KR102277866B1 - Hybrid analog and digital beamforming apparatus and method - Google Patents

Abstract

The present technology discloses a hybrid analog digital beamforming apparatus and method for a large-scale MIMO system. According to an embodiment of the present technology, hybrid analog-digital beamforming is performed by grouping the number of antennas of the sub-array differently according to the channel state, thereby improving the performance of a large-scale multi-input multi-output (MIMO) system with low computational complexity. .

Description

Hybrid analog digital beamforming apparatus and method for large-scale MIMO system {HYBRID ANALOG AND DIGITAL BEAMFORMING APPARATUS AND METHOD}

The present invention relates to a hybrid analog digital beamforming apparatus and method for a large-scale MIMO system, and more particularly, by optimizing and grouping the number of receiving antennas connected to each RF chain differently, the performance of the large-scale MIMO system can be improved and computational complexity technology that can reduce

In mobile communications, large-scale multiple-input multiple-output (MIMO) systems equipped with multiple antennas at base stations (BSs) have recently been considered to dramatically improve system performance in terms of spectral and energy efficiency.

In conventional frequency bands, precoding is typically handled only in the digital domain for interference mitigation between spatial sub-streams, whereby dedicated radio frequency (RF) chains and analog-to-digital or digital-to-analog converters (ADC/DACs) are required. used

Accordingly, the cost and power consumption of the transceiver increases in proportion to the number of antennas, and may lead to excessive power consumption in a large-scale MIMO system using a very large number of antennas.

Therefore, in next-generation mobile communication such as 5G mobile communication, the base station also utilizes hybrid beamforming that combines digital beamforming and analog beamforming. In this case, the flexibility of digital beamforming and the simplicity of multi-layer transmission and analog beamforming are combined. Thus, a large-scale MIMO system is implemented by efficiently increasing the number of antennas.

X. Gao, L. Dai, S. Han, I. Chih-Lin, and R. W. Heath, “Energy-efficient hybrid analog and digital precoding for mmWave MIMO systems with large antenna arrays,” IEEE J. Sel. Areas Commun., vol. 34, no. 4, pp. 998-1009, 2016. Y. Niu, Z. Feng, Y. Li, Z. Zhong, and D. Wu, “Low complexity and near-optimal beam selection for millimeter wave MIMO systems,” in IEEE 13th Int. Conf. Wireless Commun. Mobile Computing (IWCMC), 2017, pp. 634-639. S. Park, A. Alkhateeb, and R. W. Heath, “Dynamic subarrays for hybrid precoding in wideband mmWave MIMO systems,” IEEE Trans. Wireless Commun., vol. 16, no. 5, pp. 2907-2920, 2017.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a hybrid analog digital beamforming apparatus and method for a large-scale MIMO system capable of improving performance of the large-scale MIMO system.

Another object of the present invention is to provide a hybrid analog-to-digital beamforming apparatus and method for a large-scale MIMO system capable of reducing computational complexity.

Objects of the present invention are not limited to the objects mentioned above, and other objects not mentioned will be clearly understood by those skilled in the art from the following description.

According to one aspect for achieving the above object, a hybrid analog digital beamforming apparatus of a large-scale MIMO system of an embodiment,

A hybrid analog digital beamforming apparatus for a large-scale MIMO system, comprising:

a switching network connected to an output side of a plurality of sub-arrays including an antenna _{, a} low-noise amplifier, and a phase shift and outputting a grouped reception signal based on a switching signal supplied from the outside;

a digital receiver that digitizes and then decodes the received signal passing through the switching network; and

and a controller for generating a switching signal for grouping the antennas with a different number of antennas of the sub-array according to the channel state and transmitting the generated switching signal to the switching network.

Preferably, the switching network may be provided in an Unequal Sub Array (UESA) structure in which the number of antennas of the sub-array are grouped differently under the control of the controller.

Preferably, the control unit,

를 도출하고,By performing ordering on the channel matrix, the channel matrix

to derive,

에 대한 후보

의 서브 세트

에 각각에 대한 아날로그 결합 행렬

을 완전 탐색(ES: Exhaustive Search) 알고리즘을 수행하여 도출하고,set of the number of antennas

candidate for

a subset of

analog coupling matrix for each in

is derived by performing an exhaustive search (ES) algorithm,

및 채널 행렬

을 기반으로 기 정해진 관계식에 의거 채널 용량

을 연산하며, The derived analog combining matrix

and channel matrix

Channel capacity according to a predetermined relation based on

computes,

과 기 정해진 임계값 τ을 비교하여 비교 결과 채널 용량

이 상기 임계값 τ을 초과하는 경우 아날로그 결합 행렬

, 안테나 수의 세트

, 상기 임계값 τ 각각을 최적 아날로그 결합 행렬

, 최적 안테나 수의 세트

, 및 연산된 채널 용량

로 업데이트하도록 구비될 수 있다.The calculated channel capacity

By comparing the predetermined threshold value τ and the comparison result, the channel capacity

If this exceeds the threshold τ, the analog coupling matrix

, set of number of antennas

, each of the thresholds τ is an optimal analog combining matrix

, set of the optimal number of antennas

, and computed channel capacity

It may be provided to update to .

Preferably, the control unit,

조건을 만족하는 서브 어레이의 안테나 수의 세트

에 대한 후보

의 서브 세트

에 대해 최대 총 채널 용량을 가지는 안테나 수의 세트를 검색하는 감소된 완전 탐색 (RES: Reduced Exhaustive Search) 알고리즘을 수행하도록 구비될 수 있다.To reduce computational complexity, the number of antennas

A set of the number of antennas in a sub-array that satisfies the condition

candidate for

a subset of

It may be provided to perform a Reduced Exhaustive Search (RES) algorithm for searching a set of the number of antennas having the maximum total channel capacity for .

Preferably, the control unit,

에서 현재 찾아진 최적 안테나 세트

후보보다 더 높은 채널 용량의 안테나 세트

이 일정 횟수 반복 탐색하는 동안 발견되지 아니한 경우 감소된 완전 탐색(RES)을 조기 종료(ET)하도록 구비될 수 있다.sub-candidate set

The optimal antenna set currently found in

Antenna sets with higher channel capacity than candidates

If it is not found during the repeated search for a predetermined number of times, the reduced complete search (RES) may be terminated early (ET).

According to another aspect for achieving the above object, in the hybrid analog digital beamforming method for a large-scale MIMO system of an embodiment,

(a) grouping, in a switching network, a sub-array comprising a plurality of antennas, a low-noise amplifier, and a phase shift; and

(b) in a digital receiver, digitizing and then decoding the received signal that has passed through the switching network;

(c) generating a switching signal for grouping the number of antennas of the sub-array of the switching network differently according to the channel state by the controller and transmitting the generated switching signal to the switching network;

Preferably, step (a) may include a UESA (Unequal Sub Array) structure in which the number of antennas of the sub-array are grouped differently under the control of the controller.

Preferably, step (c) comprises:

를 도출하고,(c-1) channel matrix by performing ordering on the channel matrix

to derive,

에 대한 후보

의 서브 세트

에 각각에 대한 아날로그 결합 행렬

을 완전 탐색(ES: Exhaustive Search) 알고리즘을 수행하여 도출하고,(c-2) set of the number of antennas

candidate for

a subset of

analog coupling matrix for each in

is derived by performing an exhaustive search (ES) algorithm,

및 채널 행렬

을 기반으로 기 정해진 관계식에 의거 채널 용량

을 연산하며, (c-3) the derived analog combining matrix

and channel matrix

Channel capacity according to a predetermined relation based on

computes,

과 기 정해진 임계값 τ을 비교하여 비교 결과 채널 용량

이 상기 임계값 τ을 초과하는 경우 아날로그 결합 행렬

, 안테나 수의 세트

, 상기 임계값 τ 각각을 최적 아날로그 결합 행렬

, 최적 안테나 수의 세트

, 및 연산된 채널 용량

로 업데이트하도록 구비될 수 있다.(c-4) the calculated channel capacity

By comparing the predetermined threshold value τ and the comparison result, the channel capacity

If this exceeds the threshold τ, the analog coupling matrix

, set of number of antennas

, each of the thresholds τ is an optimal analog combining matrix

, set of the optimal number of antennas

, and computed channel capacity

It may be provided to update to .

Preferably, the step (c-1) comprises:

조건을 만족하는 서브 어레이의 안테나 수의 세트

에 대한 후보

의 서브 세트

에 대해 최대 총 채널 용량을 가지는 안테나 수의 세트를 검색하는 감소된 완전 탐색 (RES: Reduced Exhaustive Search) 알고리즘을 수행하도록 구비될 수 있다.To reduce computational complexity, the number of antennas

A set of the number of antennas in a sub-array that satisfies the condition

candidate for

a subset of

It may be provided to perform a Reduced Exhaustive Search (RES) algorithm for searching a set of the number of antennas having the maximum total channel capacity for .

Preferably after step (c-4)

에서 현재 찾아진 최적 안테나 세트

후보보다 더 높은 채널 용량의 안테나 세트

가 일정 횟수 반복 탐색하는 동안 발견되지 아니한 경우 감소된 완전 탐색(RES)을 조기 종료(ET)하도록 구비될 수 있다.sub-candidate set

The optimal antenna set currently found in

Antenna sets with higher channel capacity than candidates

It may be provided so that the reduced complete search (RES) is terminated early (ET) when it is not found during the repeated search for a predetermined number of times.

According to an embodiment, by performing hybrid analog-digital beamforming by grouping the number of antennas of the sub-array differently according to channel conditions, performance of a large-scale multi-input multi-output (MIMO) system can be improved with low computational complexity.

1 is a block diagram of a hybrid analog digital beamforming apparatus according to an embodiment.
2 is a conceptual diagram of a reduced full search (RES) process according to an embodiment.
3 is a conceptual diagram of a process of early stopping (ET) RES according to an embodiment.
4 is a graph illustrating an upper-bound of a total channel capacity and a channel capacity according to an embodiment.
5 is a graph showing total channel capacity according to an embodiment.
6 is a graph showing energy efficiency according to an embodiment.
7 is a graph of energy efficiency and total channel capacity in one embodiment.
8 is a flowchart illustrating a grouping operation of a sub-array according to an exemplary embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings and the content to be described later. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed subject matter may be thorough and complete, and that the spirit of the present invention may be sufficiently conveyed to those skilled in the art. Like reference numerals refer to like elements throughout. Meanwhile, the terms used in this specification are for the purpose of describing the embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used herein, “comprises” and/or “comprising” means that the stated component, step, operation and/or element is the presence of one or more other components, steps, operations and/or elements. or addition is not excluded.

Hereinafter, a hybrid analog digital beamforming technique capable of improving system performance by allocating and grouping unequal number of receive antennas to a sub-antenna array according to an embodiment of the present invention will be described in detail.

1 is a detailed configuration diagram of a base station BS that performs hybrid analog digital beamforming in large-scale MIMO according to an embodiment.

1, large-scale MIMO according to an embodiment is provided with a base station BS equipped with N _r receiving antennas 11, 12.. and N RF chains 51, 52, .., uplink When N < N _r , it includes K single-antenna terminal MSs.

That is, the base station BS includes N _r receiving antennas 11, 12.., and a low noise amplifier (LNA: Low Noise Amplifier: 21,22, ..) connected to the output terminals of the receiving antennas 11, 12, .. .) and a sub-antenna array including phase shifts 31, 32, .. are connected, a switching network 40 is connected to an output terminal of the sub-antenna array, and an RF chain 51 is connected to an output terminal of the switching network 40 , 52, ..) are connected. The switching network 40 is connected to the output side of a plurality of sub-arrays and is provided to transmit a grouped reception signal based on a switching signal supplied from the outside, whereby the switching network 40 sets the number of antennas of the sub-antenna array differently. It is provided in an unequal sub-array (UESA) structure.

The control unit 100 may be provided to generate the switching signal for setting the number of antennas of the sub-antenna array differently in the switching network 40 according to a channel state.

And, the base station BS is connected to an analog-to-digital converter (ADC: 61, 62, ...) that digitizes the received signal that has passed through the RF chain (51, 52, ..), and an analog combiner (ADC: 61, 62) , ...) to the output terminal of the digital receiver 70 is connected.

Accordingly, the received signal passing through the RF chains 51, 52, .. is digitized and then decoded by passing through the advanced digital receiver 70 such as old decoding and tabu search.

Accordingly, the base station BS having an unequal sub-array (UESA) structure dynamically adjusts to a radio frequency (RF) via the reception antennas 11 , 12 , ... and the switching network 40 based on the control unit 100 . Connected. Hereinafter, unless otherwise specified, the UESA system refers to a base station BS including the switching network 40 having an unequal sub-array (UESA) structure.

이고, 수신 안테나의 수 N _r 이 단일 안테나 단말 MS의 수 K 보다 크다고 가정하면, 이에 디지털 수신기(70)의 아날로그 결합 신호는 다음 식 1로 나타낼 수 있다.Here, the number of RF chains is equal to the number of single-antenna terminal MSs, that is,

, and it is assumed that the number of reception antennas N _r is greater than the number K of single-antenna terminal MSs, the analog combined signal of the digital receiver 70 can be expressed by Equation 1 below.

[Equation 1]

는 K 개 단일 안테나 단말 MS로 전송되는 벡터 심볼이고, 각 단일 안테나 단말의 평균 전송 전력은 정규화되어

이고,

는 독립적이고 균등 분산된 부가 백색 가우시안 노이즈(AWGN: additive white Gaussian noise) 샘플의 벡터이다. 여기서,

이다. 또한,

은 K 단말들과 기지국 BS 사이의 채널들을 나타내는 K 열 벡터

들로 구성된 채널 행렬로 나타낸다. 각각의 채널 엔트리

는 k 번째 단일 안테나 단말과 기지국 BS의 i 번째 수신 안테나 사이의 복소 채널 이득으로 나타낼 수 있다.here,

is a vector symbol transmitted to K single-antenna terminal MS, and the average transmission power of each single-antenna terminal is normalized.

ego,

is a vector of independent and uniformly distributed additive white Gaussian noise (AWGN) samples. here,

to be. Also,

is a K column vector representing channels between K terminals and a base station BS.

It is represented by a channel matrix composed of each channel entry

may be expressed as a complex channel gain between the k- th single antenna terminal and the i-th reception antenna of the base station BS.

는 다음과 같이 주어진다.and the analog coupling matrix

is given as

는 서브 안테나 어레이의 아날로그 가중치 벡터이고,

는

의 m 번째 아날로그 가중치 벡터이며, 일정한 진폭

를 가지나 위상이 다르다. 즉,

이다.here,

is the analog weight vector of the sub-antenna array,

is

is the mth analog weight vector of , constant amplitude

has a different status. In other words,

to be.

이라고 가정하면, 할당된 안테나의 수는

이고,

이다. On the other hand, the number of antennas allocated to the n-th sub-array in the switching network 40

Assuming that , the number of allocated antennas is

ego,

to be.

은

이 된다. 그리고 서브 어레이에 할당된 스위칭 네트워크(40)의 수신 안테나(11, 12, ...) 수 세트를

로 정의하면,

이다.And, the number of receiving antennas

silver

becomes this And a set of the number of receive antennas (11, 12, ...) of the switching network 40 allocated to the sub-array

If defined as

to be.

의 최대 고유치를

로 정의하면, 총 채널 용량(total achievable rate)

의 상계(upper bound)

는 식 2로 얻을 수 있으며, 식 2에 Jense의 부등성을 적용하면 총 채널용량의 상계는 식 3을 만족한다. And, in the UESA system

the maximum eigenvalue of

If defined as , the total channel capacity (total achievable rate)

upper bound of

can be obtained from Equation 2, and if Jense's inequality is applied to Equation 2, the upper bound of the total channel capacity satisfies Equation 3.

[Equation 2]

[Equation 3]

은

으로, 반양정치(positive semidefinite)의 에르미티안(Hermitian) 행렬로 고유 벡터 분해가 가능하고, 이에

의 고유 벡터

와 아날로그 결합 벡터

같을 때

의 최대 고유치는

를 만족할 때 식 3의 등식이 만족된다.here,

silver

, eigenvector decomposition is possible with a positive semidefinite Hermitian matrix,

eigenvector of

and analog combined vector

when the same

The maximum eigenvalue of

When , the equation of Equation 3 is satisfied.

의 최대 고유치가 식 4를 만족하도록 설계된다.To optimize the thumb rate from Equation 3, the UESA system of an embodiment is

The maximum eigenvalue of is designed to satisfy Equation 4.

[Equation 4]

Since it is difficult to optimize the thumb rate to satisfy the condition of Equation 4, Equation 4 is relaxed to the condition of Equation 5 below.

[Equation 5]

를 최대화함에 따라, 최대 고유치의 합

의 최대화 문제는 해결될 수 있다. Also

By maximizing , the sum of the largest eigenvalues

The maximization problem of can be solved.

(목표

)이고,

의 최대화(목표

) 되도록 구현되어야 한다.Accordingly, the UESA system of one embodiment is

(goal

)ego,

maximization of (goal

) should be implemented as much as possible.

는 다르므로, 일 실시 예의 UESA 시스템은 주어진 서브 어레이의 안테나 수의 세트

에 대해, 채널 용량을 최적화되어야 한다.In addition, as the number of sub-arrays and the number of receiving antennas 11, 12, .. are arranged differently, the number of antennas in the sub-array

is different, so the UESA system of one embodiment is a set of the number of antennas in a given sub-array.

For this, the channel capacity should be optimized.

를 달성하기 위해 서브 어레이에 할당된 안테나의 수의 세트

는 비 내림 차순으로 배열되어야 한다. 즉,

이다. 이에 일 실시 예의 UESA 시스템은 총 채널용량

가 상계

에 점점 가까워질수록 증가된 채널 용량을 확보할 수 있고, 이에 UESA 시스템에서 총 채널 용량은 ESA 시스템의 총 채널용량과 비교하여 상계

에 좀더 근접된다. Therefore, the goal

A set of the number of antennas allocated to the sub-array to achieve

should be arranged in non-descending order. In other words,

to be. Accordingly, the UESA system of an embodiment has a total channel capacity

virtual set-up

It is possible to secure increased channel capacity as it gets closer to , and thus, the total channel capacity in the UESA system is offset compared to the total channel capacity of the ESA system.

closer to the

에서 목표

를 만족하기 위해, 단일 안테나 단말 MS의 수 K 가 일정하게 유지되는 동안 수신 안테나의 수 N _r 이 증가될 때

의 상계

는 채널 행렬

의 열(row)을 놈(norm)의 내림 차순 정렬에 의해 최대화할 수 있다. 여기서,

는 기지국 BS 와 K 개의 단일 안테나 단말 MS들 사이의 채널

을 나타내고

의 행 벡터를 포함하는 채널 행렬이다.And,

goal in

When the number of receive antennas N _r increases while the number K of single-antenna terminal MSs is kept constant in order to satisfy

offset of

is the channel matrix

We can maximize the row of , by sorting in descending order of the norm. here,

is a channel between the base station BS and K single-antenna terminal MSs

to indicate

A channel matrix containing a row vector of .

Accordingly, the UESA system according to an embodiment performs proximity optimization on the number of antennas of a sub-array having low computational complexity.

로 정의하면, 가능한 모든 후보 안테나 수의 세트

중 가장 높은 채널용량을 제공할 수 있다.The UESA system of one embodiment determines a set of optimal number of antennas in the sub-array.

Defined as , the set of all possible candidate antenna numbers

It can provide the highest channel capacity among them.

를 결정하기 위해,

가 최대화된 채널 순서화를 수행한 다음 완전 탐색(ES : exhaustive search) 알고리즘을 통해 모든 후보들에 대해 탐색하며, 후보는 다음과 같이 정의된다.Set of optimal number of antennas in these subarrays

to determine,

After maximizing channel ordering, all candidates are searched through an exhaustive search (ES) algorithm, and the candidates are defined as follows.

를 찾은 다음 최대 썸 레이트를 제공하는 후보 세트를 검색한다. 이에 대규모 MIMO 시스템에서 후보의 크기

는 매우 커지고, 이에 높은 연산 복잡도를 가지며, 낮은 연산 복잡도를 위해 후보 세트

에서의 탐색을 실행한다.In addition, the control unit 100 according to an embodiment performs an analog coupling matrix through an exhaustive search (ES) algorithm for all candidates.

, then search for the set of candidates that give the maximum thumb rate. Accordingly, the size of candidates in large-scale MIMO systems

is very large, has high computational complexity, and is a candidate set for low computational complexity.

Execute a search in

에서 서브 어레이의 최적 안테나 수의 세트

를 검색하는 대신에

조건의 서브 어레이의 안테나 수의 세트

에 대한 후보

의 서브 세트

에 대해 ES 알고리즘을 수행하는 감소된 완전 탐색(RES:Reduced Exhaustive Search) 알고리즘을 수행한다. In order to reduce such computational complexity, the UESA system of an embodiment has most of the space

set of optimal number of antennas in the subarray

instead of searching for

A set of the number of antennas in the sub-array of conditions

candidate for

a subset of

A Reduced Exhaustive Search (RES) algorithm that performs the ES algorithm on .

The controller 100 of the UESA system shown in FIG. 1 performs a Reduced Exhaustive Search (RES) for searching the optimal analog combining matrix and the optimal number of antennas of the sub-array from each candidate with low computational complexity. can do.

의 도출한 다음 스텝 2 내지 4에서 서브 세트

의 후보들을 동등하게 검색하며, 스텝 5 및 6에서 각 후보에 대해 아날로그 결합 행렬

을 도출한 다음 도출된 채널 행렬

및 아날로그 결합 행렬

를 토대로 기 정해진 관계식에 의거 총 채널 용량

을 도출한다. 여기서, 관계식은

이다. 2 is a flowchart illustrating a reduced complete search (RES) process by the control unit 100. Referring to FIG. 2, the control unit 100 orders the norm column matrix in descending order with respect to the channel matrix to form a channel matrix.

Subsets from steps 2 to 4 after derivation of

Equally search the candidates of , and for each candidate in steps 5 and 6, the analog combining matrix

After deriving , the derived channel matrix

and analog coupling matrix

Total channel capacity according to a predetermined relation based on

to derive Here, the relation

to be.

이 정해진 임계값 τ 보다 큰 경우 아날로그 결합 행렬

는 최적 아날로그 결합 행렬

을 도출하고, 이때 안테나 세트

은 최적 안테나 세트로 설정한다.And in steps 7 to 11, the control unit 100 is the derived total channel capacity

If is greater than a given threshold τ, the analog coupling matrix

is the optimal analog combining matrix

, where the antenna set

is set as the optimal antenna set.

로 임계값을 업데이트하며, 더 높은 썸레이트를 가지는 아날로그 결합 행렬

및 안테나 세트

이 검색될 때 마다 최적 아날로그 결합행렬

및 최적 안테나 수의 세트

을 업데이트한다.After that, the control unit 100 controls the total channel capacity.

Analog combining matrix with higher thumb rate, updating the threshold with

and antenna set

The optimal analog combination matrix each time is searched for

and set of optimal number of antennas

update

에서 현재 찾아진 최적 안테나 세트

후보보다 더 높은 채널 용량의 안테나 세트

가 일정 횟수 count 반복 탐색하는 동안 발견되지 아니한 경우 감소된 완전 탐색(RES) 알고리즘을 조기 종료(ET)하도록 구비될 수 있다.3 is a flowchart illustrating an Early Termination (ET) process after a reduced complete search of the controller 100. Referring to FIG. 3 , the controller 100 controls a sub-candidate set

The optimal antenna set currently found in

Antenna sets with higher channel capacity than candidates

It may be provided to prematurely terminate (ET) the reduced complete search (RES) algorithm when is not found during the iterative search for a predetermined number of counts.

이 최적 안테나의 세트

일 가능성이 높다. 따라서, 서브 후보 세트

의 안테나 세트

후보는 안테나 세트

의 인접된 요소들 간의 차를 토대로 배열된다In this case, an antenna set having a large difference between adjacent elements

This set of optimal antennas

it is likely to be Therefore, a set of sub-candidates

Antenna set of

Candidate set of antennas

is arranged based on the difference between adjacent elements of

를 모두 0으로 설정한 다음 서브 세트

의 i 번째 후보를

로 설정한다.That is, referring to FIG. 3 , in steps 1 to 3, the control unit 100

set to all zeros and then subset

the i-th candidate of

set to

과 n 번째 서브 어레이에 할당된 안테나의 수

의 차인

으로

을 연산하고, 단계 5에서

를 연산한다. 여기서,

이다.And in step 4, the control unit 100 determines the number of antennas allocated to the n+1-th sub-array.

and the number of antennas allocated to the nth sub-array

car of

to

, and in step 5

calculate here,

to be.

가 더 크다는 의미는 후보

의 요소들 사이에 차가 크다는 것이다. 이에 결과적으로,

가 큰 후보

는 유망한 후보로 간주되어 완전 탐색을 먼저 수행된다. 따라서, 단계 7에서, 서브 후보

의 원소들은 벡터

의 원소를 내림차순으로 분류되어

이 도출되고, 단계 8 내지 26에서,

의 후보를 하나씩 완전 탐색하여 최적의 아날로그 결합 행렬

및 최적의 안테나 수

가 결정된다. 여기서 서브 세트

에 대해 완전 탐색하는 일련의 과정은 도 2에 도시된 바와 같다. here

is greater means that the candidate

There is a large difference between the elements of As a result,

a big candidate

is considered a promising candidate and a full search is performed first. Therefore, in step 7, the sub-candidate

The elements of is a vector

are sorted in descending order

is derived, in steps 8 to 26,

An optimal analog combining matrix by exhaustively searching the candidates of

and optimal number of antennas

is decided subset here

A series of processes for a complete search is shown in FIG. 2 .

를 도출한 다음 각 후보에 대해 아날로그 결합 행렬

을 도출하고, 도출된 채널 행렬

및 아날로그 결합 행렬

를 토대로 기 정해진 관계식에 의거 총 채널 용량

을 도출하며, 도출된 총 채널 용량

이 정해진 임계값

보다 큰 경우 아날로그 결합 행렬

는 최적 아날로그 결합 행렬

을 도출하고, 이때 안테나 세트

은 최적 안테나 세트로 설정한다. That is, based on steps 8 to 15, the control unit 100 orders the channel columns in descending order of the norm of the channel matrix,

and then for each candidate the analog combining matrix

, and the derived channel matrix

and analog coupling matrix

Total channel capacity according to a predetermined relation based on

, and the derived total channel capacity

this fixed threshold

Analog Combination Matrix if Greater Than

is the optimal analog combining matrix

, where the antenna set

is set as the optimal antenna set.

를 사용하여 새로운 후보가 더 이상의 성능 향상을 제공하지 않는 반복 수를 카운팅하고, 각 후보

에 대해 총 채널용량이 향상되지 아니한 각 후보의 경우, 카운터

는 1씩 증가되고, 그렇지 않으면 0으로 재설정된다. 이 후 카운터가

에 도달하면 최적의 후보가 이미 완전 탐색되었을 가능성이 높으므로 단계 24에서 감소된 완전 탐색(RES) 알고리즘은 조기 종료(ET)된다.After that, the control unit 100 in steps 16 to 23, the counter

is used to count the number of iterations where the new candidate does not provide any further performance improvement, and each candidate

For each candidate whose total channel capacity has not improved for

is incremented by 1, otherwise it is reset to 0. After this, the counter

When , it is highly likely that the best candidate has already been fully searched, so the reduced complete search (RES) algorithm in step 24 is prematurely terminated (ET).

In this case, the total power consumption of the UESA system can be expressed by Equation 6 below.

[Equation 6]

,

,

,

, 및

는 각각 LNA(21, 22, ..), 위상 시프트(31, 32,...), 스위칭 네트워크(40), RF 체인부(51, 52, ...), 및 ADC(61, 62, ..) 각각의 소비 전력이다.here,

,

,

,

, and

LNA (21, 22, ..), phase shift (31, 32, ...), switching network 40, RF chain section (51, 52, ...), and ADC (61, 62, respectively) ..) is the power consumption of each.

의 전력 소비가 추가적으로 필요하다. 그러나 스위칭 네트워크(40)에 의해 소모되는 전력은 상대적으로 작기 때문에 총 소비 전력 전체에는 아무런 영향이 미치지 아니한다. Accordingly, the UESA system uses the switching network 40 to

additional power consumption is required. However, since the power consumed by the switching network 40 is relatively small, the total power consumption is not affected at all.

Optimization simulation results

에 대해 유효 채널 경로의 수를

으로 설정된다. 도착 방위각 AoA

은

에서 균일 분포하다고 가정하고, 기지국 BS의 수신 안테나 어레이에는 반 파장의 안테나 간격의 ULA(uniform linear array)로 가정한다. 아날로그 빔형성기의 각 계수의 위상은

로 제한되고 Q는 16으로 설정된다. 마지막으로, SNR은 잡음 전력

에 대한 사용자의 평균 심볼 전력의 비율로 정의된다.Assume the same number of effective channel paths between each single antenna terminal MS and the base station BS,

number of effective channel paths for

is set to Arrival Azimuth AoA

silver

is assumed to be uniformly distributed in , and a uniform linear array (ULA) of half-wavelength antenna spacing is assumed in the receiving antenna array of the base station BS. The phase of each coefficient of the analog beamformer is

, and Q is set to 16. Finally, SNR is the noise power

It is defined as the ratio of the user's average symbol power to

와

각각에 대한 ESA 시스템 및 UESA 시스템 각각 총 채널 용량(Rate) 및

의 상계(Upper Bound: UBs)를 보인 그래프이다. 4 shows SNR = 12dB.

Wow

ESA system and UESA system respectively for each total channel capacity (Rate) and

It is a graph showing the Upper Bound (UBs) of

각각에 대해

를 가정하고, 도 4의 (a) 및 (b)를 참조하면, UESA-ES 시스템의 총 채널 용량 및 채널 용량의 상계가 다른 시스템에 비해 높으며, UESA-RES 및 UESA-RES-ET 시스템은 UESA-ES 대비 약간 낮은 것을 알 수 있다. In the UESA system performing the Early Stop (ET) algorithm of the Reduced Complete Search Algorithm (RES)

for each

, and referring to FIGS. 4 (a) and 4 (b), the total channel capacity of the UESA-ES system and the upper limit of the channel capacity are higher than other systems, and the UESA-RES and UESA-RES-ET systems are It can be seen that it is slightly lower than -ES.

For example, the UESA-ES system has approximately 10.5% higher total channel capacity compared to the ESA scheme when Nr = 32, N = K = 4, and SNR = 0 dB, whereas the UESA-ES and UESA-RES-ET systems have approximately 10%. The total channel capacity of the UESA-RES-ET system is almost the same as that of the UESA-RES system, despite the reduced search area.

을 제공 할 수 없다. 반대로 UESA 시스템은 아날로그 하이브리드 빔 형성 네트워크의 구현에 보다 유연하게 적용 할 수 있으며 스펙트럼 및 에너지 효율성간에 향상된 균형을 달성할 수 있다. 5 is a graph showing the total channel capacity of each ESA system and UESA system for Nr = 64 and RF chain N = {2, 3, 4, 5}, respectively, at SNR = 12 dB. 5, since the antenna is equally allocated to the sub-antenna arrays of the ESA system,

cannot provide Conversely, UESA systems can be more flexibly applied to the implementation of analog hybrid beamforming networks and achieve an improved balance between spectral and energy efficiency.

In addition, the UESA system shows better system performance than the ESA system, and it can be seen that the performance improvement of the system for a large value of N is more clear.

이고, p = 20 mW 의 기준 전력값이며, 각각의 총 에너지는

로 설정된다.On the other hand, in terms of energy efficiency, each

, p = a reference power value of 20 mW, and each total energy is

is set to

와

각각에 대한 ESA 시스템 및 UESA 시스템 각각의 에너지 효율을 보인 그래프로서, 도 6을 참조하면, UESA-ES 시스템은 ESA 시스템보다 높은 전력을 필요로 하지만 Nr = 64의 경우 ESA 시스템과 비슷한 에너지 효율을 달성한다는 것을 알 수 있다. 또한 Nr = 32 인 경우 낮은 SNR 값에서 UESA 시스템의 에너지 효율 향상됨을 확인할 수 있다.6 is

Wow

As a graph showing the energy efficiency of each ESA system and UESA system for each, referring to FIG. 6 , the UESA-ES system requires higher power than the ESA system, but achieves similar energy efficiency to the ESA system for Nr = 64 it can be seen that In addition, when Nr = 32, it can be seen that the energy efficiency of the UESA system is improved at a low SNR value.

, 및

에 대해 ESA 시스템, UESA-ES 시스템, UESA-RES 시스템, 및 UESA-RES-ET 시스템 각각에 대한 에너지 효율 및 총 채널 용량을 보인 그래프로서, 도 7을 참조하면, 총 채널 용량에 대한 향상은 Nr의 모든 값에 대해 확인할 수 있고, UESA 시스템의 에너지 효율면에서의 이득은 Nr이 증가함에 따라 감소됨을 알 수 있다.7 shows SNR = 6 dB,

, and

As a graph showing energy efficiency and total channel capacity for each of the ESA system, the UESA-ES system, the UESA-RES system, and the UESA-RES-ET system, referring to FIG. 7 , the improvement to the total channel capacity is Nr It can be confirmed for all values of , and it can be seen that the gain in terms of energy efficiency of the UESA system decreases as Nr increases.

On the other hand, compared with the optimal UESA-ES system, in steps 1 to 7 of the UESA-RES and UESA-RES-ET systems, candidate ordering can be performed offline, so the computational complexity for this can be ignored. In addition, the complexity of row ordering of the channel is significantly lower than that of the full search algorithm (ES) for a large number of candidates. That is, the computational complexity is proportional to the number of candidates.

, 및

에 대해, UESA-ES 시스템, UESA-RES 시스템, 및 UESA-RES-ET 시스템의 후보의 수와 UESA-RES-ET 시스템에 대한

는 하기 표 1에서 제시된다.SNR = 12dB,

, and

For the number of candidates for the UESA-ES system, the UESA-RES system, and the UESA-RES-ET system, and for the UESA-RES-ET system

is presented in Table 1 below.

Referring to Table 1, it can be seen that the UESA-ES system requires a greater number of candidates than the UESA-RES system and the UESA-RES-ET system. In addition, since the reduced complete search algorithm is performed for the UESA-RES system and the UESA-RES-ET system, it can be confirmed that the performance loss of the system occurs but the computational complexity is remarkably reduced. In large-scale MIMO equipped with Nr = 64 antennas and N = 4 RF chains, the UESA-ES system must search about 40,000 candidates to find the optimal solution. However, the UESA-RES system and the UESA-RES-ET system - A full search should be performed on 1906 and 770 candidates, which correspond to about 5% and 2% of the number of candidates in the ES system.

In the end, it can be seen that the power consumption is slightly increased in the proposed UESA system, but the total channel capacity is improved by up to 10%. In the UESA-RES and UESA-RES-ET systems, the computational complexity of determining the antenna-to-subarray connection can be reduced. , it can be seen that the computational complexity of the UESA-ES system is significantly lowered, thereby minimizing system performance loss.

8 is a flowchart illustrating a hybrid analog digital beamforming operation for a large-scale MIMO system according to an embodiment. Referring to FIG. 8, the number of antennas of a sub-array is grouped differently according to a channel state to provide a hybrid analog for a large-scale MIMO system. It may be provided for digital beamforming.

A computer-readable recording medium in which a program for executing a hybrid analog-to-digital beamforming method for a large-scale MIMO system according to an embodiment is recorded may be provided. The program may include at least one of an application program storing the hybrid analog digital beamforming method, a device driver, firmware, middleware, a dynamic link library (DLL), and an applet. The hybrid analog digital beamforming system includes a control unit, and the control unit groups the number of antennas of the sub-array differently according to the channel condition and reads the recording medium in which the hybrid analog digital beamforming method for a large-scale MIMO system is recorded. Accordingly, a hybrid analog digital beamforming method for a large-scale MIMO system may be performed by grouping the number of antennas of the sub-array differently.

를 도출할 수 있다.Referring to FIG. 8 , in step S11 , the hybrid analog digital beamforming apparatus for a large-scale MIMO system according to an embodiment performs ordering on the channel matrix to

can be derived.

에 대한 후보

의 서브 세트

에 각각에 대한 아날로그 결합 행렬

을 완전 탐색(ES: Exhaustive Search) 알고리즘을 수행하여 도출하고, 상기 도출된 아날로그 결합 행렬

및 채널 행렬

을 기반으로 기 정해진 관계식에 의거 채널 용량

을 연산한다.And, in steps S12 and S13, the hybrid analog digital beamforming apparatus according to an embodiment sets the number of antennas.

candidate for

a subset of

analog coupling matrix for each in

is derived by performing an exhaustive search (ES) algorithm, and the derived analog combining matrix

and channel matrix

Channel capacity according to a predetermined relation based on

calculate

과 기 정해진 임계값 τ을 비교하여 비교 결과 채널 용량

이 상기 임계값 τ을 초과하는 경우 아날로그 결합 행렬

, 안테나 수의 세트

, 상기 임계값 τ 각각을 최적 아날로그 결합 행렬

, 최적 안테나 수의 세트

, 및 연산된 채널 용량

로 업데이트한다.In addition, in steps S14 and S15, the hybrid analog digital beamforming apparatus according to an embodiment performs the calculated channel capacity.

By comparing the predetermined threshold value τ and the comparison result, the channel capacity

If this exceeds the threshold τ, the analog coupling matrix

, set of number of antennas

, each of the thresholds τ is an optimal analog combining matrix

, set of the optimal number of antennas

, and computed channel capacity

update to

조건을 만족하는 서브 어레이의 안테나 수의 세트

에 대한 후보

의 서브 세트

에 대해 최대 총 채널 용량을 가지는 안테나 수의 세트를 검색하는 감소된 완전 탐색 (RES: Reduced Exhaustive Search) 알고리즘을 수행할 수 있다.Here, in order to reduce computational complexity, the hybrid analog digital beamforming apparatus according to an embodiment includes the number of antennas.

A set of the number of antennas in the sub-array that satisfy the condition

candidate for

a subset of

A Reduced Exhaustive Search (RES) algorithm that searches for a set of antenna numbers with the maximum total channel capacity can be performed.

에서 현재 찾아진 최적 안테나 세트

후보보다 더 높은 채널 용량의 안테나 세트

이 일정 수 반복하는 동안 발견되지 아니한 경우 감소된 완전 탐색(RES)을 조기 종료(ET)한다.On the other hand, in step S16, the hybrid analog digital beamforming apparatus according to an embodiment is a sub-candidate set

The optimal antenna set currently found in

Antenna sets with higher channel capacity than candidates

If it is not found during this certain number of iterations, the reduced complete search (RES) is terminated early (ET).

Accordingly, according to an embodiment, hybrid analog-digital beamforming is performed by grouping the number of antennas of the sub-array differently according to channel conditions, so that large-scale multi-input multi-output (MIMO) system performance can be improved with low computational complexity.

Although the present invention has been described in detail through representative embodiments above, those of ordinary skill in the art to which the present invention pertains can make various modifications to the above-described embodiments without departing from the scope of the present invention. will understand Therefore, the scope of the present invention should not be limited to the described embodiments, and should be defined by all changes or modifications derived from the claims and equivalent concepts as well as the claims to be described later.

40: switching network
100 ; control

Claims (11)

A hybrid analog digital beamforming apparatus for a large-scale MIMO system, comprising:
a switching network connected to an output side of a plurality of sub-arrays including an antenna, a low-noise amplifier, and a phase shift and outputting a grouped reception signal based on a switching signal supplied from the outside;
a digital receiver that digitizes and then decodes the received signal passing through the switching network; and
Further comprising: a controller for generating a switching signal for grouping the number of antennas of the sub-array differently according to a channel state and transmitting it to the switching network,
The control unit is
channel matrix

Channel matrix by ordering on

to derive,
set of antennas

candidate for

a subset of

analog coupling matrix for each in

is derived by performing an exhaustive search (ES) algorithm,
The derived analog combining matrix

and channel matrix

Channel capacity based on a predetermined relational expression based on

computes,
The calculated channel capacity

By comparing the predetermined threshold value τ and the comparison result, the channel capacity

If this exceeds the threshold τ, the analog coupling matrix

, set of number of antennas

, each of the thresholds τ is an optimal analog combining matrix

, set of the optimal number of antennas

, and computed channel capacity

A hybrid analog-to-digital beamforming apparatus for a large-scale MIMO system, characterized in that it is updated to . The method of claim 1, wherein the switching network is
and a UESA (Unequal Sub Array) structure in which the number of antennas of the sub-array is grouped differently under the control of the controller. delete According to claim 2, wherein the control unit,
In order to reduce the computational complexity, the number of antennas allocated to the i-th sub-array is

When , set of the number of antennas in a given subarray

About

A set of the number of antennas in a sub-array that satisfies the condition

candidate of

a subset of

A hybrid analog-to-digital beamforming apparatus for a large-scale MIMO system, characterized in that it is provided to perform a Reduced Exhaustive Search (RES) algorithm for searching a set of the number of antennas having the maximum total channel capacity. According to claim 4, wherein the control unit,
sub-candidate set

The optimal antenna set currently found in

Antenna sets with higher channel capacity than candidates

A hybrid analog-to-digital beamforming apparatus for a large-scale MIMO system, characterized in that it is provided to terminate the reduced complete search (RES) early (ET) if it is not found during the repeated search for a predetermined number of times. A hybrid analog digital beamforming method for a large-scale MIMO system, comprising:
(a) grouping, in a switching network, _{a sub-array comprising a plurality of antennas, a} low-noise amplifier, and a phase shift; and
(b) in the digital receiver, digitizing and then decoding the received signal that has passed through the switching network;
(c) generating, by the control unit, a switching signal for grouping the number of antennas of the sub-array of the switching network differently according to a channel state, and transmitting the generated switching signal to the switching network;
The step (c) is,
(c-1) channel matrix

Channel matrix by ordering on

to derive,
(c-2) set of antenna numbers

candidate for

a subset of

analog coupling matrix for each in

is derived by performing an exhaustive search (ES) algorithm,
(c-3) the derived analog combining matrix

and channel matrix

Channel capacity according to a predetermined relation based on

computes,
(c-4) the calculated channel capacity

By comparing the predetermined threshold value τ and the comparison result, the channel capacity

If this exceeds the threshold τ, the analog coupling matrix

, set of number of antennas

, each of the thresholds τ is an optimal analog combining matrix

, set of the optimal number of antennas

, and computed channel capacity

A hybrid analog-to-digital beamforming method for a large-scale MIMO system, characterized in that it is updated to . The method of claim 6, wherein (a) step,
and a UESA (Unequal Sub Array) structure in which the number of antennas of the sub-array is grouped differently under the control of the controller. delete The method of claim 7, wherein the step (c-1) comprises:
In order to reduce the computational complexity, the number of antennas allocated to the i-th sub-array is

When , set of the number of antennas in a given subarray

About

A set of the number of antennas in the sub-array that satisfy the condition

candidate of

a subset of

A hybrid analog-to-digital beamforming method for a large-scale MIMO system, characterized in that it is provided to perform a Reduced Exhaustive Search (RES) algorithm for searching a set of the number of antennas having the maximum total channel capacity for . The method of claim 7, wherein after step (c-4)
sub-candidate set

The optimal antenna set currently found in

Antenna sets with higher channel capacity than candidates

A hybrid analog-to-digital beamforming method for a large-scale MIMO system, characterized in that the reduced complete search (RES) is terminated early (ET) if it is not found during the repeated search a predetermined number of times. [Claim 11] A recording medium having a built-in program capable of executing the hybrid analog digital beamforming method for a large-scale MIMO system according to any one of claims 6, 7, 9, and 10 on a computer.

Images