uu.seUppsala University Publications
Change search
Refine search result
1 - 29 of 29
Cite
Citation style
• apa
• ieee
• modern-language-association
• vancouver
• Other style
More styles
Language
• de-DE
• en-GB
• en-US
• fi-FI
• nn-NO
• nn-NB
• sv-SE
• Other locale
More languages
Output format
• html
• text
• asciidoc
• rtf
Rows per page
• 5
• 10
• 20
• 50
• 100
• 250
Sort
• Standard (Relevance)
• Author A-Ö
• Author Ö-A
• Title A-Ö
• Title Ö-A
• Publication type A-Ö
• Publication type Ö-A
• Issued (Oldest first)
• Created (Oldest first)
• Last updated (Oldest first)
• Disputation date (earliest first)
• Disputation date (latest first)
• Standard (Relevance)
• Author A-Ö
• Author Ö-A
• Title A-Ö
• Title Ö-A
• Publication type A-Ö
• Publication type Ö-A
• Issued (Oldest first)
• Created (Oldest first)
• Last updated (Oldest first)
• Disputation date (earliest first)
• Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
• 1.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Enhanced covariance matrix estimators in adaptive beamforming2007In: 2007 IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol II, Pts 1-3, 2007, p. 969-972Conference paper (Refereed)

In this paper a number of covariance matrix estimators suggested in the literature are compared in terms of their performance in the context of array signal processing. More specifically they are applied in adaptive beamforming which is known to be sensitive to errors in the covariance matrix estimate and where often only a limited amount of data is available for estimation. As many covariance matrix estimators have the form of diagonal loading or eigenvalue adjustments of the sample covariance matrix and as they sometimes offer robustness to array imperfections and finite sample error, they are compared to a recent robustified adaptive Capon beamforming (RCB) method which also has a diagonal loading interpretation. Some of the covariance estimators show a significant improvement over the sample covariance matrix and in some cases they match the performance of the RCB even when a priori knowledge, which is not available in practice, is used for choosing the user parameter of RCB.

Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Linear regression with a sparse parameter vector2006In: Conference Record of the 31st International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2006Conference paper (Refereed)

We consider linear regression under a model where the parameter vector is known to be sparse. Using a Bayesian framework, we derive a computationally efficient approximation to the minimum mean-square error (MMSE) estimate of the parameter vector. The performance of the so-obtained estimate is illustrated via numerical examples.

Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Linear Regression With a Sparse Parameter Vector2007In: IEEE Transactions on Signal Processing, ISSN 1053-587X, E-ISSN 1941-0476, Vol. 55, no 2, p. 451-460Article in journal (Refereed)

We consider linear regression under a model where the parameter vector is known to be sparse. Using a Bayesian framework, we derive the minimum mean-square error (MMSE) estimate of the parameter vector, and a computationally efficient approximation of it. We also derive an empirical-Bayesian version of the estimator, which does not need any a priori information, nor does it need the selection of any user parameters. As a byproduct, we obtain a powerful model (basis'') selection tool for sparse models. The performance and robustness of our new estimators are illustrated via numerical examples.

Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Adaptive equalization for frequency-selective channels of unknown length2004In: IEEE Global Telecommunications Conference (Globecom), vol. 2, (Dallas, Texas, USA), 2004Conference paper (Refereed)
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Adaptive equalization for frequency-selective channels of unknown length2005In: IEEE Transactions on Vehicular Technology, Vol. 54, no 2, p. 568-579Article in journal (Refereed)

This paper studies adaptive equalization for time-dispersive communication channels whose impulse responses have unknown lengths. This problem is important, because an adaptive equalizer designed for an incorrect channel length is suboptimal: it often estimates an unnecessarily large number of parameters. Some solutions to this problem exist (e.g., attempting to estimate the channel length'', and then switching between different equalizers); however, these are suboptimal owing to the difficulty of correctly identifying the channel length, and the risk associated with an incorrect estimation of this length. Indeed, to determine the channel length is effectively a model order selection problem, for which no optimal solution is known.

We propose a novel, systematic approach to the problem under study, which circumvents the estimation of the channel length. The key idea is to model the channel impulse response via a mixture Gaussian model, which has one component for each possible channel length. The parameters of the mixture model are estimated from a received pilot sequence. We derive the optimal receiver associated with this mixture model, along with some computationally efficient approximations of it. We also devise a receiver, consisting of a bank of soft-output Viterbi algorithms (SOVAs), that can deliver soft decisions. Via numerical simulations, we show that our new method can significantly outperform conventional adaptive Viterbi equalizers that use a fixed or an estimated channel length.

Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Subspace-based MRS data quatitation of multiplets using prior knowledge2004In: 12th meeting of the International Society for Magnetic Resonance in Medicine, 2004Conference paper (Refereed)
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Subspace-based MRS Data Quantitation of Multiplets using Prior Knowledge2004In: Journal of Magnetic Resonance, Vol. 168, no 1, p. 53-65Article in journal (Refereed)
• 8.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Parametric methods for frequency-selective MR spectroscopy - a review2004In: Journal of Magnetic Resonance, Vol. 168, no 2, p. 259-272Article in journal (Refereed)

Accurate quantitation of the spectral components in a pre-selected frequency band for magnetic resonance spectroscopy (MRS) signals is a frequently addressed problem in the MR community. One obvious application for such a frequency-selective technique is to lower the computational burden in situations when the measured data sequence contains too many samples to be processed using a standard full-spectrum method. Among the frequency-selective methods previously proposed in the literature, only a few possess the two features of primary concern: high robustness against interferences from out-of-band components and low computational complexity. In this survey paper we consider five spectral analysis methods which can be used for MRS signal parameter estimation in a selected frequency band. We re-derive the filter diagonalization method (FDM) in a new way that allows an easy comparison to the other methods presented. Then we introduce a frequency-selective version of the method of direction estimation (MODE) which has not been applied to MR-spectroscopy before. In addition, we present a filtering and decimation technique using a maximum phase bandpass FIR-filter and relate it to a similar ARMA-modeling approach known as SB-HOYWSVD (sub-band high-order Yule-Walker singular value decomposition). Finally, we study the numerical performances of these four methods and compare them to that of the recently introduced SELF-SVD (Singular Value Decomposition-based method usable in a SELected Frequency band) in several examples using simulated MR data, and discuss the benefits and disadvantages of each technique.

• 9.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Spectral analysis of multichannel MRS data2005In: Journal of Magnetic Resonance, Vol. 175, no 1, p. 79-91Article in journal (Refereed)

The use of phased-array receive coils is a well known technique to improve the image quality in magnetic resonance imaging (MRI) studies of, e.g., the human brain. It is common to incorporate proton (1H) magnetic resonance spectroscopy (MRS) experiments in these studies to quantify key metabolites in a region of interest. Detecting metabolites in vivo is often difficult, requiring extensive scans to achieve signal-to-noise ratios (SNR) that provide suitable diagnostic results. Combining the MR absorption spectra obtained from several receive coils is one possible approach to increase the SNR. Previous literature does not give a clear overview of the wide range of possible approaches that can be used to combine MRS data from multiple detector coils. In this paper we consider the multicoil MRS approach and introduce several signal processing tools to address the problem from different nonparametric, semiparametric and parametric perspectives, depending on the amount of available prior knowledge about the data. We present a numerical study of these tools using both simulated 1H MRS data, and experimental MRS data acquired from a 3T MR scanner.

• 10.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Frequency-selective SVD-based magnetic resonance spectroscopy with prior knowledge2004In: Conference Record of the 38th Asilomar Conference on Signals, Systems, and Computers, 2004Conference paper (Refereed)

We present a novel method for exploiting prior knowledge in magnetic resonance spectroscopy (MRS) based on the frequency-selective SELF-SVD method which was introduced in \cite{Sandgren}. More specifically, we use the common assumption that the magnetic resonance (MR) data is modeled by the superposition of a given number of exponentially damped sinusoids. As an application we consider the ATP (\textit{a}denosine \textit{t}ri\textit{p}hosphate) complex of an MRS signal and we use the fact that the dampings $\alpha_k$ and frequencies $\omega_k$ of the peaks of the ATP complex satisfy the following conditions: $\alpha_k = \alpha$ and $\omega_k = \omega + k\Delta$, where $\alpha$ and $\omega$ are unknown and $\Delta$ is known. Numerical examples mimicking $^{31}$P MRS data are included. The results show the superiority especially in speed of this new approach, which we will refer to as the FREEK (\textit{fre}quency-selective \textit{e}stimation with prior \textit{k}nowledge) method.

• 11.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology.
Magnetic Resonance Signal Processing2002Other (Other (popular scientific, debate etc.))

This thesis is concerned with the analysis of Nuclear Magnetic Resonance (NMR) signals.First a brief overview on the physics behind NMR is given and then the recent non-parametric adaptive filter bank methods 2D Capon and 2D CAPES are evaluated and compared with the two well known parametric methods HSVD and HTLS. This is done both with simulated and in-vivo data. HSVD and HTLS are proven to be superior to the non-parametric methods for the simulated data but the non-parametric methods have other advantages and should therefore still be considered as an alternative for in-vivo NMR analysis.The second part of the thesis presents an extension of the 2D CAPES method to allow for certain a priori information on the data, which is quite common in NMR applications. Experiments on simulated data shows that the inclusion of a priori information clearly improves the performance of the 2D CAPES method.

• 12.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Model Selection2004Licentiate thesis, monograph (Other scientific)

Before using a parametric model one has to be sure that it offers a reasonable description of the system to be modeled. If a bad model structure is employed, the obtained model will also be bad, no matter how good is the parameter estimation method. There exist many possible ways of validating candidate models. This thesis focuses on one of the most common ways, i.e., the use of information criteria. First, some common information criteria are presented, and in the later chapters, various extentions and implementations are shown. An important extention, which is advocated in the thesis, is the multi-model (or model averaging) approach to model selection. This multi-model approach consists of forming a weighted sum of several candidate models, which then can be used for inference.

• 13.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Automatic robust adaptive beamforming via ridge regression2008In: Signal Processing, ISSN 0165-1684, E-ISSN 1872-7557, Vol. 88, no 1, p. 33-49Article in journal (Refereed)
• 14.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Automatic robust adaptive beamforming via ridge regression2007In: 2007 IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol II, Pts 1-3, 2007, p. 965-968Conference paper (Refereed)

In this paper we derive a class of new parameter free robust adaptive beamformers using the generalized sidelobe canceler reparameterization of the Capon beamformer. In this parameterization the minimum variance beamformer is obtained as the solution of a linear least squares problem. In the case of an inaccurate steering vector and/or few data snapshots this marginally overdetermined system gives an ill fit causing signal cancellation in the standard minimum variance solution. By regularizing the problem using ridge regression techniques we get a whole class of robust adaptive beamformers, none of which requires the choice of a user parameter. We also propose a novel empirical Bayes-based ridge regression technique. The performance is compared to other robust adaptive beamformers.

• 15.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
An approach to sparse model selection and averaging2006In: Conference Record of the 2006 IEEE Instrumentation and Measurement Technology Conference (IMTC 2006): Sorrento, Italy 24-27 April 2006, 2006Conference paper (Refereed)

Parameter estimation when the true model structure is unknown is a commonly occurring task in measurement problems. In a sparse modeling scenario, the number of possible models grows exponentially with the total number of parameters. The full set of models therefore becomes computationally infeasible to handle. We propose a method, based on successive model reduction, for finding a sound and computationally feasible set of sparse linear regression models. Once this set of models has been found, standard model selection or model averaging techniques can be applied. We demonstrate the performance of our method by some numerical examples.

• 16.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Empirical Bayes linear regression with unknown model order2007In: 2007 IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol III, Pts 1-3, Proceedings, 2007, p. 773-776Conference paper (Refereed)

We study the maximum a posteriori probability model order selection algorithm for linear regression models, assuming Gaussian distributed noise and coefficient vectors. For the same data model, we also derive the minimum mean-square error coefficient vector estimate. The approaches are denoted BOSS (Bayesian Order Selection Strategy) and BPM (Bayesian Parameter estimation Method), respectively. Both BOSS and BPM require a priori knowledge on the distribution of the coefficients. However, under the assumption that the coefficient variance profile is smooth, we derive "empirical Bayesian" versions of our algorithms, which require little or no information from the user. We show in numerical examples that the estimators can outperform several classical methods, including the well-known AIC and BIC for order selection.

• 17.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Empirical Bayes linear regression with unknown model order2008In: Digital signal processing (Print), ISSN 1051-2004, E-ISSN 1095-4333, Vol. 18, no 2, p. 236-248Article in journal (Refereed)

We study maximum a posteriori probability model order selection for linear regression models, assuming Gaussian distributed noise and coefficient vectors. For the same data model, we also derive the minimum mean-square error coefficient vector estimate. The approaches are denoted BOSS (Bayesian order selection strategy) and BPM (Bayesian parameter estimation method), respectively. In their simplest form, both BOSS and BPM require a priori knowledge of the distribution of the coefficients. However, under the assumption that the coefficient variance profile is smooth, we derive "empirical Bayesian" versions of our algorithms which estimate the coefficient variance profile from the observations and thus require little or no information from the user. We show in numerical examples that the estimators can outperform several classical methods, including the well-known AICc and BIC for model order selection.

• 18.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Optimal Bayesian RAKE receiver for sparse channels2005In: Conference Record of the 39th Asilomar Conference on Signals, Systems, and Computers, 2005Conference paper (Refereed)

In this article we derive the optimal (soft) RAKE MRC (maximum ratio combiner) under a sparse channel model. In contrast to many previous approaches, we do not separate the problem of channel estimation and symbol detection. Instead we treat the problem of symbol detection given data and a training sequence together. We illustrate the performance of our approach by numerical examples.

• 19.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systemteknik.
Parameter Estimation and Order Selection for Linear Regression Problems2006In: 14th European Signal Processing Conference (EUSIPCO), 2006Conference paper (Refereed)

Parameter estimation and model order selection for linear regression models are two classical problems. In this article we derive the

minimum mean-square error (MMSE) parameter estimate for a linear regression model with unknown order. We call the so-obtained estimator the Bayesian Parameter estimation Method (BPM).

We also derive the model order selection rule which maximizes the probability of selecting the correct model. The rule is denoted BOSS---Bayesian Order Selection Strategy. The estimators have several advantages: They satisfy certain optimality criteria, they are non-asymptotic and they have low computational complexity. We also derive empirical Bayesian'' versions of BPM and BOSS, which do not require any prior knowledge nor do they need the choice of any user parameters''. We show that our estimators outperform several classical methods, including the AIC and BIC for order selection.

• 20.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
RAKE Receiver for Channels with a Sparse Impulse Response2007In: IEEE Transactions on Wireless Communications, ISSN 1536-1276, E-ISSN 1558-2248, Vol. 6, no 9, p. 3175-3180Article in journal (Refereed)

We derive the optimal receiver for RAKE diversity combining on channels with a sparse impulse response. The receiver is based on the Bayesian philosophy and thus it requires the knowledge of certain a priori parameters. However, we also derive an empirical Bayesian version of our receiver, which does not require any a priori knowledge, nor the choice of any user parameters. We show that both versions of our detector can outperform a classical training-based maximum-ratio-combining detector.

• 21.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
A model averaging approach for equalizing sparse communication channels2004In: Conference Record of the 38th Asilomar Conference on Signals, Systems, and Computers, 2004, p. 677-681Conference paper (Refereed)

In communication applications it is important for

equalization to have a good estimate of the channel over which

the signal was transfered.

In this paper we introduce the concept of

model averaging, where several channel estimates are used in a weighted

manner, to sparse communication channel estimation and equalization for

channels with intersymbol interference.

We show via numerical

examples that the bit-error-rate (BER) can be reduced by

our suggested method, compared to the BERs obtained using

channel estimation methods not assuming any sparsity of the channel.

• 22.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Estimation of semi-sparse radar range profiles2008In: Digital signal processing (Print), ISSN 1051-2004, E-ISSN 1095-4333, Vol. 18, no 4, p. 543-560Article in journal (Refereed)
• 23.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. AUTOMATIC CONTROL.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control.
Using prior knowledge in SVD-based NMR spectroscopy -- the ATP example2003In: Proceedings of the 7th International Symposium on Signal Processing and its Applications (ISSPA), Paris, France, July 1-4, 2003, p. 33-36Conference paper (Refereed)
• 24.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Cross-Validation Rules for Order Estimation2004In: Digital Signal Processing, Vol. 14, no 4, p. 355-371Article in journal (Refereed)

In this paper we revisit two cross-validation rules for order estimation which appear to have some advantages over commonly-used rules, yet they do not seem tohave received any significant attention so far. In an attempt to make other researchers aware of these rules, we review them here in an informal manner (with most of the technicalities in the original work being omitted) and also compare them with their closest competitors AIC, BIC (A and B Information Criteria), and MDL (Minimum Description Length) rules.

• 25.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Cyclic minimizers, majorization techniques, and the expectation-maximization algorithm: a refresher2004In: IEEE Signal Processing Magazine, Vol. 21, no 1, p. 112-114Article in journal (Refereed)
• 26.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Model-Order Selection: A review of information criterion rules2004In: IEEE Signal Processing Magazine, Vol. 21, no 4, p. 36-47Article in journal (Refereed)

The parametric (or model-based) methods of signal processing require often not only the estimation of a vector of real-valued parameters but also the selection of one or several integer-valued parameters that are equally important for the specification of a data model. Examples of these integer-valued parameters of the model include the orders of an autoregressive moving average model, the number of sinusoidal components in a sinusoids-in-noise signal, and the number of source signals impinging on a sensor array. In each of these cases, the integer-valued parameters determine the dimension of the parameter vector of the data model, and they must be estimated from the data.

• 27.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Multi-model approach to model selection2004In: Digital Signal Processing, Vol. 14, no 5, p. 399-412Article in journal (Refereed)

The single-model approach to model selection based on

information criteria, such as AIC or BIC, is omnipresent in the

signal processing literature.

However, any single-model approach picks up only

one model and hence misses the potentially significant information

associated with the other models fitted to the data. In our opinion

this is a drawback:

indeed, depending on the application, even the true model structure

(assuming that there was one) may not be the best choice for the intended

use of the model. The multi-model approach does not suffer

from such a problem:

using nothing more than the values of AIC or BIC it estimates

the a posteriori probabilities of each model under consideration

and then it goes on to use all fitted models in a weighted manner

according to their posterior likelihoods. We show via a numerical

study that the multi-model approach can outperform the single-model

approach in terms of statistical accuracy, without unduly increasing

the computational burden. The first goal of this paper is to advocate

the multi-model approach. A second goal is to introduce some guidelines

for numerically studying the performance of a model selection rule.

• 28.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
On information criteria and the generalized likelihood ratio test of model order selection2004In: IEEE Signal Processing Letters, Vol. 11, no 10, p. 794-797Article in journal (Refereed)

The Information Criterion (IC) rule and the Generalized Likelihood

Ratio Test (GLRT) have been usually considered to be two rather different

approaches to model order selection. However, we show here that a natural

implementation of the GLRT is in fact equivalent to the IC rule. A consequence

of this equivalence is that a specific IC rule, such as AIC or BIC, can be

viewed as a more direct way of implementing a GLRT with a specific

threshold. Another consequence of the equivalence, which is emphasized

herein, is a possibly original way of exploiting the information provided

by the local behavior of an IC for selecting the structure of

sparse models (the parameter vectors of which comprise many''

elements equal to zero).

• 29.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology. Uppsala University, Teknisk-naturvetenskapliga vetenskapsområdet, Mathematics and Computer Science, Department of Information Technology, Automatic control. Systems and Control.
Using prior knowledge in SVD-based parameter estimation for magnetic resonance spectroscopy--the ATP example2004In: IEEE Transactions on Biomedical Engineering, Vol. 51, no 9, p. 1568-1578Article in journal (Refereed)

We introduce the KNOB-SVD (knowledge based singular

value decomposition)

method for exploiting prior knowledge in MR

spectroscopy based on the singular value decomposition (SVD) of

the data matrix. More specifically we assume that the MR data

is well modeled by the superposition of a given number of exponentially

damped sinusoidal components, and that the dampings $\alpha_k$,

frequencies $\omega_k$ and complex amplitudes $\rho_k$

of some components satisfy the following relations:

$\alpha_k = \alpha$ ($\alpha = \textrm{unknown}$),

$\omega_k = \omega + (k-1) \Delta$ ($\omega = \textrm{unknown}$,

$\Delta = \textrm{known}$), and $\rho_k = c_k \rho$

($\rho = \textrm{unknown}$, $c_k = \textrm{known real constants}$).

which has one triple peak and two double peaks whose

dampings, frequencies and amplitudes may in some cases be known to

satisfy the above type of relations, is used as a vehicle for describing

our SVD-based method throughout the paper. By means of numerical

examples we show that our method provides more accurate parameter

estimates than a commonly-used general-purpose SVD-based method

and a previously suggested prior knowledge-based SVD method.

1 - 29 of 29
Cite
Citation style
• apa
• ieee
• modern-language-association
• vancouver
• Other style
More styles
Language
• de-DE
• en-GB
• en-US
• fi-FI
• nn-NO
• nn-NB
• sv-SE
• Other locale
More languages
Output format
• html
• text
• asciidoc
• rtf