invention may further include performing beamforming by the selected user, and further comprising calculating a postprocessing vector of the selected user after selecting the user. The performance is superior to conventional opportunistic user selection schemes, while reducing the amount of feedback information and significantly reducing the complexity of the calculation.

Description

OPPORTUNISTIC INTERFERENCE ALIGNED USER SELECTION IN MULTIUSER MIMO INTERFERENCE CHANNELS}

The present invention relates to an interference alignment technique for interference management in wireless communication, and more particularly, to an opportunistic interference alignment user selection scheme in a MIMO interference channel having multiple users.

As the demand for high data rates increases, interference management in wireless communications becomes even more important. However, conventional interference management techniques, such as frequency reuse, orthogonalization, power control, etc., have been difficult to efficiently manage the interference caused by high data rate services. This problem in interference management is exacerbated in the cell edge region in the call zone method (cellular system). Recognizing that interference management is essential to achieve high spectral efficiency, developers have developed a variety of interference management technologies.

Recently, there is a growing interest in interference alignment technology, the basic concept of which is to design a transmitter and a receiver such that interference signals at the receiver are aligned with each other. According to one conventional study, the capacity of an interference channel (IC) of two users is set within 1 bit. According to another conventional study, the degrees of freedom (DoF) of K user interference channels (ICs) are K / 2. In addition, another study proposes an iterative interference alignment algorithm for a distributed system, and a subspace interference alignment technique for an uplink cellular network system.

However, conventionally discussed interference alignment techniques not only incur significant signal overhead by sharing channel state information between transmitters, but also inaccurate channel state information in a limited feedback environment, which can greatly weaken the performance of interference alignment. In addition, there is a disadvantage that the interference alignment techniques discussed in the prior art are complicated to calculation.

In order to overcome the above-mentioned disadvantages, a novel opportunistic interference alignment method and an opportunistic interference alignment user selection method, which reduces the amount of feedback information and significantly lowers the complexity of the calculation, and outperforms conventional interference alignment schemes, Required.

In order to solve the above problems, the present invention provides a method for opportunistic interference alignment in a MIMO interference channel in which a plurality of users exist, the method comprising: at least one transmitter broadcasting a random beam; A plurality of users each feeding back a correlation of interfering channels to at least one transmitter; At least one transmitter discloses a method of opportunistic interference alignment, comprising selecting a user whose interference signals are best aligned among a plurality of users.

In this opportunistic interference alignment method, the correlation of interference channels that a plurality of users feedback to at least one transmitter indicates the amount or measure by which the interference signals are aligned.

The opportunistic interference alignment method according to the present invention may further include performing beamforming by the selected user.

The opportunistic interference alignment method according to the present invention may further include calculating a postprocessing vector of the selected user after selecting the user.

In the opportunistic interference alignment method according to the present invention, information sharing is not performed between transmitters.

In the opportunistic interference alignment method according to the present invention, the postprocessing vector can be determined to maximize the signal to interference and noise ratio (SINR).

In the opportunistic interference alignment method according to the present invention, the postprocessing vector can be determined to minimize the interference to noise ratio (INR).

In the opportunistic interference alignment method according to the present invention, the postprocessing vector is determined by projecting the user's signal in the null space of the aligned signal to eliminate the interference using a separate degree of freedom (Extra DoF). Can be.

In the opportunistic interference alignment method according to the present invention, the interference signals may be quantized into one unit vector. This quantization may be optimal quantization or suboptimal quantization.

The present invention further discloses a hybrid interference alignment method in a MIMO interference channel having a plurality of users, the hybrid interference alignment method comprising the steps of: at least one transmitter each broadcasting a random beam; Multiple receivers project their signals into null spaces of the aligned signals to eliminate interference using a first postprocessing vector that maximizes the signal-to-noise ratio (SNR) and a separate extra doF Finding a second postprocessing vector determined by the application; Calculating a first and second signal to interference and noise ratio (SINR) corresponding to the first and second postprocessing vectors by a plurality of receivers of the user; A receiver of a plurality of users feeding back a greater of the first and second signal to interference and noise ratios (SINRs) to at least one transmitter; The at least one transmitter comprises selecting a user among the plurality of users with the largest feedback signal to interference and noise ratio (SINR).

The hybrid interference alignment method is a mixture of the novel interference alignment method of the present invention and the conventional maximum signal-to-noise ratio (SNR) user selection method, and may further include performing beamforming by the selected user.

The present invention proposes practical interference alignment techniques based on opportunistic user selection in a limited feedback environment. Unlike conventional opportunistic beamforming, the random beam broadcast at the transmitter is capable of interference alignment user selection for other transmitters. help. Once the interference alignment user is selected, the selected user performs beamforming with the provided receive antenna. Since the postprocessing vector is multiplied after the interference alignment user has been selected, there is no need for all users to calculate the postprocessing vector, thereby greatly reducing the complexity of the calculation compared to conventional opportunistic user selection schemes.

As described above, the opportunistic interference alignment method including the OAPUS (Opportunistic Interference Aligned User Selection) according to the present invention is superior in performance and feedback compared to conventional opportunistic user selection methods. The amount of information is reduced and the computational complexity is significantly reduced.

1 is a schematic diagram illustrating a system model with a 3-user interference channel according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating the performance of various opportunistic user selection schemes according to an embodiment of the present invention compared with the performance of conventional schemes.
FIG. 3 is a diagram comparing the performance of the OIAUS schemes proposed in the present invention and the conventional schemes when ISR = 0.6 according to an embodiment of the present invention.
4 is a diagram illustrating performance of user selection schemes according to a change in the number K of users in each user group when SNR = 10dB and ISR = 0.7, in comparison with the performance of conventional schemes, according to an embodiment of the present invention. to be.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be understood, however, that the invention is not limited to the disclosed embodiments, but may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, Is provided to fully inform the user. In addition, for convenience of explanation, components may be exaggerated or reduced in size.

I. System Model

Referring now to Figure 1, as a system model according to an embodiment of the present invention, three transmitters (101, 102, 103) with two antennas and corresponding three user groups (201, 202) 203 is shown where there is a three-user interference channel (IC). In the first, second and third user groups 201, 202 and 203, there are K users each having two receive antennas. The exchange of information between the transmitters 101, 102, 103 is not allowed. Each transmitter (101, 102, 103) broadcasts a random beam and opportunistically selects among the K users to communicate with that transmitter. Thus, the transmitter and the selected receiver build a three-user MIMO IC. Although the transmitters 101, 102, 103 can support up to two independent streams over a 2 x 2 MIMO channel, each transmitter only has a single stream to protect the dimensions for interference alignment at each receiver. Send. When M transmit antennas and N receive antennas are used, MN / (M + N−1) streams may be transmitted to achieve maximum degrees of freedom (DoF) that can be achieved.

을 만족하고, 이때 i번째 송신기에 대한 i번째 사용자 그룹 내의 k번째 사용자에서 수신된 신호는 다음과 같이 주어진다.If we set the random unit beamforming vector at the i th transmitter to w _i , w _i is a condition.

Is satisfied, and the signal received at the k-th user in the i-th user group for the i-th transmitter is given by

는 j번째 송신기로부터 i번째 사용자 그룹 내의 k번째 사용자로의 채널 행렬을 나타내고,

의 각 성분은 평균이 0이고 분산이 1인 i.i.d(independent and identically distributed) 복소 가우시안 확률 변수이다. i번째 송신기로부터의 송신 신호는 x_i로 표시되고 이는

로 주어지는 송신 파워 제약조건을 만족하며,

는 0의 평균과 단위 복소 공분산 행렬

을 갖는 복소 가우시안 분포를 따르는 가우시안 잡음을 나타낸다. α는 간섭 채널들의 상대적인 전파 경로 손실을 나타낸다. 각 간섭 송신기로부터 수신된 평균 간섭 파워가

가 되므로, α는

와 같이 정해지고, 간섭 대 신호 파워 비(ISR)를 나타낸다. 이 식에서, INR은 간섭 대 잡음 비를 나타내고, SNR은 신호 대 잡음비를 나타낸다. 포스트프로세싱 벡터

를 이용하여, 포스트프로세싱 후에 수신된 신호는 다음과 같은 식으로 나타낼 수 있다:In the above equation,

Denotes the channel matrix from the j th transmitter to the k th user in the i th user group,

Each component of is an iid (independent and identically distributed) complex Gaussian random variable with mean 0 and variance 1. The transmitted signal from the i th transmitter is denoted by x _i , which is

Satisfies the transmit power constraint given by

Is a unit complex covariance matrix with mean of 0

Represents a Gaussian noise that follows a complex Gaussian distribution. α represents the relative propagation path loss of the interfering channels. The average interference power received from each interfering transmitter

Since α is

And denotes the interference-to-signal power ratio (ISR). In this equation, INR represents the interference-to-noise ratio and SNR represents the signal-to-noise ratio. Postprocessing vector

Using, the signal received after postprocessing can be represented as:

는 복소 공액 전치(conjugate transpose)를 나타낸다. 따라서, 신호 대 간섭 및 잡음비(Signal-to-Interference plus Noise Ratio; SINR)는 다음과 같은 식으로 정해진다.Where superscript

Denotes conjugate conjugate transpose. Therefore, signal-to-interference plus noise ratio (SINR) is determined as follows.

Ⅱ. Conventional Opportunity User Choice

를 최적화할 수 있다. 본 발명에 따른 간섭 정렬 사용자 선택 방식의 이점을 설명하기 위하여, 본 섹션에서는 현재 논의되고 있는 3가지 기회적 사용자 선택 기준과 그에 대응하는 포스트프로세싱 벡터 설계를 언급하고자 한다.Since each transmitter only transmits a single stream, receivers postprocess the vector according to a given user-selected scheme.

Can be optimized. To illustrate the benefits of the interference alignment user selection scheme in accordance with the present invention, this section refers to three opportunistic user selection criteria that are currently discussed and corresponding postprocessing vector designs.

A. Maximum SNR User Selection (MAX-SNR)

과 같이 주어지고, 이에 대응되는 SNR은

이 된다.In the maximum SNR user selection scheme, the user with the highest SNR is selected regardless of the interference. Each user uses a postprocessing vector that maximizes the SNR and feeds back the post-processing SNR to the transmitter. The postprocessing vector of the j user in the i group is

And the corresponding SNR is

.

B. Maximum INR user selection (MIN-INR)

In this way, user selection is made based on the INR after postprocessing, where each user feeds back their INR to the transmitter and the user with the lowest INR is selected. For INR minimization, the postprocessing vector of the k th user in the i th user group is given by:

는 최소 고유값

에 대응되는 행렬

의 고유벡터이다. 포스트프로세싱 후의 INR은

이 된다. 모든 사용자들은 최소 고유값과 그에 대응되는 고유벡터들을 찾아내기 위해 고유값분해(EVD)를 수행할 필요가 있다.here,

Is the minimum eigenvalue

Matrix corresponding to

. INR after postprocessing

. All users need to perform eigenvalue decomposition (EVD) to find the minimum eigenvalues and their corresponding eigenvectors.

C. Maximum SINR User Choice (MAX-SINR)

In this selection scheme, the user may be selected to maximize SINR after postprocessing. The postprocessing vector of the k-th user in the i-th user group is given by

Here, Vmax (?) Is the eigenvector of the matrix (?) Corresponding to the maximum eigenvalue.

이다.Also,

to be.

이 되고, 여기서 λ_max(?)는 행렬 (?)의 최대 고유값이다. MIN-INR 선택과 유사하게, 모든 사용자들은 최대 SINR을 가진 사용자를 선택하기 위하여 행렬 역변환을 따라 EVD를 수행하여야 한다.The SINR value fed back to the transmitter

Where λ _max (?) Is the maximum eigenvalue of the matrix (?). Similar to the MIN-INR selection, all users must perform an EVD along the matrix inverse transformation to select the user with the maximum SINR.

Ⅲ. Opportunity Interference Alignment User Selection Method According to the Invention

The opportunistic interference alignment user selection (OIAUS) according to the present invention selects a user whose interference signals are best aligned with each other. The amount of alignment is measured by the correlation between the interfering signals. In particular, the correlation of the interference signals of the k-th user in the i-th user group is given by the following equation.

Although the number of interfering signals is greater than the degree of freedom for canceling interference at the receiver, interfering signals from other transmitters can be more easily canceled as the number of users increases, which increases the chance of selecting a user with well-aligned interfering signals. Because. Unlike opportunistic beamforming in MIMO BC, in the opportunistic interference alignment user selection according to the present invention, the random beam of the transmitter helps other transmitters other than the transmitter to select the users whose interference signals are best aligned. .

Once the transmitter selects the interference aligned user, the postprocessing vector on the selected user's side can be determined according to various criteria, as several embodiments are presented below as reference.

A. Maximizing SINR after OIAUS (OIAUS-SINR)

는 다음 식과 같이 정해진다.In this embodiment, the selected user maximizes SINR by applying the appropriate postprocessing vector. Postprocessing vector to maximize SINR for selected user k in the i th user group

Is determined by the following equation.

Looking at the above equation, the final equation is the same as the maximum SINR user selection (MAX-SINR) introduced above and requires the same matrix inverse transform and eigenvalue decomposition (EVD), but unlike the MAX-SINR calculation of the postprocessing vector It is to be understood that the difference is that the operation only needs to be performed for the selected user and not all users.

B. Minimize INR after OIAUS (OIAUS-INR)

In this embodiment, when the design criterion of the postprocessing vector is to minimize INR, the postprocessing vector of the kth user in the ith user group is given by the following equation.

This postprocessing vector is the same as the postprocessing vector at the maximum INR user selection (MIN-INR) introduced above, with the difference that unlike MIN-INR, not all users need to calculate the postprocessing vector. Eigenvalue decomposition only needs to be performed on the selected user side.

C. Projection on nullspaces of aligned interference channels

, i번째 사용자 그룹 내의 k번째 사용자)로 양자화되어야 한다. 이때, 투사 행렬

은

과 같이 정해진다.Since only a single data stream is transmitted from the transmitter over a 2 x 2 MIMO channel, in this embodiment, the selected user has his or her own in the nullspace of the aligned signal to eliminate interference using extra degrees of freedom (extra DoF). Project the signal. However, when the number of users is limited, the interference signals are not easy to be perfectly aligned.

, the k-th user in the i-th user group). Where the projection matrix

silver

Is determined as follows.

Where I is the eigen matrix. Accordingly, the postprocessing vector is derived as in the following equation.

는, 하기와 같이 최적으로 얻어지거나 서브옵티멀(suboptimal)하게 얻어질 수 있다.Quantization Vector of Two Interfering Signals

May be optimally obtained or suboptimally obtained as follows.

)가 구축되어,

과 같은 식을 얻어낼 수 있다. 여기서,

는 i번째 사용자 그룹 내의 k번째 사용자 측에서 인지된 간섭이고, 위에서 소개한 MIN-INR의 식에 기술되어 있다. 투사는 간섭으로부터의 영향을 최소화하기 위한 것이므로, 최적 양자화벡터(

)를 구비한 포스트프로세싱 벡터는 OIAUS-INR에서 INR을 최소화하는 포스트프로세싱 벡터와 동일해진다.1) Optimal Quantization: The optimal quantization vector (

) Is built,

You can get something like here,

Is the perceived interference at the k-th user side in the i-th user group and is described in the MIN-INR equation introduced above. Since the projection is intended to minimize the effects from the interference, the optimal quantization vector (

Postprocessing vector with) becomes the same as postprocessing vector that minimizes INR in OIAUS-INR.

)를 얻어내는 과정은 고유값분해(EVD) 수행을 필요로 하므로, 계산의 복잡성을 감소시키기 위해 2개의 간섭 신호들의 양자화를 보다 단순화시키는 방안을 고려할 수 있다. 선택된 사용자 측에서의 2 개의 간섭 채널 벡터들이

과

라면, 평균 간섭 벡터는

이 되고, 이는 효율적 간섭 신호 벡터로 간주된다. 그러므로, 간섭 신호들에 대한 양자화된 기저 벡터는

가 되고, 포스트프로세싱 벡터는

및

프로세서를 거쳐 정해진다. 이 단순화된 양자화는 단지 선형 연산만을 필요로 하므로, 다른 방식들에 비해 계산이 훨씬 단순해진다. 사용자들의 수가 무한하게 커짐에 따라, 간섭 신호들의 서브옵티멀 양자화는 최적 양자화에 가까워지는데, 이는 간섭 신호들의 평균 각도가 K와 함께 감소하기 때문이다. 2) Suboptimal Quantization (OIAUS-AVG): Optimal Quantization Vector (

) Requires performing eigenvalue decomposition (EVD), so we may consider a way to simplify the quantization of the two interfering signals to reduce the computational complexity. Two interference channel vectors at the selected user side

and

, The mean interference vector is

This is regarded as an efficient interference signal vector. Therefore, the quantized basis vector for interfering signals

And the postprocessing vector

And

Determined by the processor. This simplified quantization only requires linear operations, which makes the calculation much simpler than the other schemes. As the number of users grows infinitely, the sub-optimal quantization of the interfering signals approaches the optimal quantization because the average angle of the interfering signals decreases with K.

D. Hybrid with OIAUS-AVG and MAX-SNR

The OIAUS proposed in the present invention is expected to show excellent performance due to a significant reduction in complexity compared to the conventional MAX-SNR user selection scheme described in Section II A in an interference limited environment, that is, an area with high SNR. However, it is known that MAX-SNR user selection may be optimal in low SNR regions.

와

를 찾아낸다. 이때, 각 수신기는 프로세싱 벡터들에 대응되는 SINR을 계산하고

와 같이 둘 중 더 큰 값을 송신기에 피드백한다. 여기서, SINR(v')는 포스트프로세싱 벡터로서 v'를 사용한 SINR 값이다. 최종적으로, 각 송신기는 최대 SINR을 가진 사용자를 선택한다.MRC receivers require only linear operations, similar to OIAUS-AVG, so we can consider a hybrid scheme that combines OIAUS-AVG and MAX-SNR to improve performance while minimizing the increase in complexity. In particular, each receiver has two processing vectors,

Wow

Find it. In this case, each receiver calculates an SINR corresponding to the processing vectors.

Feedback the larger of the two to the transmitter. Here, SINR (v ') is an SINR value using v' as a postprocessing vector. Finally, each transmitter selects the user with the maximum SINR.

E. Complexity Analysis

이 되고,

는 big-O 표기법이다. 일반적으로, n x n 정방행렬의 채널 반전은

연산들을 필요로 하고, m x n 행렬의 SVD는

연산들을 필요로 한다. 따라서, MAX-SINR 방식 및 MIN-INR 방식의 복잡도는

이 된다.Referring to Table 1 below, the complexity of each approach described above is summarized. N represents the number of receive antennas, and CI represents the operation for channel inversion. In conventional MAX-SNR user selection, all users calculate their SNR using a postprocessing vector that maximizes the SNR, so the complexity is

Lt; / RTI &

Is the big-O notation. In general, the channel inversion of the nxn square matrix is

Operations, and SVD of an m-by-n matrix

Requires operations. Therefore, the complexity of the MAX-SINR and MIN-INR methods

.

차수(order)에서 연산을 해야하므로, OIAUS는

연산들을 필요로 한다. 포스트프로세싱 벡터에 따라, OIAUS의 복잡도는 증가한다. OIAUS-SINR 방식 및 OIAUS-INR 방식에 있어서, 선택된 사용자는 빔 형성 벡터들을 계산하는데, 이 계산에 있어서 채널 반전 및 EVD를 위해

연산들이 요구된다. OIAUS-AVG 방식에서는, 포스트프로세싱 벡터를 찾는 과정의 복잡도는 간섭 신호 벡터들의 단순화된 양자화로 인하여

이 된다. 모든 사용자가 2개의 빔 형성 벡터들을 찾아야 하지만, 하이브리드 방식에서는 각 사용자가 2 개의 빔 형성 벡터를 찾아내는 과정은 단지

연산만을 필요로 하므로, 하이브리드 방식의 복잡도는

이 된다. For all users to calculate the correlation of the interfering channels

Since we need to operate on orders, OIAUS

Requires operations. Depending on the postprocessing vector, the complexity of OIAUS increases. In the OIAUS-SINR scheme and the OIAUS-INR scheme, the selected user calculates beamforming vectors, for channel inversion and EVD in this calculation.

Operations are required In the OIAUS-AVG scheme, the complexity of finding the postprocessing vector is due to the simplified quantization of the interfering signal vectors.

. Every user must find two beamforming vectors, but in a hybrid approach, the process for each user finding two beamforming vectors is only

Because it only requires computation, the complexity of the hybrid approach

.

IV. Numerical Analysis and Results

In this section, we will compare and evaluate the performance of the OIAUS of the present invention with conventional user selection schemes. 2 illustrates the performance of various opportunistic user selection schemes in accordance with one embodiment of the present invention, where the ISR (α) is set to 0.2 and the number of users K is set to 10. FIG. In FIG. 2, it is shown that the interference alignment user selection scheme shows similar performance with much lower complexity than the MIN-INR user selection scheme. It can be seen that the performance degradation of OIAUS-AVG by simplified quantization is insignificant compared to OIAUS-INR. In addition, in FIG. 2, the OIAUS proposed in the present invention shows a much better performance than the conventional MAX-SNR in a high SNR region. In addition, it can be seen that a hybrid scheme for improving performance in a low SNR region has better performance in almost all SNR regions than the conventional MAX-SNR and MIN-INR user selection schemes. That is, it is to be understood that the OIAUS schemes proposed in the present invention show notable performance but are much lower in complexity than the conventional MAX-SNR and MIN-INR.

3 illustrates the performance of the OIAUS schemes and conventional schemes proposed in the present invention when ISR = 0.6, according to an embodiment of the present invention. The number of users and other configurations are the same as in FIG. Since the ISR is increased compared to Fig. 2, the interference becomes more important and the performance of the MAX-SNR is greatly weakened. The OIAUS schemes proposed in the present invention exhibit similar performance to the conventional MIN-INR user selection scheme.

4 illustrates the performance of user selection schemes according to the change in the number K of users in each group when SNR = 10dB and ISR = 0.7, according to an embodiment of the present invention. Referring to FIG. 4, as the number of users increases, the OIAUS schemes according to the present invention show a performance scale similar to that of the conventional MIN-INR user selection scheme. The conventional MAX-SINR user selection scheme and the present invention are proposed. While the performance difference between the hybrid schemes increases with K, as the K value increases, the complexity of the conventional MAX-SINR user selection scheme increases much faster than the hybrid scheme proposed in the present invention.

In the above, according to the present invention, a so-called opportunistic interference alignment user selection scheme (OIAUS), which is a practical interference alignment scheme based on opportunistic user selection in MIMO interference channels, has been proposed and described. The optimal postprocessing vector and the sub-optimal postprocessing vector were also analyzed, and the mathematical and numerical results presented above provide feedback, while the OIAUS according to the present invention shows notable performance compared to conventional opportunistic user selection schemes. Not only does it significantly reduce the amount of information, it also shows that it has significantly reduced the complexity of the computation. In addition, the hybrid method, which combines OIAUS and MAX-SNR methods, has shown tremendous performance improvements, especially in areas of low SNR, with minimal increase in complexity.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the invention. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

Claims (9)

An opportunistic interference alignment method in a MIMO interference channel having a plurality of users,
At least one transmitter each broadcasting a random beam;
The receiver of the plurality of users respectively feeding back correlation of interfering channels to the at least one transmitter;
Selecting, by the at least one transmitter, a user whose interference signals are best aligned among the plurality of users;
Calculating a postprocessing vector of the selected user,
The user with the best alignment of the interference signals is determined by the correlation between the interference signals,
The postprocessing vector is an opportunistic interference alignment method, wherein the selected user is determined by projecting his or her signal in a null space of an aligned signal to eliminate interference by using an extra degree of freedom. . The method of claim 1,
And performing beamforming by the selected user. delete The method according to claim 1 or 2,
Opportunity interference alignment method characterized in that no information sharing between the at least one transmitter. The method according to claim 1 or 2,
Wherein the postprocessing vector is determined to maximize signal-to-interference and noise ratio (SINR). The method according to claim 1 or 2,
Wherein the postprocessing vector is determined to minimize an interference-to-noise ratio (INR). delete The method according to claim 1 or 2,
Opportunity interference alignment method characterized in that the interference signals are quantized into one unit vector. A hybrid interference alignment method in a MIMO interference channel having a plurality of users,
At least one transmitter each broadcasting a random beam;
The receivers of the plurality of users each have their first postprocessing vector maximizing signal-to-noise ratio (SNR) and separate signals in null spaces of the aligned signals to eliminate interference using separate extra degrees of freedom. Finding a second postprocessing vector determined by projection;
Calculating, by the receivers of the plurality of users, first and second signal to interference and noise ratios (SINRs) corresponding to the first and second postprocessing vectors;
The receiver of the plurality of users feeding back a greater of the first and second signal to interference and noise ratios (SINRs) to the at least one transmitter;
And selecting, by the at least one transmitter, a user among the plurality of users having the largest feedback signal-to-interference and noise ratio (SINR).

Images