SIMO之 FIR系統的可適性盲蔽式通道鑑定：使用互參考最小平方和演算法

Blind Identification of SIMO FIR Systems : Using Mutual-Reference RLS Algorithm

白豐綱(Feng-Geng Pai)

指導教授 ： 黃 正 光

元智大學/工程學院/電機與資訊工程研究所/碩士(1998年)

摘要

近年來無線通訊科技一日千里，其盲蔽式通道鑑定及等化方法亦有許多突破性的發展，其中主要的一類是利用在時間或空間上對接收之數位調變訊號做過度取樣的單輸入多輸出 SIMO 之架構，這種性質為盲蔽式通道鑑定及等化技術開了一扇新的大門，並使其在無線通訊上的實用性大增。
在本論文中，所提出新的盲蔽式SIMO通道鑑定法，是基於利用上述之 SIMO架構且加入以相互參考性質的觀念為基礎，並使用推廣之限制型之RLS方法來鑑定出多個子通道路之脈衝響應，其讓吾人能使用任意的輸入訊號源和用最少的資料量，來求出通道參數；接著吾人提出在通道鑑定之前使用以階數遞迴最小平方和的方法和EDC階數判定法則來自動求得子通道的最大階數；其次在時變通道的環境下，吾人提出利用可適性記憶的RLS演算法來改進其通道追蹤的能力，使其可快速收斂及擁有高性能。最後我們利用上述鑑定出之通道參數，使用最小均方誤差(MMSE)法的有限長度之線性等化器及MMSE DFE的有限長度之非線性等化器二種形式，並使用它們來還原訊號；另外吾人提出雙向鑑定與等化的方法來降低資料接收的錯誤率，並以電腦模擬方式來分析驗證所提的方法的各種性能。

關鍵字

相互參考 ； 通道鑑定 ； 盲蔽式 ； 最小平方和 ； 單一輸入多輸出

並列摘要

Recently, the progress in wireless communications technology is even faster than before. Methods of new blind channel identification and equalization have lots of breakthrough too. It is noticed that the single-input and multiple output (SIMO) structure is resulted from over-sampling technique in time or space domain. Such a structure lend itself to identification and equalization and is promising to develop a new field and be very usefulness in the wireless communications.
In this thesis, we base on the use of the mutual-reference property of SIMO structure to develop a new channel identification. And we identify the impulse response of multiple subchannels by use of constrained RLS algorithm which can identify channel coefficient for any input statistics and using short data record. Next, we suggest to use the order-recursive least squares and EDC methods to detect the channel order before channel estimation. Third, we use adaptive memory RLS method to improve channel tracking ability in practical real-time application. Fourth, use the finite length equalizer based on the MMSE criteria, and the estimated channel response and the finite length DFE to restore data. Finally, a forward-backward channel identification and equalization method is proposed to decrease the error rate in receiver data. Computer simulation are included to demonstrate various features of our proposed approach.

並列關鍵字

Mutual-Reference ； channel Identification ； Blind ； LS ； SIMO

參考文獻

[1] J. K. Tugnait, "Identification of linear stochastic system via second and fourth-order cumulant matching," IEEE Trans. Inform. Theory, Vol. IT-33, pp. 393-407, May 1987.

[2] Y. Sato, "A method of self-recovering equalization for multilevel amplitud-modulation," IEEE Trans. Commun., Vol 23, pp. 679-682, June 1975.

[3] A. Benvenist, M. Goursat, "Blind equalizers," IEEE Trans. Commun., Vol 60, pp. 275-296, Feb. 1981.

[4] D. N. Godard, "Self-recovering equalization and carrier tracking in two dimensional data communication systems," IEEE Trans. Commun., Vol. COM-32, No. 8, pp. 871-883, Aug. 1984.

[5] L. Tong, G. Xu and T. Kailath, "A new approach to blind identification and equalization of multipath channel," Proc. 25th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, pp. 856-860, Nov. 1991.