雙碼本波束成形相關(guān)技術(shù)研究
發(fā)布時間：2018-08-18 16:21
【摘要】：多天線技術(shù)能夠利用空間自由度實現(xiàn)空間分集和復(fù)用，有效提高頻譜利用率。其中，波束成形是在多天線系統(tǒng)中實現(xiàn)空分多址的主要技術(shù)，主要分為兩類，，一類是長時的波束成形技術(shù)，它應(yīng)用于小間距的多根天線陣列，主要原理是利用空間信道的強(qiáng)相關(guān)性及波的干涉原理產(chǎn)生強(qiáng)方向性的輻射方向圖，使輻射方向圖的主瓣自適應(yīng)地指向用戶來波方向；另一類是短時的波束成形技術(shù)，它的主要原理是基于實時的信道沖擊響應(yīng)，利用天線陣元之間的弱相關(guān)性構(gòu)建信號處理算法，當(dāng)發(fā)送端和接收端都具有完整的信道信息，短時波束成形技術(shù)能夠簡單有效地實現(xiàn)MIMO信道的最大分集增益。當(dāng)天線距離在半波長到10倍波長之間，角度擴(kuò)散不大時，為長時波束成形和短時波束成形結(jié)合起來提供了條件，因而，能使二者結(jié)合的雙碼本技術(shù)被引入到LTE-Advanced（LTE-A）系統(tǒng)中，它同時用兩個碼本分別指示長時波束成形特性（碼本1）和短時波束成形特性（碼本2）。此外，由于長時特性和短時特性的變化周期不相同，決定了二者反饋周期也不同，這樣就可以延長長時特性碼本的反饋周期，從而減小平均反饋量，提高資源利用率。雖然在LTE-A中定義了雙碼本發(fā)送方案，但是雙碼本估計和反饋等都是雙碼本技術(shù)中亟待解決的關(guān)鍵問題。 本論文就是圍繞著雙碼本波束成形技術(shù)的研究展開的，通過對相關(guān)理論研究，提出了一系列解決方案�？偟膩碚f，本文主要研究內(nèi)容和貢獻(xiàn)如下： 首先，研究了雙碼本波束成形技術(shù)的碼本設(shè)計特性和碼本估計算法。在寬帶信號的正交頻分多路復(fù)用（Orthogonal Frequency Division Multiplexing，OFDM）系統(tǒng)中，不同子載波之間存在頻率偏移，所以在利用不同載波上的信道信息進(jìn)行獨立碼本估計時存在偏差，本文提出相關(guān)估計算法，利用寬帶信號中多個載波之間的相關(guān)性，將不同載波的信道信息聚焦到某個參考頻點上獲得等效的信道信息。此外，在將多采樣點估計獲得的信道矩陣進(jìn)行平均來獲得統(tǒng)計信道信息時，引入采樣點時刻的信噪比作為加權(quán)因子進(jìn)行加權(quán)平均，可以克服由于個別噪聲干擾較大的采樣信息帶來的估計性能惡化現(xiàn)象。通過理論分析和數(shù)值仿真得出，在信號帶寬較小時，優(yōu)化算法對系統(tǒng)性能提升較小，主要得益于加權(quán)平均；當(dāng)信號帶寬較大時，系統(tǒng)性能提升較明顯。 接著，在雙碼本波束成形系統(tǒng)中，分析由于IQ（in-phase/quadrature-phase）不平衡對波束成形系統(tǒng)帶來的影響，推導(dǎo)了在IQ不平衡情況下引起的碼本估計偏差以及雙碼本估計特性。為了消除IQ不平衡對碼本估計的影響，本文給出了一種完整的收發(fā)端IQ不平衡估計和補償算法，通過該算法能夠?qū)κ瞻l(fā)端IQ不平衡參數(shù)進(jìn)行分步估計補償，校正該因素引起的碼本估計誤差，從而提高系統(tǒng)性能。 然后，通常情況下為了減小上行反饋資源，對代表長時信道特征的碼本1采用較長的反饋周期，然而信道變化快慢具有時變性，如果采用固定的反饋周期會影響碼本1的準(zhǔn)確性，因此本文利用修正的非等間距灰色預(yù)測算法對信道方向指示（CDI）信息進(jìn)行預(yù)測，提出一種自適應(yīng)反饋策略。此外，由于碼本2估計實時性要求較高，過時的信道估計會嚴(yán)重降低碼本2的估計性能，因此本文利用卡爾曼預(yù)測算法對信道估計矩陣進(jìn)行預(yù)測，并且引入灰色預(yù)測算法對預(yù)測時用到的狀態(tài)轉(zhuǎn)移矩陣進(jìn)行修正，從而提高碼本2的估計性能。 最后，重點研究在大規(guī)模天線陣列（Massive MIMO）系統(tǒng)模型下，隨著天線數(shù)的增加，信道信息估計與反饋信息的復(fù)雜度成倍增加，同時，Massive MIMO系統(tǒng)中天線陣通常呈二維陣列分布，信道的方向性估計也需要擴(kuò)展成二維方向估計。針對上述特性，本文提出基于在二維信道方向估計下的雙碼本波束成形技術(shù)方案，包括雙碼本設(shè)計和雙碼本估計方案。最后通過仿真驗證該方案的有效性，并對多種碼本估計算法進(jìn)行對比仿真，從復(fù)雜度和性能兩方面進(jìn)行比較分析，其中似ESPRIT原理的碼本估計算法復(fù)雜度較低，且性能也接近改進(jìn)的二維雙碼本估計算法。
[Abstract]:Multi-antenna technology can realize spatial diversity and multiplexing by using spatial degree of freedom, and improve spectrum efficiency effectively. Beamforming is the main technology to realize space division multiple access in multi-antenna system. It can be divided into two types. One is long-time beamforming technology, which is applied to multi-antenna arrays with small spacing. The main principle is to utilize space. The strong correlation of the time channel and the interference principle of the wave produce a strong directional radiation pattern, which makes the main lobe of the radiation pattern self-adaptively point to the user's direction of arrival; the other is the short-term beam forming technology, whose main principle is based on the real-time channel impulse response, and uses the weak correlation between the antenna elements to construct the signal location. When the distance between the antenna is between half-wavelength and 10-fold wavelength, and the angle spread is small, it provides the conditions for combining long-time beam forming with short-time beam forming. The dual codebook technique, which combines the two codebooks, is introduced into LTE-Advanced (LTE-A) system. The two codebooks are used to indicate the long-time and short-time beamforming characteristics (codebook 1) respectively. In addition, the feedback periods of the two codebooks are different because of the different change periods of the long-time and short-time characteristics. Extending the feedback period of the long-time characteristic codebook to reduce the average feedback and improve the resource utilization. Although the dual codebook transmission scheme is defined in LTE-A, the estimation and feedback of the dual codebook are the key problems in the dual codebook technology.
This dissertation focuses on the research of double codebook beamforming technology, and puts forward a series of solutions through the related theoretical research. In general, the main research contents and contributions of this dissertation are as follows:
Firstly, the codebook design characteristics and the codebook estimation algorithm of the dual codebook beamforming technique are studied. In the Orthogonal Frequency Division Multiplexing (OFDM) system of broadband signals, frequency offset exists between different subcarriers, so the channel information on different carriers is used for independent codebook estimation. In this paper, a correlation estimation algorithm is proposed to obtain the equivalent channel information by focusing the channel information of different carriers on a reference frequency point according to the correlation between multiple carriers in a wideband signal. Carved signal-to-noise ratio (SNR) as a weighted average factor can overcome the deterioration of estimation performance caused by sampling information with large noise interference. Theoretical analysis and numerical simulation show that when the signal bandwidth is small, the optimization algorithm improves the system performance less, mainly because of the weighted average; when the signal bandwidth is large. The performance of the system is improved obviously.
Then, the influence of IQ (in-phase/quadrature-phase) imbalance on beamforming system is analyzed in a dual codebook beamforming system. The error of codebook estimation caused by IQ imbalance and the characteristics of double codebook estimation are deduced. In order to eliminate the influence of IQ imbalance on codebook estimation, a complete transceiver and receiver is given in this paper. The IQ unbalance estimation and compensation algorithm can compensate the IQ unbalance parameters of the transceiver step by step, correct the code book estimation error caused by this factor, and improve the system performance.
Then, in order to reduce the upstream feedback resources, a long feedback period is usually used for codebook 1, which represents the characteristics of long-term channel. However, the speed of channel change is time-varying. If a fixed feedback period is adopted, the accuracy of codebook 1 will be affected. Therefore, the modified unequal-interval grey prediction algorithm is used to indicate the direction of the channel. In addition, because of the high real-time requirement of codebook 2 estimation, outdated channel estimation will seriously degrade the performance of codebook 2 estimation. Therefore, this paper uses Kalman prediction algorithm to predict the channel estimation matrix, and introduces grey prediction algorithm to predict the state transition used in the prediction. The matrix is modified to improve the estimation performance of codebook 2.
Finally, with the increase of the number of antennas in Massive MIMO system, the complexity of channel information estimation and feedback information increases exponentially. At the same time, the antenna array in Massive MIMO system is usually two-dimensional array distribution, and the channel direction estimation also needs to be extended to two-dimensional direction estimation. In this paper, a two-codebook beamforming scheme based on two-dimensional channel direction estimation is proposed, including two-codebook design and two-codebook estimation scheme. Finally, the validity of the scheme is verified by simulation, and a variety of codebook estimation algorithms are compared and simulated. Comparisons and analyses are made from the aspects of complexity and performance. The codebook estimation algorithm has low complexity and the performance is close to the improved two dimensional codebook estimation algorithm.
【學(xué)位授予單位】：重慶大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：TN919.3
本文編號：2190017
[Abstract]:Multi-antenna technology can realize spatial diversity and multiplexing by using spatial degree of freedom, and improve spectrum efficiency effectively. Beamforming is the main technology to realize space division multiple access in multi-antenna system. It can be divided into two types. One is long-time beamforming technology, which is applied to multi-antenna arrays with small spacing. The main principle is to utilize space. The strong correlation of the time channel and the interference principle of the wave produce a strong directional radiation pattern, which makes the main lobe of the radiation pattern self-adaptively point to the user's direction of arrival; the other is the short-term beam forming technology, whose main principle is based on the real-time channel impulse response, and uses the weak correlation between the antenna elements to construct the signal location. When the distance between the antenna is between half-wavelength and 10-fold wavelength, and the angle spread is small, it provides the conditions for combining long-time beam forming with short-time beam forming. The dual codebook technique, which combines the two codebooks, is introduced into LTE-Advanced (LTE-A) system. The two codebooks are used to indicate the long-time and short-time beamforming characteristics (codebook 1) respectively. In addition, the feedback periods of the two codebooks are different because of the different change periods of the long-time and short-time characteristics. Extending the feedback period of the long-time characteristic codebook to reduce the average feedback and improve the resource utilization. Although the dual codebook transmission scheme is defined in LTE-A, the estimation and feedback of the dual codebook are the key problems in the dual codebook technology.
This dissertation focuses on the research of double codebook beamforming technology, and puts forward a series of solutions through the related theoretical research. In general, the main research contents and contributions of this dissertation are as follows:
Firstly, the codebook design characteristics and the codebook estimation algorithm of the dual codebook beamforming technique are studied. In the Orthogonal Frequency Division Multiplexing (OFDM) system of broadband signals, frequency offset exists between different subcarriers, so the channel information on different carriers is used for independent codebook estimation. In this paper, a correlation estimation algorithm is proposed to obtain the equivalent channel information by focusing the channel information of different carriers on a reference frequency point according to the correlation between multiple carriers in a wideband signal. Carved signal-to-noise ratio (SNR) as a weighted average factor can overcome the deterioration of estimation performance caused by sampling information with large noise interference. Theoretical analysis and numerical simulation show that when the signal bandwidth is small, the optimization algorithm improves the system performance less, mainly because of the weighted average; when the signal bandwidth is large. The performance of the system is improved obviously.
Then, the influence of IQ (in-phase/quadrature-phase) imbalance on beamforming system is analyzed in a dual codebook beamforming system. The error of codebook estimation caused by IQ imbalance and the characteristics of double codebook estimation are deduced. In order to eliminate the influence of IQ imbalance on codebook estimation, a complete transceiver and receiver is given in this paper. The IQ unbalance estimation and compensation algorithm can compensate the IQ unbalance parameters of the transceiver step by step, correct the code book estimation error caused by this factor, and improve the system performance.
Then, in order to reduce the upstream feedback resources, a long feedback period is usually used for codebook 1, which represents the characteristics of long-term channel. However, the speed of channel change is time-varying. If a fixed feedback period is adopted, the accuracy of codebook 1 will be affected. Therefore, the modified unequal-interval grey prediction algorithm is used to indicate the direction of the channel. In addition, because of the high real-time requirement of codebook 2 estimation, outdated channel estimation will seriously degrade the performance of codebook 2 estimation. Therefore, this paper uses Kalman prediction algorithm to predict the channel estimation matrix, and introduces grey prediction algorithm to predict the state transition used in the prediction. The matrix is modified to improve the estimation performance of codebook 2.
Finally, with the increase of the number of antennas in Massive MIMO system, the complexity of channel information estimation and feedback information increases exponentially. At the same time, the antenna array in Massive MIMO system is usually two-dimensional array distribution, and the channel direction estimation also needs to be extended to two-dimensional direction estimation. In this paper, a two-codebook beamforming scheme based on two-dimensional channel direction estimation is proposed, including two-codebook design and two-codebook estimation scheme. Finally, the validity of the scheme is verified by simulation, and a variety of codebook estimation algorithms are compared and simulated. Comparisons and analyses are made from the aspects of complexity and performance. The codebook estimation algorithm has low complexity and the performance is close to the improved two dimensional codebook estimation algorithm.
【學(xué)位授予單位】：重慶大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：TN919.3
【參考文獻(xiàn)】
相關(guān)期刊論文 前5條
1 徐以濤;王呈貴;;OFDM系統(tǒng)基于自適應(yīng)定階的MMSE信道估計[J];電子與信息學(xué)報;2007年01期
2 周兆經(jīng);蔡璐菁;;采用最大熵譜分析法計算天線方向圖[J];火控雷達(dá)技術(shù);1983年02期
3 仝蕊;申卯興;任俊亮;陳疆萍;;基于反三角函數(shù)變換的振蕩序列灰色預(yù)測模型研究[J];統(tǒng)計與決策;2010年24期
4 趙耀東;呂曉德;;基于特征值分解的穩(wěn)健波束形成技術(shù)[J];微波學(xué)報;2010年S1期
5 Yan Liang;Feng Shu;Yijin Zhang;Junhui Zhao;;High-performance compensation scheme for frequency-dependent IQ imbalances in OFDM transmitter and receiver[J];Journal of Systems Engineering and Electronics;2013年02期
本文編號：2190017
本文鏈接：http://sikaile.net/kejilunwen/wltx/2190017.html
最近更新
教材專著
