KR102612385B1 - Cooperative Relaying method for Multi-User MIMO Wireless Backhaul Networks - Google Patents

Abstract

A cooperative communication method for a wireless backhaul network using a multiple relay broadcasting system including one transmitter, N repeaters, and K destination receivers, wherein the transmitter transmits to the N repeaters and K destination receivers in a first time slot. Streams are equally distributed and transmitted, but in the first time slot, the repeaters do not transmit the received signal stream, and the transmitting end equally distributes the streams to be transmitted to the N repeaters and K destination receiving ends in the second time slot. The stream is distributed and transmitted, and the repeaters extract a stream to be transmitted to each target receiving end from their own received signal in the first time slot and then control the stream to be transmitted to the corresponding target receiving end in the second time slot.

Description

Cooperative Relaying method for Multi-User MIMO Wireless Backhaul Networks}

The present invention relates to a cooperative communication method for a multi-user wireless backhaul network.

5G and 6G communication systems aim to improve the 4G communication system in various aspects including data rate, latency, energy, and cost, including increasing the transmission rate by more than 1,000 times compared to the existing 4G communication system.

To achieve this, methods such as cell densification, use of the wide mmWave frequency band, and ways to increase frequency efficiency using multi-antenna transmission techniques can be mainly considered.

Cell densification is an efficient means of increasing area capacity, which improves spectrum reuse and reduces the effective number of users within each cell (HS Dhillon, RK Ganti, F. Baccelli, and JG Andrews, “and analysis of K- tier downlink heterogeneous cellular networks,”IEEE J. Sel. Areas Commun., vol. 30, no. 3, pp. 550-560, Apr. 2012.) When cell density becomes too high, it is difficult to use wired backhaul due to high costs. In these cases, wireless backhaul is presented as an attractive and practical solution (S. Chia, M. Gasparroni, and P. Brick, “next challenge for cellular networks: Backhaul,” IEEE Microw. Mag. , vol. 10, no. 5, pp. 54-66, Aug. 2009). In other words, for cell densification, there is a need to replace the existing wired backhaul with a wireless backhaul in terms of cost, efficiency, and performance. Wireless backhaul mainly uses line-of-sight communication, and the line-of-sight channel environment creates an environment in which the dominant signal path between the transmitter and destination receiver is finite and multiple paths are not sufficiently secured. . am. Accordingly, it was not easy to increase frequency efficiency because sufficient multiplexing gain could not be obtained even by using multiple antennas.

Considering that backhaul is communication between multiple base stations and center units, there is a need to support high data rate transmission, and the development of wireless transmission technology is necessary to overcome this problem.

Millimeter wave (mmWave) systems used in communication channels can be achieved using massive multiple-input multiple-output (MIMO) or beamforming with highly directional antennas to overcome high path loss in mmWaving channels. Requires large directivity gain. As a result, the effective channel of millimeter wave wireless backhaul has an inherently large line-of-sight (LoS) component. For these poor distribution environments, in contrast to rich distribution environments where the gain is proportional to the number of antennas, the multiplexing gain from MIMO is severely limited even when large numbers of antennas are used. This is analyzed to be due to the channel matrices becoming rank-deficient due to the keyhole effect. Therefore, considering the need for wireless backhaul to support huge data rates, it is necessary to improve the spectral efficiency of MIMO transmission despite rank-starvation of the channel.

In a multi-antenna system, the data transmission rate can be increased by utilizing the same frequency and time resources, but transmitting data in parallel to multiple layers in space through multiple antennas. When transmitting data in parallel in space through multiple antennas, rank information provides prior information about how many layers can be transmitted simultaneously. In this case, the terminal (UE) or MIMO receiver analyzes the downlink wireless channel environment, determines the rank of the multi-antenna channel, and reports it to the base station as a reverse channel. The more accurate this rank information is, the more the terminal can receive parallel data transmitted by the base station without error and the optimal downlink data transmission rate can be achieved.

Existing 3rd and 4th generation mobile communication systems such as LTE/LTE-A utilize MIMO technology, which transmits using multiple transmitting and receiving antennas to increase data transmission rate and system capacity. MIMO technology utilizes a plurality of transmitting and receiving antennas to spatially separate and transmit a plurality of information streams. In this way, spatially separating and transmitting a plurality of information streams is called spatial multiplexing. In general, how many information streams can be applied to spatial multiplexing depends on the number of antennas of the transmitter and receiver. In general, the number of information streams to which spatial multiplexing can be applied is referred to as the rank of the transmission.

Republic of Korea Patent Publication No. 10-1123222 provides a method of estimating a wireless channel rank by determining the number of available spatially parallel communication streams between the transmitting antenna array and the receiving antenna array from the mutual ratios of the eigenvalues of the channel matrix. I'm doing it.

Meanwhile, there have been several studies to solve the data transmission rate by adopting MIMO broadcast channel technology with a repeater (「B.Wang, J. Zhang, and A. Host-Madsen, “the capacity of MIMO relay channels,” IEEE Trans. Inf . Theory , vol. 51, no. 1, pp. 29-43, Jan. 2005.」, 「R. Zhang, CC Chai, and Y. Liang, “beamforming and power control for multiantenna relay broadcast channel with QoS constraints, ” IEEE Trans. Signal Process. , vol. 57, no. 2, pp. 726-737, Feb. 2009.”, 「C. Chae, T. Tang, RW Heath, and S. Cho, “relaying with linear processing for multiuser transmission in fixed relay networks,” IEEE Trans. Signal Process. , vol. 56, no. 2, pp. 727-738, Feb. 2008.”, 「 H. Shi, T. Abe, T. Asai, and H. Yoshino, “schemes using matrix triangularization for MIMO wireless networks,” IEEE Trans. Commun. , vol. 55, no. 9, pp. 1683-1688, Sep. 2007.”, 「S. Chen and RS Cheng, “ the degrees of freedom of K -user MIMO interference channel with a MIMO relay,” in Proc. IEEE Global Telecommun. Conf. (GLOBECOM) , Dec. 2010, pp. 1-5.”). However, most of the previous studies assumed a rich distributed path environment where the channel matrix has the entire rank, so these systems cannot be directly applied to a channel environment with insufficient paths.

In addition, in Korean Patent Publication No. 2017-0100317, according to a resource allocation request, when any two adjacent links among the links connecting the child nodes are in a violation relationship by occupying the same time slot, the two adjacent links are A resource allocation method in a wireless sensor network that allocates different time slots is proposed. However, the time slot-based resource allocation method does not provide an efficient transmission method for a rank-deficient channel system.

Republic of Korea Patent Publication No. 10-2151973 (Initial access and wireless resource management for integrated access and backhaul wireless networks) Republic of Korea Patent Publication No. 10-1123222 (Method and device for wireless channel rank estimation) Republic of Korea Patent Publication No. 2017-0100317 Resource allocation method and node device in wireless sensor network

1.B.Wang, J. Zhang, and A. Host-Madsen, “the capacity of MIMO relay channels,”IEEE Trans. Inf. Theory, vol. 51, no. 1, pp. 29-43, Jan. 2005. 2.R. Zhang, C. C. Chai, and Y. Liang, “beamforming and power control for multiantenna relay broadcast channel with QoS constraints,” IEEE Trans. Signal Process., vol. 57, no. 2, pp. 726-737, Feb. 2009. 3. C. Chae, T. Tang, R. W. Heath, and S. Cho, “Relaying with linear processing for multiuser transmission in fixed relay networks,” IEEE Trans. Signal Process., vol. 56, no. 2, pp. 727-738, Feb. 2008. 4. H. Shi, T. Abe, T. Asai, and H. Yoshino, “Schemes using matrix triangularization for MIMO wireless networks,” IEEE Trans. Commun., vol. 55, no. 9, pp. 1683-1688, Sep. 2007. 5. S. Chen and R. S. Cheng, “the degrees of freedom of K-user MIMO interference channel with a MIMO relay,” in Proc. IEEE Global Telecommun. Conf. (GLOBECOM), Dec. 2010, pp. 1-5. 6.H. S. Dhillon, R. K. Ganti, F. Baccelli, and J. G. Andrews, “and analysis of K-tier downlink heterogeneous cellular networks,” IEEE J. Sel. Areas Commun., vol. 30, no. 3, pp. 550-560, Apr. 2012. 7. S. Chia, M. Gasparroni, and P. Brick, “Next challenge for cellular networks: Backhaul,” IEEE Microw. Mag., vol. 10, no. 5, pp. 54-66, Aug. 2009.

The purpose of the present invention is to provide a cooperative relay method for a multi-user wireless backhaul network that efficiently transmits data in a rank-deficient channel environment.

Another object of the present invention is to provide a cooperative relaying method for a multi-user wireless backhaul network that can overcome rank deficiency, eliminate interference between users and repeaters, and provide an alternative signal path by using multiple repeaters as active reflectors in a MIMO transmission system. is to provide.

According to one aspect of the present invention, a cooperative communication method for a wireless backhaul network using a multiple relay broadcasting system including one transmitting end, N repeaters, and K destination receiving ends, wherein the transmitting end has the N number of destinations in a first time slot. Streams to be transmitted are equally distributed and transmitted to repeaters and K target receiving ends, but the repeaters do not transmit the received signal stream in the first time slot, and the transmitting end transmits the N repeaters and K in the second time slot. Streams to be transmitted to each target receiving end are equally distributed and transmitted, and the repeaters extract the stream to be transmitted to each target receiving end from its received signal received in the first time slot and then transmit it to the corresponding target receiving end in the second time slot. It is characterized by including control to do so.

In addition, signals transmitted from the transmitter in the first and second time slots are characterized in that they are transmitted using zero-forcing beam forming.

In addition, the signal transmitted by the repeaters to the target receiving end in the second timeslot is characterized in that it is transmitted by zero-forcing beam forming.

In addition, when M > L(K·N), - where 1 to N are the number of repeaters, 1 to K are the number of destination receiving ends, 1 to L are the number of transmission paths, and 1 to M are the number of antennas, the first In a timeslot, each transmitting end transmits L independent streams that do not reach any repeater to the destination receiving end, and at the same time transmits L independent streams to the destination receiving end through a relay [1: N] other than the dedicated repeater, and the destination receiving end Characterized in that the KL stream is recovered in the first time slot, and in the second time slot, each transmitter transmits the same as the first time slot, and each repeater transmits the signal received in the first time slot as Zero- Forcing beamforming is used to transmit the signal to the target receiving end, and at the same time, each repeater receives the signal transmitted from the transmitting end, and the target receiving end receives and recovers the L(K + N) stream in the second time slot. It is characterized by

According to an embodiment of the present invention, the cooperative relaying method for a multi-user wireless backhaul network system is, in contrast to a rich distributed environment where the sum DoF cannot be improved, the sum DoF is achieved in a poor distributed environment through appropriate cooperative relaying. DoF can be increased dramatically.

In addition, the cooperative relay method according to an embodiment of the present invention is a rank-deficient MIMO multiple access channel (MAC) with a repeater, that is, an industry with a general antenna configuration in which the number of antennas at the repeater and transmitter, repeater, and target receiver are arbitrary. It can be easily expanded and applied to link MU-MIMO rank-fault channels.

In addition, the relaying method proposed according to an embodiment of the present invention transmits directly without cooperative relaying at both the sum DoF in the high signal-to-noise ratio (SNR) region and the sum rate in the finite signal-to-noise ratio (SNR) region. It has improved performance compared to

Additionally, it indicates that the scheme proposed according to an embodiment of the present invention is a good candidate solution for millimeter wave wireless backhaul networks, and can improve system throughput by supporting huge data traffic by improving spectral efficiency.

Additionally, like the mmWave wireless backhaul network, the system method according to an embodiment of the present invention can be applied to maritime communications or satellite-to-terrestrial networks, and may have a very small angular spread effect.

Figure 1 illustrates a multi-user MIMO relay broadcasting system for explaining a cooperative communication method for a multi-user wireless backhaul network according to an embodiment of the present invention.
Figure 2a shows a cooperative communication transmission method in a first timeslot according to an embodiment of the present invention.
Figure 2b illustrates a cooperative communication transmission method in a second timeslot according to an embodiment of the present invention.
Figure 3 graphically shows the sum DoF of examples for the proposed method and the existing method considered for M when L, K, and N are fixed at L = 2, K = 3, and N = 2.
Figure 4 graphically shows the sum DoF for the proposed and existing methods considered for N when M, K, and L are fixed at M = 16, K = 3, and L = 2.
Figure 5 graphically shows the sum DoF for the proposed method and the existing method considered for channel rank L when M, K, and N are fixed to M = 12, K = 3, and N = 2.
Figures 6a and 6b graphically show the achievable sum data rate for the case of perfect self-interference cancellation for the proposed existing method.
Figures 7 and 8 graphically show the summed transmission rate versus transmission power (P) for the proposed method and the existing method in the case of incomplete self-interference cancellation.

The terms used in this application are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, when a part “includes” a certain component, this means that it may further include other components rather than excluding other components unless specifically stated to the contrary.

In addition, terms such as "...unit", "...unit", "module", and "device" used in the specification refer to a unit that processes at least one function or operation, which refers to hardware, software, or a combination of hardware and software. It can be implemented as:

Additionally, when describing components of embodiments of the present invention, terms such as first and second may be used. These terms are only used to distinguish the component from other components, and the nature, sequence, or order of the component is not limited by the term. When a component is described as being 'connected', 'coupled' or 'connected' to another component, that component may be directly connected, coupled or connected to that other component, but that component and that other component It should be understood that another component may be 'connected', 'combined', or 'connected' between elements.

Throughout the specification of the present invention, bold lowercase and uppercase letters are used to indicate vectors and matrices, respectively. For rational number α, operation gives the integer part of α . For matrix A , and represents the transpose and complex conjugate transpose of A , respectively. Also, |A| and rank (A) represent the determinant and rank of A , respectively. and are respectively identity matrix and Represents the zero matrix. message , ,... is used to represent a finite set, [1:n] = {1,2,..,n}.

The expected value is It is displayed as . mean 0 and variance To represent a circularly symmetric complex Gaussian distribution with Use . Additionally, Unif(a, b) represents a continuous uniform distribution over the interval (a, b) .

Hereinafter, a cooperative communication method for a multi-user wireless backhaul network according to the implementation of the present invention will be described in detail.

According to one embodiment of the present invention, cooperative relaying for multi-user multiple-input multiple-output (MU-MIMO) wireless backhaul networks is proposed, wherein a source is transmitted to K destination destinations with the help of N repeaters. Transmits independent messages, where the repeaters are characterized by operating in full-duplex mode or half-duplex mode. It provides a method of data transmission in a line-of-sight dominated environment where each of the channel matrices between nodes is rank-deficient due to the lack of multiple paths. For rank-deficient environments, the multiplexing gain from using multiple antennas is severely limited, and capacity cannot be improved with the number of antennas.

In one embodiment of the present invention, a novel cooperative relaying method using a repeater as an active reflector is provided to overcome the rank deficiency defect.

In a method according to an embodiment of the present invention, the transmitting end distributes the streams through the repeaters as uniformly as possible, and the repeaters extract the streams of each target receiving end based on their received signals and transfer them to their corresponding target receiving ends. It includes a method for forwarding to , while any interference between users and between repeaters is appropriately removed and nullified.

In one embodiment of the present invention, the sum degrees of freedom (DoFs) and sum rates that can be achieved using full-duplex or half-duplex repeaters are analyzed, and the repeaters are full-duplex. When operating in this mode, a strict upper limit of the degrees of freedom (DoF) was derived. From these results, it was demonstrated that the method according to an embodiment of the present invention significantly improves system throughput compared to a system that transmits directly without cooperative relay.

Ⅰ-1 (Example 1: MIMO relay broadcast channel system)

Figure 1 illustrates a multi-user MIMO relay broadcasting system for explaining a cooperative communication method for a multi-user wireless backhaul network according to an embodiment of the present invention.

Referring to FIG. 1, the MIMO (Multiple-Input Multiple-Output) relay broadcasting system according to an embodiment of the present invention includes one transmitter (Source), N repeaters (Relay 1, ... Relay N), and K destination receivers. Includes (Destination). In Figure 1, the solid line represents a direct link between a transmitting end (Source) and a destination receiving end (Destination), and the dotted line represents a cooperative link connected to a relay (Relay).

The relay broadcasting system according to an embodiment of the present invention is characterized in that a system control unit (not shown) controls signal processing of each transmitting end, repeater, and destination receiving end.

The relay according to an embodiment of the present invention is a full duplex method that processes transmission and reception simultaneously, or a half duplex method that processes transmission and reception separately. It is characterized by operation.

Each node of the transmitting end (Source), relay (Relay), and destination receiving end (Destination) uses M antennas. Each communication link consists of L signal paths (L paths).

The MIMO relay broadcasting system according to an embodiment of the present invention takes into account an environment with insufficient multipath, and the signal path is set to be smaller than the number of antennas.

However, the system operates even if the signal path is larger than the number of antennas.

Referring to Figure 1, at time t, set and It is assumed that the relays within each transmit and receive in full duplex (in the case of half duplex) am.).

here, and means a set of relays performing a transmitting operation and a set of repeaters performing a receiving operation, respectively.

and, Relay in timeslot t, denoted by The received signal vector is expressed in Equation 1 below.

here, represents the signal vector transmitted from the transmitting end, represents the transmitted signal vector of the repeater (i), is the channel matrix from the transmitter (source) to the repeater (j), is the channel matrix from repeater (i) to repeater (j), is the noise vector at repeater (i). and, am.

Similarly, The destination receiver at timeslot t, denoted by The received signal vector is given by Equation 2 below.

here, is the channel matrix from the transmitting end to the destination receiving end (k), is the channel matrix from the repeater (i) to the destination receiving end (K), and is the noise vector at the destination receiving end (k). here am. All noise vectors are assumed to be independent of each other, signal vectors, and time slots.

Additionally, all transmitters are subject to their average transmit power constraints, i.e.

must be satisfied, where: is a constant independent of P.

In the following, we focus on rank-deficient channel environments where there are only L ≤ M dominant signal paths between nodes. and are respectively and Assume that it represents the channel matrix corresponding to the kth signal path of (where am.).

Due to the keyhole effect, all and For , regardless of the number of antennas M, rank( F _j,k ( t )) = rank( E _j,i,k ( t )) = rank( H i _,k ( t )) = rank( G _{i,j, k} ( t )) = 1.

As a result, a general rank-deficient MIMO channel is assumed as follows:

here and is an M × 1 vector, and

and is a 1 × M vector, and all coefficients are extracted from a continuous distribution. Therefore, the ranks of F j ( t ), E j,i ( t ), H i ( t ) and G i,j ( t ) are all equal to L with probability 1 (ignoring signal paths other than the L dominant path). Assuming it can be done).

Here, in Equation 3 and are respectively ) represents the receiving and transmitting antenna array response for the kth signal path, and in Equation 4 and , Equation 5 and in Equation 6 and can also be defined similarly. also, Is represents the attenuation of the kth signal path. In Equation 4 , in equation 5 In equation 6 is defined the same way.

In the channel mattress structure described in Equations 3 to 6, a key hole-based channel model was introduced to explain the physical reasons why rank-deficient channels may occur, but the results of one embodiment of the present invention are based on a specific rank-deficient channel model. It is not limited to but can also be applied to other general rank-deficient channel models.

This rank-deficient channel assumption is valid in poor distributed environments such as wireless backhaul. In a rank-deficient channel environment, the angular spread is very small and the number of dominant multipaths is not sufficient compared to the number of conventional antennas.

From this, the multiplexing gains of MIMO are somewhat limited in poor distributed environments.

In the following embodiment, a cooperative relaying method is proposed to overcome channel rank deficiency and provide improved multiplexing gain.

Regarding channel state information (CSI), it is assumed that the entire CSI on the transmitter and receiver sides is available to all nodes, that is, all channel coefficients are perfectly known to all nodes. We primarily consider wireless backhaul networks where channels are likely to be down, and this is due to the fact that all nodes are fixed and there are only a few dominant signal paths between nodes.

Therefore, in the setting according to an embodiment of the present invention, it is possible to obtain total channel state information (CSI) from each node.

Ⅰ-2 ( encoding, decoding achievable speed and degree of freedom)

A source wants to transmit a set of messages {W ₁ , W ₂ ,..., W _K } to each of a set of dedicated destinations (Destinations) {1, 2,., K} . W _k is , It is assumed that it is randomly and uniformly selected from . Here, n is the code length and R _k is the transmission rate of W _k . At timeslot t, the source transmits an encoded signal x(t) that is a function of {W ₁ , W ₂ ,...,W _K }. Then, after n-time transmissions, the destination receiver Attempts to restore W _k based on its received signal vectors {y _k (1), y _k (2),...,y _k (n)}.

The rate tuple {R ₁ , R ₂ ,...,R _K } is such that as the code length n becomes infinite, the average probability of decoding error at all destination receivers becomes 0 (2 ^nR1 ,..., 2 ^nRK , n) It is assumed that this can be achieved if a sequence of codes exists. Capacity area C means closure of the set of achievable rate tuples {R ₁ ,...,R _K }. And the total degree of freedom (DoF) was calculated by applying the same method as in previous studies as follows (「L. Zheng and DNC Tse, “and multiplexing: A fundamental tradeoff in multiple-antenna channels,” IEEE Trans. Inf. Theory , vol. 49, no. 5, pp. 1073-1096, May 2003.」, 「SA Jafar and MJ Fakhereddin, “of freedom for the MIMO interference channel,” IEEE Trans. Inf. Theory, vol. 53, no. 7, pp. 2637-2642, Jul. 2007.”

Here, P is the average transmission power constraint of the source.

Ⅱ (Degree of freedom analysis)

In one embodiment of the present invention, the sum DoF of (M, K, L) MIMO broadcast channels with N repeaters is analyzed. The upper bound on the achievable sum of DoF and sum of networks is mentioned in the following theorem.

(Theorem 1)

For a (M, K, L) MIMO broadcast channel with full-duplex N repeaters, the sum DoF of Equation 8 is achievable.

Here, M, K, L, and N are natural numbers.

Additionally, for a (M, K, L) MIMO broadcast channel with half-duplex N repeaters, the sum DoF of Equation 9 is achievable.

(Theorem 2)

The upper limit of the sum DoF of (M, K, L) MIMO broadcast channels with N repeaters (full or half duplex) is obtained by the following equation 10.

It should be noted that the sum DoF achievable from Theorem 1 and Theorem 2 has an upper bound matching when the repeater operates in full-duplex mode. Therefore, in one embodiment of the present invention, the sum DoF for that case is completely characterized. Theorem 1 can be proven by explaining the details according to an embodiment of the present invention.

A proof of Theorem 1

(1) When M≤L(K+N) and using a full-duplex repeater

In an embodiment of the present invention, first consider the case of using M≤L(K+N) and full-duplex repeater mode.

represents the jth data stream to be delivered to the ith destination receiver (target receiver (i)).

Assume that represents the jth independent data stream for the destination receiving end (i) in timeslot (t). here is independent of i, j and t. In each timeslot (t), the transmitting end sends not only the LN stream through cooperative relaying, which is delivered to the target receiving end in the next timeslot, but also the KL stream through direct transmission to the target receiving end (a timeslot actually means a block of signals, but complexity (block notation is omitted to avoid ).

Specifically, if Equation 11 represents a stream vector for direct transmission exclusively for the destination receiving end i∈, Equation 12 is a stream vector for cooperation dependence in timeslot t, indicating the location where the signal stream will be delivered from the transmitting end, that is, each repeater ( It shows how to properly distribute the stream according to the relay and destination.

here, means the L+1th signal data stream to be delivered to the first destination receiving end, means the L+2th signal data stream to be delivered to the Kth destination receiver, 1~N are the number of repeaters, 1~K are the number of destination receivers, 1~L are the number of transmission paths, and L, N, K are natural Acceptance, , and is a repeater sent to of in the second It means that it is composed of the first element.

Then, in the first timeslot, the transmitter sends the signal matrix of Equation 13.

Equation 13 presents a method of transmitting a signal to eliminate multi-user interference during stream transmission through zero-forcing beamforming at the transmitting end.

here, silver It means the transmission beam forming matrix in the first time slot for delivering to the receiving end i, means the data stream vector to be delivered to the receiving end i, silver It means the transmission beam forming matrix in the first time slot for transmitting to repeater i, means the data stream vector to be delivered to repeater i.

also, Is , and , This is the beam forming matrix for the destination receiving end (i) for direct transmission. rank and is the rank headed to repeater (i) for cooperative relay during , and , is the beam forming matrix of .

Since M≥ L(K + N), V _i (1) and U _i (1) matrices that satisfy the above conditions can be constructed.

Additionally, V _i (1) and U _i (1) can be properly normalized to meet the average power limit P.

The main purpose of the cooperative relaying method proposed here is to distribute each target's stream as evenly as possible to each repeater in order to maximize the effective channel rank by efficiently using multiple repeaters as active reflectors, as implied in Equation 12.

In the present invention, a time slot refers to a time interval periodically allocated to each channel when sending data using time division multiplexing, where the first time slot refers to the first time slot and the second time slot refers to the second time slot. do.

First, in the first timeslot, the transmitter transmits the streams to be transmitted in a way that distributes them to each repeater as equally as possible.

On the other hand, all repeaters while all The first timeslot for, that is, and does not transmit a signal to The repeater does not transmit a signal in the first timeslot.

destination receiving end The signal vector received in the first time slot is given by Equation 14.

here means the channel matrix between the receiving end k and the transmitting end, silver refers to the transmission beam forming matrix for delivering to the receiving end k, means the channel matrix between repeater i and receiving end k, means the noise vector at the receiving end k.

Here, so that streams associated with beamforming vectors other than V _k (1) do not reach the destination receiving end (k). Is , and (b) is due to the zero forcing properties of Vi(1) and U _i (1). As a result, rank = Therefore, the destination receiving end (k) is Based on , The stream can be recovered. here refers to a set of repeaters that perform transmission operations.

Additionally, the repeater in the first timeslot The signal vector received from is given by Equation 15.

Equations 15 and 16 present a method for each relay to separate and extract a stream to be transmitted to each destination from its received signal received in the first timeslot.

here means the channel matrix between the transmitter and repeater j, means the channel matrix between repeater i and repeater j, means the noise vector at repeater j.

In equation 15: is due to the zero forcing properties of V _i (1) and U _i (1), as shown in Equation 14. Therefore, in Equation 15 related to U _j (1) The streams can reach the repeater (j).

to the next, Based on, the repeater (j) according to the destination receiving end indexes We attempt to classify the streams in s _r,j (1). silver At the destination receiving end represents the stream for

after that, at To extract the repeater (j) is a zero-forcing receiver beamforming matrix Apply to y _r,j (1). here is given as the following equation: is the cardinality of .

here ego, silver not associated with Invalidates all column vectors in .

Therefore, equation 17 can be obtained.

here, and = am.

In this way, the repeater (j) For s _r,j,i (1), we can obtain a noisy version.

In one embodiment of the present invention, a cooperative relay strategy is considered in the second time slot. This is similar to the method of the first timeslot, where the signal of Equation 18 is transmitted from the transmitting end.

The second timeslot is designed in such a way that V _i (2) and U _i (2) are the same as the first timeslot.

Equations 19 and 20 present a method in which each relay delivers a stream to the corresponding destination destination without interference from noise between users and relays in the second timeslot according to an embodiment of the present invention.

repeater silver Zero forcing beam forming is performed as shown in Equation 19 below.

here is the beam forming matrix of the repeater (j) that delivers the stream to the destination receiving end (i) that satisfies the conditions of Equation 20 below.

here means the channel matrix between repeater j and receiving end k in the second timeslot, means the channel matrix between repeater j and repeater k in the second time slot, is a matrix It means the matrix coefficient (rank) of .

Because of Appropriate to satisfy all can be found.

Specifically, the beamforming matrix Is It is designed to nullify both user-to-user interference and relay-to-relay interference while delivering only to the intended receiving end (i). Additionally, the average power constraint to satisfy can be properly normalized. also You can see that it is.

Then the destination receiving end The signal vector received in the second timeslot is given by Equation 21 below.

here Is This is due to the zero forcing properties of .

and Is and The results given by each indicate the valid channels.

and,

am.

also,

also,

thus The rank of the matrix Equation 23 indicated by is given by the following Equation 24.

Equation 24 is It is caused by.

Therefore, the K purpose receiving end is In this case, (α + L + 1) desired streams can be recovered, otherwise, (α + L) desired streams can be recovered in the second timeslot.

As a result, considering all K destination receiving ends, a total of αK + β + KL = L(K + N) streams can be transmitted during the second timeslot.

on the other side When set to , the signal at repeater j∈[1:N] of the received second timeslot is given by Equation 25.

here Is Due to its zero-forcing characteristics, interference between repeaters is completely eliminated.

As a result, only the stream related to U _j (2) can reach the repeater (j), similar to Equation 15.

Moreover, by performing the same steps as in the first timeslot, The stream can be extracted and classified according to the destination index classification before being transmitted in the next timeslot.

In summary, the total (2KL + NL) stream can be recovered over two timeslots according to the method according to an embodiment of the present invention.

Moreover, if we apply this transmission strategy recursively to T timeslots, a total of (TKL + (T- 1)NL) streams can be transmitted in a similar manner.

As a result, if T →∞, the sum DoF of Equation 26 can be obtained.

(2) When M≥ L(K + N) and using a half-duplex repeater,

The methods that can be achieved with a half-duplex repeater can be achieved by slightly modifying the full-duplex repeater, so to avoid repetition, the main changes caused by the half-duplex repeater will be described below.

In half duplex mode, the repeater can transmit or receive in each timeslot.

Therefore, unlike the full duplex method, all repeaters are set to operate in transmission mode in the second time slot. in other words, , which naturally causes Additionally, unlike full-duplex schemes that operate over sufficiently large T timeslots, we limit T to T = 2 for achievable schemes.

The method that can be achieved with a full duplex repeater can be applied to a network with a half duplex repeater, and as a result, the following Equation 27 can be achieved.

Therefore, the cooperative relay method according to an embodiment of the present invention can provide a cooperative transmission method for a backhaul network based on a half-duplex repeater in a rank-deficient channel environment with insufficient multipath.

(3) If M<L(K + N) and using a full duplex or half duplex repeater,

First, consider a network with full duplex repeaters. In this case, in the network, only K ₁ of the K destination receivers is activated and only N ₁ of the N repeaters is activated so that the conditions K ₁ ≤ K and N ₁ ≤ N, M ≥ L(K ₁ + N ₁ ), are satisfied.

The network then becomes a (M,K ₁ , L) broadcast channel with N1 full duplex repeaters, and the proposed achievable scheme can be directly applied using full duplex repeaters as explained earlier.

also, under conditions Maximizing the sum DoF of Optimal, denoted by Choosing a pair is easy.

added last You can also get DoF.

here and If , one additional destination receiver is activated and transmitted directly, otherwise, one additional relay is activated and transmitted through cooperative relay transmission (if K ₁ = K, M < L (K + N) is assumed and thus activated Because there is at least one additional repeater left to do. must be less than N.).

As a result, the sum DoF of M is achievable.

Now let's look at a network with half-duplex repeaters.

First, consider the case where KL < M < L(K + N). In this case, only N ₂ repeaters among N repeaters are activated. here am.

The network then becomes a (M,K,L) broadcast channel with N ₂ half-duplex repeaters, and the proposed achievable approach can again be applied using direct half-duplex repeaters as described above. Additionally, one additional repeater can be activated to send an additional M - L(K + N ₂ ) stream over two timeslots through cooperative relay transmission.

As a result, the sum DoF of equation 28 is achievable.

finally, In this case, the sum DoF of M can be achieved only through direct transmission without cooperative relaying. This proves Theorem 1.

Now we prove Theorem 2.

B. Proof of Theorem 2

Consider a repeater with full cooperation with all destination receivers. Then the network determines the channel rank It becomes a single-user MIMO channel with M transmit antennas and M(K + N) receive antennas. Additionally, we assume that there is a noise-free feedback link from the receiver to the transmitter with infinite link capacity in a single user MIMO channel. This new channel is obviously larger than the original channel, as allowing full collaboration or adding feedback links does not reduce capacity. It is well known from previous research that feedback cannot increase the capacity of a point-to-point channel without memory, and that the sum DoF of a MIMO channel is determined by the channel rank.

Therefore, the sum DoF of the new network is It can be specified as . As a result, the sum DoF of the original network is The upper bound is specified as , and the proof of Theorem 2 is proven.

Next, the present invention will be described in detail through examples.

Figure 2a shows a cooperative communication transmission method in a first timeslot according to an embodiment of the present invention.

Figure 2b illustrates a cooperative communication transmission method in a second timeslot according to an embodiment of the present invention.

Referring to Figures 2a and 2b, the number of transmitting antennas M = 5, the number of target receiving ends K = 2, the number of repeaters N = 3, the signal path L = 1, and the repeater operates in full duplex mode in the first and second timeslots. The case was considered. represents the jth data stream to be delivered to the ith destination receiver, Is s ₁ , ₂ represents the noise version of (1).

Referring to Figure 2a, in the first timeslot, according to the achievable method described in the previous [Proof of Theorem 1], the transmitting end (souece) sends s ₁ , ₁ (1) and s ₂ , ₁ (1) to the destination receiving end, respectively. Deliver to Destination 1 and Destination 2 and distribute equally. The transmitting end (souece) transmits the signal data streams s ₁ , ₂ (1), s ₂ , ₂ (1), s ₁ , ₃ (1) through zero-forcing beamforming, respectively, to relays Relay 1, Relay 2, Relay 3. send to This zero forcing transmission method can be implemented as M≥L(N+K). Then, each destination receiving terminal Can recover the desired stream s _i , ₁ (1) in the first timeslot. Additionally, the relay stores the received signal.

That is, in the first timeslot, the transmitting end distributes streams one by one to each repeater and the destination receiving end. At this time, zero forcing beam forming is used to accurately deliver each stream to the desired location without causing interference elsewhere. Each destination receiving end can decode and restore streams one by one.

Referring to Figure 2b, in the second timeslot, the transmitting end transmits signal data streams s _1,1 (2), s _2,1 (2), s ₁ , ₂ (2), s ₂ , ₂ (2), and s _1. , ₃ (2) is transmitted similarly to the transmission method of the first timeslot to the destination receiving end Destination 1, the destination receiving end Destination 2, and the repeater Relay 1, Relay 2, and Relay 3, respectively. On the other hand, relays Relay 1 and Relay 3 send noise versions of signal data streams s ₁ , ₂ (1) and s ₁ , ₃ (1) to destination 1, respectively, and relay 2 sends s ₂ , ₂ (1). The noise version is sent to destination 2. The repeater uses zero forcing beamforming and does not cause interference between users or between repeaters. Additionally, the repeater receives a stream signal from the transmitter transmitted in the second timeslot.

As a result, destination destination Destination 1 can recover the {s _1,2 (1), s _1,3 (1), s _1,1 (2)} streams. The intended receiving end, Destination 2, can recover the {s _2,2 (1), s _2,1 (2)} stream in the second timeslot.

That is, in the second time slot, as in the first time slot, the transmitting end distributes and transmits one stream to each repeater and the destination receiving end. The repeater uses zero forcing beamforming to accurately deliver the stream received in the first timeslot only to the target receiving end without causing interference elsewhere. As a result, target receiver 1 can recover three streams, and target receiver 2 can recover two streams.

In summary, a total of 7 streams can be transmitted over 2 timeslots. When used as a half-duplex repeater, the achieved multiplexing gain (sum DoF) is 7/2=3.5. And if you use a full duplex repeater that can transmit and receive at the same time, the multiplexing gain (sum DoF) increases to 5 (when using a duplex repeater, it can always operate in the second timeslot except for initial transmission) do.)

Considering the fact that the sum of DoF is only 2 when transmitting directly without a repeater, the cooperative communication repeater method according to an embodiment of the present invention is simple, but can significantly increase transmission capacity in a poor distributed environment.

Additionally, by comparing the achievable DoF with full duplex and half duplex repeaters, operating in full duplex mode can further improve the total DoF.

C. Theorem 1: Applying MIMO repeater to multiple access channels

The cooperative communication method according to an embodiment of the present invention can be easily extended to a rank-deficient MAC (Multiple Access Channel) such as a rank-deficient channel in an uplink MUM IMO system with a repeater in a similar manner.

A (M,K,L) multiple access channel with N repeaters represents a network in which K transmitting ends attempt to send independent messages to the destination receiving end with the help of N repeaters, where each of the destination receiving ends, transmitting ends and repeaters has M Antenna is installed. Here, each channel rank is L ≤ M. Then, the achievable sum DoF and the upper bound on the sum DoF of (M,K,L)multiple access channels with N repeaters are the same as in Theorems 1 and 2, respectively.

Considering the case where M ≥ L(K + N) and the repeater operates in full duplex, each transmitter in the first timeslot Sends L independent streams to the destination. It never reaches a repeater, and simultaneously sends L independent streams to each repeater. send to does not reach the destination like repeaters other than dedicated repeaters through zero-forcing beam-forming similar to that of broadcast channels.

Then, the destination receiving end can recover the KL stream in the first timeslot. This transmission strategy is feasible because M≥L(K + N), similar to the reasons explained previously.

In the first timeslot, each transmitting end transmits L independent streams that do not reach any repeater to the destination receiving end, and at the same time, repeaters other than the dedicated repeater [1: N] transmit L independent streams [ 1cing: K] is transmitted to the destination receiving end. Afterwards, the destination receiving end can recover the KL stream in the first timeslot. This transmission strategy is feasible because M > L(K·N), similar to the reasons given in Section III-A.

The transmission method for each transmitter in the second timeslot is the same as the first timeslot. At the same time, each repeater transmits the previously received signal in the first timeslot to the destination receiver and prevents it from reaching other repeaters using Zero-Forcing beamforming. Additionally, at the same time, each repeater receives the signal transmitted from the transmitter. As a result, the destination receiving end can receive and recover the L(K + N) stream in the second timeslot.

In summary, a total of (2KL + LN) streams can be transmitted over two timeslots.

Additionally, by recursively applying this transmission method to T timeslots where T is sufficiently large, the sum DoF of L(K + N) can be obtained.

The achievable sum DoF for other cases can be similarly proven.

Conversely, these systems now allow full cooperation between repeaters and transmitters, and the network has M(K + N) transmit antennas and M receive antennas with channel rank min{M,L(K + NS)}. It becomes a single user MIMO channel.

Therefore, by applying the cooperative communication method according to an embodiment of the present invention, it can be easily expanded to a rank-deficient MAC (Multiple Access Channel) such as a rank-deficient channel in an uplink MUMIMO system.

Theorem 2 : Apply to general antenna configuration

The cooperative relay communication method according to an embodiment of the present invention can be extended to cases where the number of antennas at the transmitting end, repeater, and destination is arbitrary.

Ms, Mr, and Md represent the number of antennas at the transmitting end, repeater, and destination, respectively. Then, by applying the relay communication method according to an embodiment of the present invention, the total DoF can be expressed by the following equation in each achievable full duplex and half duplex.

here,

, Equation 29 represents the case of a full duplex repeater, and Equation 30 represents the case of a half duplex repeater. Here, K* and N* mean the optimal values of K and N, respectively.

(proof)

The description below focuses on the modifications introduced due to generalization of antenna configuration to avoid duplication of explanation.

First, the system control unit selects K* and N* that satisfy the above condition Equation 31. At this time, two time slots are considered. Then, in a manner similar to that previously described in Equation 11, the transmitting end Send the stream to the destination receiving end through direct transmission, The stream is Afterwards, the sending step through cooperative relay is performed in each timeslot. Likewise, each repeater appropriately extracts the stream of each destination receiver based on the received signal and Afterwards, the step of transmitting to the target receiving end without causing interference between users and repeaters is performed.

Additionally, the destination receiving end silver If the desired stream is Otherwise, you can recover the desired stream in the second timeslot. It can be restored with here , , and due to the third condition of Equation 31, am.

As a result, total The stream may be transmitted during the second timeslot.

Considering these facts, it can be seen that for a typical antenna configuration, the summed DoF specified in Equation 29 and Equation 30 can be achieved with a full duplex and half duplex repeater, respectively.

By extending Theorem 2 to a general antenna configuration, it can be seen that the sum DoF for a general antenna configuration reaches an upper limit as shown in Equation 32.

Therefore, the achievable sum DoF in Equation 29 is the number of antennas at each node. If it is large enough, it matches the upper limit of Equation 32.

Next, different application examples are explained.

(Application example 1, direct transmission)

In the case of only direct transmission without cooperative relaying (i.e., when N = 0), the sum of the DoF from Theorems 1 and 2 is This happens.

This result shows that the necessary condition for obtaining DoF sum improvement in the cooperative communication method according to an embodiment of the present invention is It means that

(Application example 2. Full-rank case)

For a full-rank channel, i.e., L = M, the sum DoF becomes M from Theorems 1 and 2, which is the same as the sum DoF that can be obtained through direct transmission without cooperative relaying. This result shows that cooperative relaying cannot improve the sum DoF for full-rank channels, as discussed in previous studies. However, as can be seen in Theorem 1, cooperative relaying can increase the sum DoF for poor sparsely distributed environments under certain conditions.

(Application example 3: Time division)

Consider time-division transmission in which only one destination is active in each timeslot. In this case, the sum DoF achievable with full duplex and half duplex repeaters are respectively and It is expressed as

Time division using half-duplex repeaters can be recovered in the manner proposed in previous studies.

(Application example 4: When there is no forward channel state information (CSI) in the repeater)

If forward CSI is not available at the repeaters, that is, each repeater (i) and Consider the case where the coefficient of is not known. In this case, since the repeater does not have CSI, it cannot perform interference cancellation between users or between repeaters. Instead, in this case, to obtain better performance, the direct transmission in Application Example 1 or the time division transmission method in Application Example 3 can be adapted and selectively applied.

Therefore, the sum DoG achievable in full duplex and bottom duplex repeaters is respectively and It can be expressed as

Therefore, it can be seen that benefits can be obtained by using repeaters when the number N of repeaters is sufficiently large compared to the number of target receiving ends K.

Ⅲ. SUM RATE analysis

According to an embodiment of the present invention, in addition to the sum DoF, the sum rate of the proposed method can also be derived in the finite SNR region. First, consider the case of a full duplex repeater. For simplicity, we assume M/L is an integer. Additionally, according to Theorem 1, we select the optimal K and N that maximize the sum DoF in full duplex mode and assume that the communication occurs in a sufficiently large timeslot T. Then, the average achievable aggregate transmission rate in Equation 21 is given by Equation 33.

here is the destination receiving end It is a ratio for Is is the covariance of . And expectations are made over time slots.

We now consider the aggregate rate achievable when the repeater operates in half-duplex mode. Again, we assume that we choose the optimal K and N that maximize the sum DoF of Theorem 1 in half-duplex mode. Unlike the transmission method of full duplex mode, in the case of half-duplex mode, communication occurs over two timeslots. Therefore, the average achievable combined transmission rate in Equation 14 and Equation 21 can be calculated by the following Equation 34.

Meanwhile, the average combined transmission rate of direct transmission without a repeater described in Application Example 1 is expressed in Equation 35 below.

Next, we look at the average combined transmission rate of time division transmission using the full duplex repeater described in Application Example 3. In this case, it is assumed that communication takes place through KT timeslots and that only one destination receiver is activated for each T timeslot in a round robin manner.

and is the destination receiving end If only the destination receiver in Equation 21 is activated, represents the effective channel matrix for represents the covariance of the corresponding noise vector. The following equation 36 represents the average combined transmission rate.

Lastly, when there is no forward CSI in the repeater described in Application Example 4, the average sum rate is given by Equation 37 below.

Ⅳ. Numerical results

Next, the sum DoF and sum rate of the technique proposed in the present invention are numerically evaluated and the results are explained. Specifically, the DoF and combined transmission rate of the proposed method and the full duplex repeater, the proposed method and the half duplex repeater, the direct transmission method without cooperative relay mentioned in Application Example 1, and the time division using the full duplex repeater method described in Application Example 3. Compare transmission methods. We can see from the previous discussion that the upper bound on the sum DoF in Theorem 2 is consistent with the DoF achievable with full duplex specified in Theorem 1. In addition, the DoF and sum rate that can be achieved in the repeater when there is no forward CSI in each case follow the maximum values that can be obtained in direct transmission and time division transmission, as explained in Application Example 4 and Equation 37.

A. Sum DoF Comparison

(1) Comparison of sum DoF for M:

First, the sum DoF trend for the number of antennas M is analyzed.

Figure 3 graphically shows the sum DoF of examples for the proposed method and the existing method considered for M when L, K, and N are fixed at L = 2, K = 3, and N = 2.

Referring to Figure 3, the results show that the proposed cooperative communication scheme significantly improves the sum DoF in both full-duplex repeaters and half-duplex repeaters compared to direct transmission or time-division transmission when the number of antennas M is greater than a certain threshold. Specifically, unlike direct transmission where the sum DoF is proportional to M only until M is less than or equal to 6, in the proposed scheme the sum DoF increases as M ≤ 10 until M.

Therefore, when M is greater than 6, the method proposed in one embodiment of the present invention outperforms direct transmission without using a repeater, showing that the proposed relay method successfully overcomes the ranking defect. In addition, in the cooperative communication method according to an embodiment of the present invention, by allowing a full duplex operation mode by comparing the DoF sum with full duplex and half duplex repeaters, the sum DoF can be further improved in a MIMO channel with insufficient rank. was analyzed.

In addition, when comparing the DoF sum of the cooperative communication method proposed according to an embodiment of the present invention and the full duplex repeater with the time division transmission method, it is found that multi-user cooperative transmission along with interference management between users according to an embodiment of the present invention It shows that the sum DoF can be significantly improved compared to single-user communication.

(2) Comparison of sum DoF for N:

Next, the sum DoF for the number N of repeaters is analyzed and explained.

Figure 4 graphically shows the sum DoF for the proposed and existing methods considered for N when M, K, and L are fixed at M = 16, K = 3, and L = 2.

Referring to Figure 4, clearly the sum DoF of direct transmission does not change with N. As expected, the results show that the sum DoF of all methods increases with N up to a certain threshold except for direct transmission. The signal path provided by the repeater increases by N and consequently the rank of the effective desired channel matrix increases by N up to the threshold. In other words, performance increases depending on the number N of repeaters and then converges above a certain threshold. In particular, the proposed full duplex repeater method provides the same sum DoF as in the case of a full rank channel if the number of repeaters is sufficiently large, i.e. It shows that it can be achieved. Even in the case of time division, which is single-user communication, an improved sum DoF can be obtained compared to direct transmission (multi-user communication) when N>4 thanks to the cooperative relay proposed in an embodiment of the present invention.

(3) Comparison of sum DoF for L:

Figure 5 graphically shows the sum DoF for the proposed and existing methods considered for channel link L when M, K, and N are fixed at M = 12, K = 3, and N = 2.

Referring to Figure 5, it is shown that improved sum DoF can be provided when the channel rank L is less than a certain threshold (e.g., L < 4 in this case).

That is, when the value L of the number of multiple paths is small, the performance difference of the cooperative communication method according to an embodiment of the present invention increases compared to the existing communication method. Therefore, the cooperative communication method according to an embodiment of the present invention has multiple paths. It can be seen that it is especially effective when it is small.

In addition, if the channel rank L is sufficiently large, the sum DoF of M, which is the upper limit of the sum DoF, can be achieved even when transmitting directly without cooperative relaying, so in this case, the proposed relaying method is not necessary to improve the sum DoF.

B. Sum Rate Comparison

1) Complete magnetic interference cancellation e.g.

Consider the case M = 10, K = 3, L = 2, N = 2. The example described this time adopts the LoS channel model described previously, where a half-wavelength uniform linear antenna array is used at each node.

Specifically, in Equation 3 and Specifically, each and It is given by Equation 38, and is the receiving antenna array of repeater j, respectively, in the transmitting antenna array of the transmitting end, and Indicates the angle of incidence of the jth LoS path in Equation 3 described above.

Additionally, in Equation 3 described above, , and all and for t Assume:

Other channel matrices equations 4, 5 and 6 are defined similarly. In addition, as described in Embodiment 1 (MIMO relay broadcast channel system) described above, all and for t Assume perfect self-interference cancellation at each repeater.

Figures 6a and 6b graphically show the achievable sum data rate for the case of perfect self-interference cancellation for the proposed existing method.

Referring to Figures 6a and 6b, the power for the perfect self-interference cancellation case when M = 10, K = 3, L = 2, and N = 2 in an example using time-division transmission in conventional direct transmission and full duplex mode. It represents the achievable sum data rate using gain P.

Figure 6 shows the proposed scheme using full duplex repeater, half duplex repeater, direct transmission and time division transmission given by equations (33), (34), (35) and (36), respectively.

Figure 6a This shows the case, and Figure 6b shows This shows the case.

Referring to Figures 6a and 6b, the results show that the proposed technique outperforms both direct transmission and time-sliced transmission over the entire range of SNR, and the performance gap increases with P. Of these, repeaters are given as 10 and 8, respectively, but direct transmission and time division are only given as 6. also When is small, that is, each transmission power is the repeater in Figure 6(b). = 0.2, which is smaller than the transmission stage, but the proposed scheme can still provide improved performance when P is greater than a certain threshold (e.g., P > 5 dB).

That is, when the transmission power of the repeater is less than that of the transmitter ( When , is small, the performance gain is somewhat reduced, but it is analyzed that it still maintains better performance than the existing method.

2) Example of incomplete magnetic interference cancellation:

Although it is assumed that the self-interference of each full duplex repeater is completely eliminated, in reality, there may be residual self-interference that cannot be ignored due to incomplete self-interference cancellation. Here we investigate the effect of residual self-interference on the summed transmission rate.

Specifically, the coefficients of the self-interference channel matrix of repeater j, i.e. The coefficients of all About It is assumed that it is derived from . here compared to other channels represents the effective channel strength.

is determined by the amount of relative path loss, which is significantly smaller than that of other channels because the distance between the transmitter and receiver within the repeater is very short, and the amount of self-interference suppression in a full duplex repeater.

Residual magnetic interference in Equation 17 Because when the repeater operates in full duplex mode,

It becomes,

Figures 7 and 8 graphically show the summed transmission rate versus transmission power (P) for the proposed method and the existing method in the case of incomplete self-interference cancellation.

Figures 7 and 8 respectively and Indicates when.

Here other parameters are the same as in Figure 6.

Comparing this figure with Figure 6, it can be seen that the aggregate data rate performance of the full duplex repeater is significantly degraded.

Since our proposed relaying method is based on an amplification-then-forwarding strategy, additional noise amplification due to residual magnetic interference may occur.

Referring to Figures 7a and 7b, in the proposed full duplex repeater, self-interference cancellation is not perfect, resulting in a situation where the transmitted signal affects the received signal ( is not 0), it can still be seen that better performance is maintained than the conventional transmission method.

Certain thresholds, e.g. = -20dB), the half duplex repeater method proposed according to an embodiment of the present invention appears to have better performance than the full duplex repeater method, as shown in FIG. 7(a) or FIG. 8(a). and As , increases, the benefits of cooperative full duplex repeaters are expected to almost disappear. In this case, it is analyzed that the cooperative relay method according to an embodiment of the present invention must operate in half-duplex repeater mode rather than full-duplex repeater mode.

According to an embodiment of the present invention, a cooperative relay technique for MU-MIMO wireless backhaul network is provided. The cooperative communication method according to an embodiment of the present invention uses multiple repeaters as active reflectors to overcome rank deficiency and provide an alternative signal path that can eliminate interference between users and between repeaters.

The cooperative communication method according to an embodiment of the present invention can significantly improve both the sum DoF and the sum data rate compared to direct transmission only without cooperative communication.

The cooperative communication method according to an embodiment of the present invention can significantly improve performance while being based on linear beam-forming with low complexity.

The cooperative communication method according to an embodiment of the present invention can be an attractive practical solution for improving the capacity of rank-deficient channels, especially to support high-performance networks such as mmWave wireless backhaul.

The above description and abstract of various embodiments of the disclosure and corresponding drawings of one embodiment of the invention are intended to be illustrative and not intended to limit or exhaust the disclosed embodiments to the precise form disclosed. That's not what I'm trying to do. Those skilled in the art may realize that other embodiments, including modifications, substitutions, combinations, and additions, may be implemented to perform the same, similar, alternative, or alternative functions of the disclosed subject matter. Accordingly, it should be understood that such possible extensions may be recognized as being considered within the scope of the present disclosure.

The subject matter disclosed in one embodiment of the invention is not limited to any single embodiment described herein.

Source: Transmitter
Relay 1, N: Repeater
Destination1,...K: Destination receiving end

Claims (15)

delete delete In a cooperative communication method for a wireless backhaul network using a multi-user relay system including one transmitter, N repeaters, and K destination receivers,
The cooperative communication method is,
The transmitting end equally distributes and transmits streams to be transmitted to the N repeaters and K destination receiving ends in the first time slot, but the repeaters do not transmit the transmitted signal stream in the first time slot,
The transmitting end equally distributes and transmits the streams to be transmitted to the N repeaters and K target receiving ends in the second time slot, and the repeaters transmit the streams to be transmitted to each target receiving end from their received signals in the first time slot. After extracting, it is characterized in that it includes controlling the transmission to the target receiving end in the second time slot,
When the repeater is full-duplex in a rank-deficient channel under the condition of M≤L(K+N), the signal stream (Sr(t)) transmitted by the transmitting end to each repeater and the destination receiving end in the time slot (t) is A cooperative communication method characterized in that it is formed by the following Equation 12.
(Equation 12)

here, means the L+1th data stream to be delivered to the first destination receiving end, means the L+2th data stream to be delivered to the Kth destination receiver, 1~N are the number of repeaters, 1~K are the number of destination receivers, 1~L are the number of transmission paths, 1~M are the number of antennas, and L, N and K are natural numbers. Also, and is a repeater sent to of in the second It means that it is composed of the first element. According to paragraph 3,
A cooperative communication method, wherein the signal matrix transmitted from the transmitting end to each repeater and the destination receiving end in the first time slot is according to Equation 13 below.
(Equation 13)

here, silver It means the transmission beamforming matrix in the first time slot for delivering to the receiving end i, means the data stream vector to be delivered to the receiving end i, silver It means the transmission beamforming matrix in the first time slot for transmitting to repeater i, means the data stream vector to be delivered to repeater i. According to paragraph 4,
destination receiving end A cooperative communication method, characterized in that the signal (y _k (1)) received in the first timeslot is given by the following equation 14.
(Equation 14)

here means the channel matrix between the receiving end k and the transmitting end, silver refers to the transmission beamforming matrix for delivering to the receiving end k, means the channel matrix between repeater i and receiving end k, means the noise vector at the receiving end k. According to clause 5,
In Equation 14 above, so that streams associated with beamforming vectors other than V _k (1) do not reach the destination receiving end (k). Is and (b) is characterized by being due to the zero forcing properties of Vi (1) and U _i (1) - where refers to a set of repeaters that perform transmission operations,
The destination receiving end (k) is Based on rank = A cooperative communication method characterized by recovering the stream of s _d , _k (1), ∀k ∈ [1: K ] under the conditions. According to clause 6,
repeater Signal vector received in the first time slot is a cooperative communication method characterized by being obtained by the following equation 15.
(Equation 15)

here means the channel matrix between the transmitter and repeater j, means the channel matrix between repeater i and repeater j, means the noise vector at repeater j. In clause 7,
The repeater i receives the signal stream from the transmitter In s _r,j (1), the destination receiving end Characterized by invalidating all column vectors of F j (1)U _j (1) that are not associated with s _r,j,i (1) to extract the signal stream s _r, j, _i (1) to be transmitted to A collaborative communication method. In clause 7,
In Equation 15, the streams of s _r,j (1) arrive at the repeater (j), and in Equation 15 is due to the zero forcing properties of V _i (1) and U _i (1), and the repeater (j) is based on the y _r,j (1) and according to the target receiver indices. Classify the streams in s _r,j (1) and the destination receiver Extract the signal streams s _r , _j , i (1) for, and the repeater (j) is a zero-forcing receiver to extract s _r,j,i (1) from the signal stream s r, _{j (} 1). Beamforming Matrix A cooperative communication method characterized by applying to y _r,j (1) above.
here is the cardinality of s _r,j,i (1) given in Equation 16 below.
(Equation 16)
here ego, is to invalidate all column vectors of F _j (1)U _j (1) that are not associated with s _r,j,i (1). According to clause 9,
Signal vector received in the first time slot from the repeater (j) Signal vector to be transmitted from to the destination receiving end (i) is a cooperative communication method characterized by being obtained by the following equation 17.
(Equation 17)

here, and = lim. According to clause 10,
A cooperative communication method, wherein the signal matrix transmitted by the transmitting end to each repeater and the destination receiving end in the second time slot is given by the following equation 18.
(Equation 18)
According to clause 11,
the repeater above A cooperative communication method characterized by transmitting to the destination receiving end (i) without interference from noise by performing zero forcing beam forming as shown in Equation 19 below.
(Equation 19)

here is the beam forming matrix of the repeater (j) delivering the stream to the destination receiving end (i) that satisfies the conditions of Equation 20 below.
(Equation 20)

here means the channel matrix between repeater j and receiving end k in the second time slot, means the channel matrix between repeater j and repeater k in the second time slot, is a matrix It means the matrix coefficient (rank) of . In a cooperative communication method for a wireless backhaul network using a multi-user relay system including one transmitter, N repeaters, and K destination receivers,
The cooperative communication method is,
The transmitting end equally distributes and transmits streams to be transmitted to the N repeaters and K destination receiving ends in the first time slot, but the repeaters do not transmit the transmitted signal stream in the first time slot,
The transmitting end equally distributes and transmits the streams to be transmitted to the N repeaters and K target receiving ends in the second time slot, and the repeaters transmit the streams to be transmitted to each target receiving end from their received signals in the first time slot. After extracting, it is characterized in that it includes controlling the transmission to the target receiving end in the second time slot,
Under the condition of M≥L(K + N), when the repeater is in half-duplex mode, the repeater is controlled to transmit a signal to the destination receiving end in the second time slot, and the total degree of freedom (sum DoF) is expressed by the following equation A cooperative communication method obtained by 27.
(Equation 27)

Here, 1~N are the number of repeaters, 1~K are the number of target receiving ends, 1~L are the number of transmission paths, and 1~M are the number of antennas. In a cooperative communication method for a wireless backhaul network using a multi-user relay system including one transmitter, N repeaters, and K destination receivers,
The cooperative communication method is,
The transmitting end equally distributes and transmits streams to be transmitted to the N repeaters and K destination receiving ends in the first time slot, but the repeaters do not transmit the transmitted signal stream in the first time slot,
The transmitting end equally distributes and transmits the streams to be transmitted to the N repeaters and K target receiving ends in the second time slot, and the repeaters transmit the streams to be transmitted to each target receiving end from their received signals in the first time slot. After extracting, it is characterized in that it includes controlling the transmission to the target receiving end in the second time slot,
If M > L(K·N), - where 1~N is the number of repeaters, 1~K is the number of target receiving stations, 1~L is the number of transmission paths, 1~M is the number of antennas,
In the first timeslot, each transmitting end transmits L independent streams that do not reach any repeater to the destination receiving end, and at the same time transmits L independent streams to the destination receiving end through a relay [1: N] other than the dedicated repeater, and the destination receiving end Characterized by recovering the KL stream in the first timeslot,
In the second time slot, each transmitting end transmits in the same way as the first time slot, and each repeater transmits the signal received in the first time slot to the destination receiving end using zero-forcing beamforming, and at the same time, each repeater transmits the signal received in the first time slot to the destination receiving end. A cooperative communication method characterized in that the repeater receives the signal transmitted from the transmitting end, and the destination receiving end receives and recovers the L(K + N) stream in the second time slot. In a cooperative communication method for a wireless backhaul network using a multiple relay broadcasting system including one transmitter, N repeaters, K destination receivers, and L signal paths,
The cooperative communication method is,
a) the control unit of the system selects K* and N* that satisfy the following equation;

Here, Ms, Mr, and Md represent the number of antennas at the transmitting end, repeater, and destination, respectively, and K* and N* mean the optimal values of K and N, respectively.
b) The control unit of the system is located at the transmitting end. Send the stream to the destination receiving end through direct transmission, The stream is Afterwards, the step of sending through cooperative relay in each timeslot;
- In the cooperative relay, the transmitting end equally distributes and transmits streams to be transmitted to the N repeaters and K target receiving ends in the first time slot, but the repeaters do not transmit the received signal stream in the first time slot. and
The transmitting end equally distributes and transmits the streams to be transmitted to the N repeaters and K target receiving ends in the second time slot, and the repeaters transmit the streams to be transmitted to each target receiving end from their received signals in the first time slot. After extracting, it is characterized in that it includes controlling the transmission to the target receiving end in the second time slot,
c) Each repeater extracts the stream of each target receiver based on the received signal and Subsequently, transmitting to the target receiving end without causing interference; and
d) destination receiving end silver If , the desired stream can be restored to (α + L + 1), otherwise, the desired stream can be restored to (α + L) in the second timeslot; - here and and (α + L + 1) ≤ Md,
A cooperative communication method comprising: