應用於快速衰落多輸入多輸出正交分頻多工系統使用內插式QR分解之可擴充式多輸入多輸出檢測器

Scalable MIMO Detector with Interpolation-based QR Decomposition for Fast-Fading MIMO-OFDM Systems

邱柏霖(Po-Lin Chiu)

指導教授 ： 周復芳 ； 黃元豪

國立交通大學/電機學院/電信工程系所/博士(2012年)

https://doi.org/10.6842/NCTU.2012.00177

摘要

用來檢測所傳送資料的多輸入多輸出檢測器在多輸入多輸出接收機中扮演了重要的角色，特別是基於QR分解的多輸入多輸出檢測器可用來避免複雜的反矩陣運算已被廣泛地認為是一種具有潛力的方法。然而，在多載波系統中，每個次載波都需QR分解使得運算複雜度大幅地增加，此外，QR前置處理與後續多輸入多輸出檢測通常被獨立分開來研究而造成運作上的限制。在本論文中，我們不僅研究低複雜度基於QR分解的多輸入多輸出檢測器，而且考慮QR前置處理與多輸入多輸出檢測器的整合設計。

根據結合頻域通道估測與QR分解運算在多輸入多輸出正交分頻多工系統中的原始研究，我們提出一個修改的演算法，此內插式的QR分解擁有可擴充的特性，在可變秩的多輸入多輸出機制中更可節省功率消耗。同時，對於所提出的QR分解的硬體設計應用在更高維度的多輸入多輸出系統中，我們也發展出一套通式及運算時間的安排方法。實驗數據顯示在與硬體成本取得平衡的情況下，可達到更高的資料吞吐量。

在QR分解之後，根據基於QR分解的連續干擾消除檢測演算法，我們提出使用多候選選擇的基於QR分解的連續干擾消除檢測，透過修改限制函式使其具有可變範圍以改善檢測效能。換句話說，這個方法有多種的候選組態，可在複雜度與效能間作靈活的取捨。另外，與其他樹搜尋檢測機制不同的是，所提出的機制不需要排序運算來決定每層的存活節點，因此除了擁有多候選的好處外，多候選選擇基於QR分解的連續干擾消除檢測所擁有的不需排序特性不僅降低硬體成本而且能縮短處理時間。

本論文開發一個應用於空間多工多輸入多輸出正交分頻多工系統之具有QR分解與通道內插的可配置多輸入多輸出檢測器。對於硬體實作，管線式處理單元及遞迴式處理單元是兩種分別針對高吞吐量與低成本應用的架構，然而這些架構通常限制運作環境僅適用於緩慢平坦衰落通道。因此，根據我們提出的QR分解與多輸入多輸出檢測演算法，本論文提出針對QR前置處理與多輸入多輸出檢測的一個混合遞迴式與管線式處理單元的架構，用以消除兩處理器間的介面緩衝記憶體及等待時間，同時也運算通道內插。此架構支援第三代合作計劃長期演進技術(3GPP-LTE)系統中的2x2、2x4及4x4多輸入多輸出組態以及64QAM、16QAM與QPSK調變。本論文所提出的處理器對於一個正交分頻多工次載波可執行一次QR分解與一次多輸入多輸出檢測，是第一個應用於頻率選擇性快速衰落多輸入多輸出正交分頻多工系統中支援實時QR分解與多輸入多輸出檢測的多輸入多輸出檢測器。

關鍵字

多輸入多輸出 ； 內插 ； QR分解 ； 次載波群組排序 ； 多候選選擇 ； 多輸入多輸出檢測器

並列摘要

Multiple-input multiple-output (MIMO) detector used for detecting the transmitted data streams plays an important role in MIMO receivers. Especially, the QR decomposition (QRD) based MIMO detection is widely recognized as a promising method to avoid the complicated pseudo-inverse matrix calculations. However, in the multi-carrier system, the tone-by-tone QRD increases the computational complexity dramatically. Besides, the QR preprocessing and MIMO detector are usually investigated independently so that there are some operating assumptions. In this dissertation, we aim to not only the low complexity QR-based MIMO detector but also the joint QR preprocessing and MIMO detector designs.

Based on the original research that integrates the calculations of the frequency-domain channel estimation and the QRD for the MIMO orthogonal frequency division multiplexing (OFDM) system, we propose a modified algorithm that possesses a scalable property to save the power consumption for interpolation-based QRD (IQRD) in the variable-rank MIMO scheme. Moreover, we also develop the general equations and a timing scheduling method for the hardware design of the proposed QRD for the higher-dimension MIMO system. The experimental results show that much higher data throughput is achieved at balancing the hardware cost.

After the QRD, based on the QR successive interference cancellation (SIC) detection algorithm, we propose the QR-SIC detection with multiple candidates selection (MCS) to improve the detection performance by modifying slicing function with a variable range. In other words, this approach has various candidate configurations and forms a flexible trade-off between complexity and performance. Furthermore, in contrast to other tree-search-based detection schemes, the proposed scheme does not need sorting operations to determine the survival nodes in each layer. Besides possessing the benefit of multiple candidates, the MCS-QRSIC detection with the non-sorted property not only reduces the hardware cost but also shortens the processing time.

A configurable MIMO detector with QRD and channel interpolation for spatial multiplexing (SM) MIMO-OFDM systems is developed in this dissertation. For hardware implementation, pipelined processing element (PE) and iterative PE are two basic architectures for high-throughput and low-cost applications, respectively. These architectures usually limit the operating condition to only flat-fading and slow-fading channel. Therefore, based on our proposed QRD and MIMO detection algorithms, this dissertation proposes a mixed iterative and pipelined PE architecture for QR preprocessing and MIMO detection to eliminate interface memory buffer and idling time between two processors. The channel interpolation is also performed at the same time. The architecture supports 2x2, 2x4 and 4x4 MIMO configurations and 64QAM, 16QAM and QPSK for 3rd Generation Partnership Project Long Term Evolution (3GPP-LTE) system. The proposed processor, which performs one QRD and one MIMO detection for one OFDM subcarrier, is the first MIMO detector that supports real-time QRD and MIMO detection for frequency-selective fast-fading MIMO-OFDM systems.

並列關鍵字

Multiple-Input Multiple-Output (MIMO) ； Interpolation ； QR Decomposition ； Subcarrier-Grouped-Ordering ； Multiple Candidates Selection ； MIMO Detector

參考文獻

[1] G. J. Foschini, “Layered space-time architecture for wireless communication in a fading envirenment when using multiple antennas,” Bell Laboratories Technical Journal, vol. 1, pp. 41–59, Autumn 1996.

[3] “Part 16: Air Interface for Fixed and Mobile Broadband Wireless Access Systems–Advanced Air Interface,” P802.16m D5, IEEE, New York, NY, Apr. 2010.

[5] E. Agrell, T. Eriksson, A. Vardy, and K. Zeger, “Closest point search in lattices,” IEEE Trans. Inf. Theory, vol. 48, pp. 2201–2214, Aug. 2002.

[6] K. W. Wong, C. Y. Tsui, R. S. K. Cheng, and W. H. Mow, “A VLSI architecture of a K-best lattice decoding algorithm for MIMO channels,” in Proc. IEEE Int. Symp. Circuits and Systems, vol. 3, pp. 273–276, May 2002.

[7] P. W. Wolniansky, G. J. Foschini, G. D. Golden, and R. A. Valenzuela, “V-BLAST: an architecture for realizing very high data rates over the rich-scattering wireless channel,” in Proc. IEEE Int. Symp. Signals, Systems and Electronics, pp. 295–300, Sep. 1998.