EFFICIENT EQUALIZATION OF NONLINEAR MULTICHANNEL SYSTEMS. PhD thesis, King Fahd University of Petroleum and Minerals.

PDF (PhD Dissertation) Abdulmajid_Lawal PhD Dissertation.pdf Restricted to Repository staff only until 5 July 2022. Download (1MB)

Arabic Abstract

هي التعرف على نظام معين و اعادة نمذجته. نتناول الانظمه احادية البعد و هي التي تعالج اشارات مثل اشارات الاتصالات و اثنائية البعد مثل الصور. في هذهالدراسه نتاول النظمه متعددة المدلات و متعددة المخرجات. لقد قمنا بتطوير خورازميات محسنه بحيث تتناول احد اهم المعلومات الجانبيه الموجدوه بالنظمه و هي النظام انائي في ما يدعى\LR{Toeplitz Structure} . قمنا باستخدام الفضاء المرافق للوسط الناقل و الفضاء االممثل للاشارات المرسله لتطوير و اقتراح مجموعه من الخوارزميات. و باستخدام برامج المحاكاه قمنا بفحص هذه الخوارزميات و مقارنتها مع خوارزميات اخرى و قد اظهرت كفائتها. من ناحيه اخرى اخرى قمنا باقتراح مجموعه من الخوارزميات التي تعالج الانظمه الغير خطيه والتي قد تظهر لعدة اسباب. ايظا استخدما التركيب الموجود داخل الانظمه لتحسين اعادة الحساب و النمذجه للانظمه. و في النهايه تطرقنا لاقتراح خوارزميه للتعامل مع الانظمه ثنائية البعد.

English Abstract

{This work addresses the problem of blind system/signal identification and estimation in the context of 1-dimensional (1D) and 2-dimensional (2D) single input multiple outputs (SIMO) and multiple-input multiple-output (MIMO) system. The proposed methods exploit the inherent Toeplitz structure that exists in most communication systems. In the first part, a SIMO structural channel subspace (SCS) that exploits the Toeplitz structure of a convolutive model to identify a channel matrix is developed. The method utilizes the channel subspace and performs well especially under ill-channel conditions, i.e, poor channel diversity. In this method, the channel is estimated first, the information of the channel is then utilized to estimate the signal. However, in practical systems, the use of channels for signal estimation may lead to error propagation due to the inversion of the channels. This motivation leads to the development of another novel algorithm termed the structured signal subspace (SSS) method that directly estimates the signal with no required knowledge of the channel. The algorithm utilizes the signal subspace matrix to estimate the signal. To produce a more robust algorithm, we developed a SIMO bilinear structure channel subspace method (B-SCS) which uses the information from both signal and channel subspace to estimate the channel. The SIMO SCS, SIMO SSS, and B-SCS have been extended to solve MIMO signal/channel estimation problems. On the other hand, the influence of some side information on the channel is investigated. The work also extends the aforementioned methods to an adaptive case, where the channel is estimated adaptively, the choice of an adaptive system is to ensure the utility of series of algorithms in real-time systems. Furthermore, we devised a novel approach to investigates nonlinear systems equalization. To solve the identification and estimation problem in 2D systems, we introduced a novel approach for 2D blind multichannel identification using the helix transform in conjunction with the cross relation (CR) method. The helix transform is used to convert the 2D convolution of a 2D signal and channels into 1D convolution. The CR method, known for its simplicity, efficiency, and low computational cost, is then adapted and used to estimate the unknown channel coefficients. The numerical simulations of the devised methods are appealing and show that the structure-based algorithms outperform the state-of-the-art methods in different cases.

Item Type:	Thesis (PhD)
Subjects:	Engineering Electrical
Department:	College of Engineering and Physics > Electrical Engineering
Committee Advisor:	Zerguine, Azzedine
Committee Co-Advisor:	Karim, Abed-Meraim
Committee Members:	Muqaibel, ALi and Zidouri1, Abdelmalek and Alghadhban, Samir
Depositing User:	ABDULMAJID LAWAL (g201403800)
Date Deposited:	29 Jul 2021 09:04
Last Modified:	29 Jul 2021 09:04
URI:	http://eprints.kfupm.edu.sa/id/eprint/141923