A fast quasi-Newton adaptive filtering algorithm

Marshall, D. F.; Jenkins, W.K.

doi:10.1109/78.143437

Cited by 59 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case, the iteration number to converge with ε affects the complexity of the Kalman filter based on quasi-Newton methods. In [4]- [6], we can see the efficiency of the quasi-Newton methods for Kalman filter algorithm.…”

Section: Simulation Environment and Expected Resultsmentioning

confidence: 99%

Quasi-newton based Kalman filter channel estimation for OFDM systems

Nam

Kim

Kang

et al. 2010

2010 Digest of Technical Papers International Conference on Consumer Electronics (ICCE)

View full text Add to dashboard Cite

In this paper, Kalman filter method has been used for the estimate of time variant channels in OFDM systems. The estimation is based on Comb-type pilot in frequency domain. The channel coefficients that belong to the pilot subcarriers are estimated by Kalman filter. To reduce complexity, quasi-Newton methods are adapted for updating the gain of Kalman filter. Then, we can estimate channel parameter except for the inverse matrix which needs high complexity by using quasi-Newton methods.

show abstract

Section: Simulation Environment and Expected Resultsmentioning

confidence: 99%

Quasi-newton based Kalman filter channel estimation for OFDM systems

Nam

Kim

Kang

et al. 2010

2010 Digest of Technical Papers International Conference on Consumer Electronics (ICCE)

View full text Add to dashboard Cite

show abstract

“…To derive this relation, consider the a posteriori error vector (10) and substitute , which is obtained from (6). After some algebra, the desired relation is obtained as (11) Therefore, the a posteriori errors are identically zero if , which underscores the optimality of the least squares solution. This optimality may not be desirable in some applications, where large observation noise is present, as, for example, in acoustic echo cancellation in a hands-free telephone at low signal-to-ratio (SNR) levels [23].…”

Section: B Properties Of the Urls Algorithmmentioning

confidence: 99%

“…The fast conjugate gradient algorithm [10] is a suboptimal approximation to it. The suboptimal approximation of the RLS algorithms (the fast quasi Newton (FQN) algorithm [11]) updates the covariance matrix of the input in every instants by assuming that the covariance matrix changes slowly with time.…”

mentioning

confidence: 99%

Underdetermined-order recursive least-squares adaptive filtering: the concept and algorithms

Baykal

Constantinides²

1997

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

Abstract-The concept of underdetermined recursive leastsquares (URLS) adaptive filtering is introduced. In particular, the URLS algorithm is derived and shown to be a direct consequence of the principle of minimal disturbance. By exploiting the Hankel structure of the filter input matrix, the fast transversal filter (FTF) version of the URLS algorithm (URLS-FTF) is derived including sliding window and growing window types, which allow alteration of the order of the URLS algorithm (which is equivalent to the linear prediction order of the input) in real time. The computational complexity is reduced to O(N) + O(m), where N is the adaptive filter length, and m is the order of the URLS algorithm. In addition, the efficient URLS (EURLS) algorithm, which does not compute the filter coefficients explicitly, thereby significantly reducing the computational load, is presented. Some earlier adaptive algorithms such as the averaged LMS, filtered-X LMS, and fast conjugate gradient are shown to be suboptimal approximations of the URLS algorithm. Instrumental variable approximations are also discussed. The URLS algorithm has a whitening effect on the input signal, which provides immunity to the eigenvalue spread of the input signal correlation matrix. Although the algorithm is sensitive to observation noise, it has good tracking characteristics, and tradeoffs can be found by tuning the step size. The utility of the URLS algorithms, in its basic form and FTF realization, depends heavily on the practical applicability of the mth-order sliding window estimate of the covariance matrix and mth-order FTF relations. The feasibility of the URLS family in practical applications is demonstrated in the simulations, which include channel equalization and acoustic echo cancellation in hands-free terminals with real data sets.

show abstract

“…Typically, we choose q much smaller than N . As in [7], the filter vector ~( n ) is computed only every N time steps . Thus the complexity of Algorithm 3 averages out to O(q log N) operations per step n, where q is a small integer.…”

Section: Commentsmentioning

confidence: 99%

“…Comparisons of this proposed FFr-based RLS approach with the standard, O ( N 2 ) , RLS updating and downdating methods, and with fast, O ( N ) , RLS methods will be reported in Ng and Plemmons [9]. Other methods being considered include a frequency domain FFT-based frequency domain version of the quasi-Newton LMS adaptive filtering algorithm by Marshall and Jenkens [7], along with a hybrid LMS-RLS scheme that is also based on FFT computations [9]. Preliminary results indicate that these iterative methods can compete with direct methods in an adaptive signal processing environmen t .…”

Section: Commentsmentioning

confidence: 99%

FFT-based RLS in signal processing

Plemmons

1993

IEEE International Conference on Acoustics Speech and Signal Processing

View full text Add to dashboard Cite

This paper proposes a new algorithm for fast adaptiue filtering. The algorithm applies an FFT-based iterative method and uses sliding data windows inuoluing block updating and downdating computations. The method is stable and robust, and computes the tap weight filter vector in O(L log N ) operations, where the sliding window Toeplitz data matrix X is L-by-N. The complexity thus generally lies between those ofthe family of unstable but fast, O ( N ) , methods and the stable but slow, O ( N 2 ) , Cholesky factor updating methods.

show abstract

A fast quasi-Newton adaptive filtering algorithm

Cited by 59 publications

References 24 publications

Quasi-newton based Kalman filter channel estimation for OFDM systems

Quasi-newton based Kalman filter channel estimation for OFDM systems

Underdetermined-order recursive least-squares adaptive filtering: the concept and algorithms

FFT-based RLS in signal processing

Contact Info

Product

Resources

About