Gami Hiren

Propagator Method for joint time delay and frequency estimation

A novel method of estimating the differential delay of a sinusoidal signal is considered. The new method utilizes the propagator method (PM) which does not require the eigen-decomposition of the cross-spectral matrix (CSM) in estimating the delay of a signal received at two separated sensors. Moreover, the proposed method would generate estimates of the signal frequencies. Such estimates of time delay and frequencies are based on the observation and/or covariance matrices. Computer simulation is performed to validate the new procedure.

Efficient Structure-Based Carrier Offset Estimator for OFDM System

In this paper, we proposed a closed form solution for blind Orthogonal Frequency Division Multiplexing (OFDM) Carrier Frequency Offset (CFO) estimation employing the Rank- Revealing QR triangular factorization Method (RRQR). The advantage of using the RRQR it gives precious information about numerical rank and efficiently separates the signal space from the noise space. Furthermore, the RRQR does not involve the eigenvalue decomposition (EVD) or singular value decomposition (SVD) of co-variance matrix of the received signals and is a valuable tool in linear algebra to meet compatibility requirement of real time environment. Computer simulations shows the superior performance and much less processing time of RRQR compared with the method employing ESPRIT Algorithm Index-Terms: OFDM, Carrier Frequency Offset, DFT, Estimation, RRQR Method.

Rank Revealing QR Factorization for Jointly Time Delay and Frequency Estimation

The Rank-Revealing QR factorization (RRQR) is a valuable tool in numerical linear algebra because it provides accurate information about rank and numerical null-space. In this paper, we addressed the problem of estimating the time delay and the frequencies of noisy sinusoidal signals received at two spatially separated sensors using the well known RRQR, subspace decomposition technique. Although eigenvalue decomposition (EVD) of cross spectral matrix or Singular value decomposition SVD for the data matrix based techniques provide accurate estimation, they are hard to meet real time constraints due to computational load and cost. To explore compatibility with real time applications, we proposed a RRQR method in association with the well-known MUSIC/root-MUSIC algorithm to estimate unknown parameters without using any EVD or SVD. The simulation results verify that the proposed method provide better performance than the well known EVD or SVD based methods with less computational complexity. Index-Terms: Delay and Frequency estimation, Rank-Revealing

Time Delay Estimator for Frequency Hopping System using Rank-Revealing Triangular Factorization

In this paper, the multipath time delay estimation (TDE) problem for a slow frequency hopping (SFH) system using rank revealing QR factorization method (RRQR) is considered. It gives precious information about numerical rank and null space. By applying the RRQR in association with the well-known MUSIC algorithm we achieved a highly efficient estimator. The proposed methods would generate estimates of the unknown delay parameters. Such estimates are based on the observation and/or covariance matrices. Moreover, the RRQR does not require the eigenvalue decomposition (EVD) of the cross-spectral matrix (CSM) or singular value decomposition (SVD) of the data matrix of received signals. Computer simulations are also included to demonstrate the effectiveness of the proposed method.

Subspace based blind CFO estimation for OFDM by exploiting used carriers

Orthogonal Frequency Division Multiplexing (OFDM) is a promising technique to handle impairments of multipath channel. Alternatively, one of its major drawbacks is the drift in reference carrier, which is known as Carrier Frequency Offset (CFO). Hence, the CFO should be estimated and compensated with a sufficient accuracy. In this paper, a new algorithm for blind CFO-OFDM estimation is obtained by introducing the Propagator Method (PM) in conjunction with the well-known MUSIC based high resolution searching algorithm. Furthermore, the PM does not require the Eigenvalue Decomposition (EVD) or Singular Value Decomposition (SVD) of the covariance matrix of the received signals; Simulations are also included to demonstrate the effectiveness of the proposed method in comparison with other conventional methods.

High resolution Joint Time Delay and Frequency Estimation

Joint Time Delay and Frequency Estimation (JTDFE) problem of complex sinusoidal signals received at two separated sensors is an attractive problem that has been studied for many engineering applications. In this paper, the Rank-Revealing QR factorization is applied to the real data matrix obtained via the unitary transformation of the square Toeplitz complex data matrix. Then the MUSIC spectrum estimation function is used to estimate the frequencies. The time delay is estimated by applying RRQR to the complex data matrix. The unitary transformation from complex to real would reduce the processing time of frequency estimation by almost a factor of four, since the cost of complex manipulations is four times the real manipulations. Also RRQR is an important tool in numerical linear algebra because it provides accurate information about rank and numerical null space. The simulation results validate the performance of the proposed method.

RRQR based closed-form blind carrier offset estimator for OFDM systems

Carrier Frequency Offset (CFO) is one of the major drawbacks of the Multi Carrier Modulation (MCM) technique. Blind, semiblind and data-aided techniques have been introduced to estimate the CFO. In this paper, a new blind CFO estimator based on the Matrix Pencil (MP) method is proposed. A closed-form solution of this generalized eigenvalue problem using the MP method is solved by Rank Revealing QR factorization (RRQR). The RRQR reveals information about numerical rank and efficiently divides the space into signal and noise subspaces. Our algorithm is proficient to estimate CFO by not involving Eigenvalue Decomposition (EVD) or Singular Value Decomposition (SVD) of the spectral data matrix. The proposed method is characterized by the highly accurate CFO estimation especially with a very small number of OFDM frames. Computer simulations are included to show the dominant performance of our proposed method by comparing it with the ESPRIT algorithm, particularly at lower numbers of available data blocks.

Semiblind Multiuser MIMO Channel Estimators Using PM and RRQR Methods

Estimating the channel state information (CSI) of several transmitters that use orthogonal space-time block codes (OSTBC) to communicate with a single receiver is considered. Based on Rank Revealing QR (RRQR) factorization and Propagator method (PM), two new algorithms to estimate multiuser MIMO channels are proposed. The algorithm estimates the subspace spanned by the user channels use only a few training blocks to extract the users CSI from this subspace. Our both algorithms achieve better performance in comparison with reference LS based approach. Computer simulations are included to validate proposed methods.