KR20190086248A - Adaptive Perturbation-aided Opportunistic Beamforming method and MIMO system using the same - Google Patents

Abstract

The present invention relates to a transmission beamforming technique of a multi-antenna system. According to the present invention, in an opportunistic beamforming using only limited channel information after random beamforming, data transmission rate can be improved by making minor modifications to an additional memory portion of a base station (BS) and a decision block in a scheduler without additional information exchange between the BS and a mobile station (MS) or change in MS hardware design.

Description

[0001] The present invention relates to an opportunistic transmission beamforming method using adaptive perturbation, and a MIMO system using the opportunistic transmission beamforming method.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmission beam forming technique of a multi-antenna system.

Many previous studies have addressed diversity and opportunistic beamforming of multiple users to improve the performance of multi-user based wireless communication systems.

The diversity of multi-users is based on the fact that the channel conditions observed by other users include independent fading changes.

Each mobile station (MS) estimates an instantaneous channel state, which is referred to as a signal-to-noise ratio (SNR) and sent to a base station (BS).

The scheduler at the BS then uses this information to allocate resources to the MS with the best channel quality.

Therefore, the overall system can be greatly improved compared to the case where the processing performance does not utilize the diversity of multi-users.

In order for this idea to be implemented in a real system, the scheduled delay and fairness issues must be resolved.

As a simple algorithm to solve these problems, proportional fair scheduling (PFS) technique can be adopted, which not only achieves fairness among MSs but also selects MS in favorable channel conditions, Utilize user diversity gain.

In slow fade channels, however, small channel fluctuations decrease the multi-user diversity gain over time.

To solve this problem, an opportunistic beamforming (OBF) algorithm has been proposed.

In the OBF, the dynamic fluctuation of the effective channel condition observed by the MS through signal transmission of multiple antennas in which the amplitude and phase of the composite beamforming coefficients change in a pseudo random manner increases, which results in a larger multiuser diversity gain do.

However, even in OBF, the multi-user diversity gain is severely reduced in a small number of MSs, which degrades system performance.

To improve the overall system performance of the OBF, an advanced version of time-coherent-opportunistic beamforming (TC-OBF) has been introduced.

In the TC-OBF, the beamforming vector is stored in the memory of the BS according to the achievable data rate of each MS, and instead of transmitting the data using the current beamforming vector, the scheduler searches in memory for the maximum Data corresponding to the data rate is adopted.

In general, channel information is required for beamforming of a multi-antenna system. However, in a large multi-antenna system using a very large number of antennas, acquisition of channel information is difficult or limited.

Therefore, it is difficult to obtain the beam forming gain. To solve this problem, the OBF technique can be used, but it has a disadvantage that it has a lower data transmission rate than the conventional adaptive beam forming technique.

Korean Patent Laid-Open No. 10-2016-0081779 discloses a data transmission method using physical network coding in a relay station, comprising: an operation of receiving a first signal from a first node and a second signal from a second node; Generating a third signal to which a physical network coding is applied based on the first signal and the second signal, generating a third signal to which a physical network coding is applied based on an effective power of a first channel between the relay station and the first node and a second signal between the relay station and the second node Forming a beam that maximizes the low effective power of the channel's effective power, and transmitting the third signal to the first node and the second node based on the beam, Which is different from the existing adaptive beam forming technique.

The present invention improves performance by applying adaptive perturbation instead of generating beamforming weight completely randomly in order to obtain a high data transmission rate while utilizing only limited channel information We propose an opportunistic transmission beamforming method using adaptive perturbation.

The present invention relates to an opportunistic transmission beamforming method for a multi-antenna system, comprising the steps of: generating a beamforming weight for opportunistic beamforming, the beamforming weight having a highest data transmission rate, The method includes the steps of: generating an opportunistic transmission beam using an adaptive perturbation;

According to an aspect of the present invention, the perturbation value uses a value obtained by adding an adaptive perturbation weight to a random perturbation weight.

According to an aspect of the present invention, the adaptive perturbation weight is applied to a random weight when a random weight of a previous time brings about a transmission speed, and to a value of an opposite sign when a transmission speed is decreased .

According to another aspect of the present invention, there is provided a Multiple Input Multiple Output (MIMO) system to which an opportunistic transmission beam forming method (AP-OBF) using the adaptive perturbation is applied.

According to the present invention, an improved transmission rate can be obtained by applying the adaptive perturbation technique to existing opportunistic beamforming, and high performance can be obtained with low complexity by operating the multi-antenna system using only limited feedback information, It is possible to operate in a realistic large-scale multi-antenna system in which channel information is limited, thereby making it easy to apply the technology.

Also, it can be used as an effective signal transmission method for a 5G mobile communication system requiring a high transmission rate with low complexity.

FIG. 1 is a processing flowchart of an opportunistic transmission beam forming method using adaptive perturbation according to the present invention.

In the OBF and TC-OBF algorithms, beamforming vectors are randomly generated, which requires a very large number of MSs in order to obtain near optimal beamforming performance.

The present invention proposes a technique for generating a new beamforming vector according to current channel conditions rather than being generated randomly.

In other words, it aims to utilize feedback more efficiently and track channels that change over time.

When perturbations are applied to the beamforming vector, it can lead to performance improvement or degradation.

In the adaptive perturbation-aided OBF (AP-OBF) method using the adaptive perturbation according to the present invention, the BS adopts a perturbation that provides a higher transmission rate than the maximum of the previous transmission rates stored in the memory, It is abandoned.

In an opportunistic transmit beamforming method of a multi-antenna system, when generating a beamforming weight of opportunistic beamforming, the beamforming weight perturbing the beamforming weight that yields the highest data transmission rate among the beamforming weights generated for the immediately preceding M- If the perturbation by this method is accepted, the beamforming vector is likely to be close to the optimal value with the maximum transmission rate.

In addition, in addition to the random perturbation, the AP-OBF utilizes adaptive perturbation, which is generated based on the perturbation in the previous time slot.

Similar to the distributed beamforming technique based on the one-bit feedback information, for a selected MS in each time slot, if the perturbation leads to a performance improvement as compared to the maximum transmission rate of the memory, Which can lead to improved performance.

Conversely, if the perturbation causes a performance degradation, the opposite direction of this perturbation provides a performance improvement in the next time slot.

According to this idea, the AP-OBF technique can improve the received SNR in the current time slot using information about perturbation including the case where performance degradation occurred in the previous time zone.

를 생성하고, 이는 이전 시간의 랜덤 섭동과 동일 또는 반대 방향으로 설정된다.The proposed AP-OBF technique in time slot t (> 0) uses a random perturbation and adaptive perturbation to generate a new beamforming vector

Which is set in the same or opposite direction as the random perturbation of the previous time.

That is, the present invention uses a value obtained by adding an adaptive perturbation weight to a random perturbation weight as a total perturbation value, and the adaptive perturbation weight applies a random weight when a random weight of a previous time brings about a transmission speed improvement, When the speed is lowered, it is applied as the value of the opposite sign.

And transmits the pilot signal to the MS using the beamforming vector.

Since the improvement or the decrease of the transmission rate can be determined using the SNR feedback information transmitted from each MS, each MS estimates the received SNR and reports it to the BS.

Similar to the TC-OBF, the BS stores in the memory of size M an achievable data rate corresponding to the reported SNR as well as the beamforming vector.

를 선택한다.For scheduling, we use a modified PFS strategy, which is based on the following MS index < RTI ID = 0.0 >

.

<Formula 1>

는 최근 시간 슬롯 M동안의 MS k의 최대 데이터 속도를 나타낸다.here,

Lt; / RTI &gt; represents the maximum data rate of the MS k for the most recent time slot M.

의 경우, BS는 메모리에서 최대 데이터 전송속도

를 위한 빔형성 벡터를 발견한다.

, The BS can determine the maximum data rate

&Lt; / RTI &gt;

에 상응하는 빔형성 벡터가 데이터를

로 전송하는데 사용된다.Then, instead of the beamforming vector used for the pilot transmission, w (t)

Lt; RTI ID = 0.0 &gt; beamforming &lt; / RTI &

Lt; / RTI &gt;

The selected beamforming vector w [t] is also used to generate a new beamforming vector for time slot t + 1.

Referring to FIG. 1, the detailed operation of the AP-OBF according to the present invention will be described below.

)]는 랜덤하게 생성된다.The initial beamforming vector generation step (s1) generates the beamforming vector w [t] (=

)] Are randomly generated.

과 적응 섭동

은 빔형성 벡터 w[t-1]에 적용되는데, 이는 시간 슬롯(T-1)에서

로 데이터를 전송하는 데 사용된다.The new beamforming vector generation step s2, at time slot t (> 0), generates a random perturbation

And adaptive perturbation

Is applied to the beamforming vector w [t-1], which in time slot (T-1)

Lt; / RTI &gt;

는

복합 매트릭스의 공간을 나타낸다.here,

The

Represents the space of the composite matrix.

Specifically, the new beamforming vector is expressed by Equation (2).

<Formula 2>

은 표준화되지 않은 빔형성 벡터로서 식3에 의해 계산되며,

는 벡터값을 나타낸다.here,

Is calculated as an un-standardized beamforming vector by Equation 3,

Represents a vector value.

<Formula 3>

를 위해, 먼저 i.i.d, 영평균, 단위 변환 복합 가우시안 랜덤 변수로 구성되는

벡터를 생성한다. 그리고 나서 그것을 기준값(

)으로 나누어 정규화한다.Here,? And? Represent a perturbation coefficient.

, First of all, it is composed of iid, zero mean, and unit conversion complex Gaussian random variable

Create a vector. Then,

).

는 이전 시간슬롯에서 적용되었던 랜덤 섭동과 동일하거나 반대 방향으로 설정된다. Adaptive perturbation vector

Is set to be the same as or opposite to the random perturbation applied in the previous time slot.

이

에 대한 SNR 증가를 만들었다면 식4가 성립하고, 그렇지 않으면 식 5를 적용한다.Obviously, if the random perturbation of time slot t-1

this

If the SNR increase is made for Eq. (4), then Eq. (4) is established, otherwise Eq. (5) is applied.

<Formula 4>

&Lt; EMI ID =

로 설정된다.In the initial stage,

.

를 활용하여 모든 MS에 파일럿 신호를 전송한다.The step (s3) of transmitting the pilot signal to the MS is performed by the BS using the new normalized beamforming vector

To transmit pilot signals to all MSs.

The received SNR feedback step (s4) uses the pilot signal, and each MS measures the received SNR, which is fed back to the BS.

와 함께 메모리에 저장한다.In the transmission rate storage and scheduling step (s5), the BS collects SNRs in all MSs, calculates the data transmission rates achievable using Equation (6)

In the memory.

를 스케쥴링한다.The BS calculates

.

의 경우, BS는 메모리 내에서 최대 전송속도에 상응하는 빔형성벡터 w[t]를 발견한다.In the beamforming vector search step (s6) of the maximum transmission rate,

, The BS finds a beamforming vector w [t] corresponding to the maximum transmission rate in the memory.

에 전송하고 식6에 따른 각 MS의 평균 성능을 갱신한다.By using w [t], the BS sends the data

And updates the average performance of each MS according to Equation (6).

&Lt; EMI ID =

는 스케줄링 딜레이와 다중 사용자 다이버시티 이득 간의 트레이드오프 이다.here,

Is a tradeoff between the scheduling delay and the multi-user diversity gain.

로 진행해 가면서 단계 2)에서 단계 6)까지 반복된다.7) The algorithm

The process is repeated from step 2) to step 6).

을 사용하는 것보다는 메모리 내에 있는 가장 좋은 값을 사용하는데, 이는 이전 빔형성 벡터에 대해 랜덤하게 생성된 것뿐 아니라 적응적 구성 요소를 추가함으로써 생성된다.In Equation 3, to generate a new beamforming vector,

, Which is generated by adding an adaptive component as well as a randomly generated one for the previous beamforming vector.

This implies that the successful SNR of the scheduled user in the slow fading channel, where the same MS is likely to be scheduled over several time slots in a short time, .

In addition, by using both the performance improvement and the deteriorating perturbations, the AP-OBF technique can quickly adapt the beamforming vector to the channel, which leads to overall system performance improvement.

An advantage of the technique according to the invention is that it does not require any additional information exchange or MS hardware design changes between the BS and the MS, only a slight modification in the additional memory portion of the BS and the decision block in the scheduler.

The technical scope of the present invention includes a Multiple Input Multiple Output (MIMO) system to which the opportunistic transmission beam forming method (AP-OBF) described above is applied.

Claims (5)

An opportunistic transmit beamforming method for a multi-antenna system,
The method of claim 1, wherein, when generating the beam forming weight of the opportunistic beam forming, the beam forming weight is caused to perturb the beam forming weight which has the highest data transmission speed among the beam forming weights generated for the immediately preceding M hours . The method according to claim 1,
Wherein the perturbation value is a random perturbation weight plus an adaptive perturbation weight. 3. The method of claim 2,
Wherein the adaptive perturbation weight is applied to a random weight when a random weight of a previous time brings about a transmission speed and to a value of an opposite sign when a transmission speed is decreased. The method of claim 3,
Wherein the improvement or decrease of the transmission rate is determined using the SNR feedback information transmitted from each terminal. A Multiple Input Multiple Output (MIMO) system to which the opportunistic transmission beam forming method (AP-OBF) of any one of claims 1 to 4 is applied.

Images