Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Fundamentals Of Statistical Signal Processing

Monte Carlo methods are revolutionizing the on-line analysis of data in fields as diverse as financial modeling, target tracking and computer vision. These methods, appearing under the names of bootstrap filters, condensation, optimal Monte Carlo filters, particle filters and survival of the fittest, have made it possible to solve numerically many complex, non-standard problems that were previously intractable. This book presents the first comprehensive treatment of these techniques, including convergence results and applications to tracking, guidance, automated target recognition, aircraft navigation, robot navigation, econometrics, financial modeling, neural networks, optimal control, optimal filtering, communications, reinforcement learning, signal enhancement, model averaging and selection, computer vision, semiconductor design, population biology, dynamic Bayesian networks, and time series analysis. This will be of great value to students, researchers and practitioners, who have some basic knowledge of probability. Arnaud Doucet received the Ph. D. degree from the University of Paris-XI Orsay in 1997. From 1998 to 2000, he conducted research at the Signal Processing Group of Cambridge University, UK. He is currently an assistant professor at the Department of Electrical Engineering of Melbourne University, Australia. His research interests include Bayesian statistics, dynamic models and Monte Carlo methods. Nando de Freitas obtained a Ph.D. degree in information engineering from Cambridge University in 1999. He is presently a research associate with the artificial intelligence group of the University of California at Berkeley. His main research interests are in Bayesian statistics and the application of on-line and batch Monte Carlo methods to machine learning. Neil Gordon obtained a Ph.D. in Statistics from Imperial College, University of London in 1993. He is with the Pattern and Information Processing group at the Defence Evaluation and Research Agency in the United Kingdom. His research interests are in time series, statistical data analysis, and pattern recognition with a particular emphasis on target tracking and missile guidance.

/pdf/sequential-monte-carlo-methods-in-practice-3kfxeebn8e.pdf

Sequential Monte Carlo methods in practice

物件導向軟體之架構(Object-Oriented Software Construction)探討

This paper presents an overview of progress in the area of multiple input multiple output (MIMO) space-time coded wireless systems. After some background on the research leading to the discovery of the enormous potential of MIMO wireless links, we highlight the different classes of techniques and algorithms proposed which attempt to realize the various benefits of MIMO including spatial multiplexing and space-time coding schemes. These algorithms are often derived and analyzed under ideal independent fading conditions. We present the state of the art in channel modeling and measurements, leading to a better understanding of actual MIMO gains. Finally, the paper addresses current questions regarding the integration of MIMO links in practical wireless systems and standards.

/pdf/from-theory-to-practice-an-overview-of-mimo-space-time-coded-431vesj02q.pdf

From theory to practice: an overview of MIMO space-time coded wireless systems

Recently, an analytical expression for the mean bit error probability of a binary modulated signal has been derived for a receiver performing maximum ratio combining of correlated diversity branches based on imperfect channel state information. We extend these results to the general case of colored additive noise. Furthermore, an alternative solution is derived which yields numerically stable results also in the case of the respective channel and interference covariance matrices having closely-spaced or multi-fold eigenvalues. Finally, the results are applied to a spatio-temporal W-CDMA RAKE receiver, where the effects of multiple-access-interference (MAI) and interpath interference (IPI), due to multipath propagation, on the mean bit error probability are investigated.

/pdf/maximum-ratio-combining-of-correlated-diversity-branches-32k8v3zlyy.pdf

Maximum ratio combining of correlated diversity branches with imperfect channel state information and colored noise

Architecture description languages (ADL) have been established to aid the design of application-specific instruction-set processors (ASIP). Their main contribution is the automatic generation of a software toolkit, including C compiler, assembler, linker, and instruction-set simulator. Hence, the challenge in the design of such ADLs is to unambiguously capture the architectural information required for the toolkit generation in a single model. This is particularly difficult for C compiler and simulator, as both require information about the instructions' semantics, however, while the C compiler needs to know what an instructions does, the simulator needs to know how. Existing ADLs solve this problem by either introducing redundancy or by limiting the language's flexibility.This paper presents a novel, mixed-level approach for ADL-based instruction-set description, which offers maximum flexibility while preventing from inconsistencies. Moreover, it enables capturing instruction- and cycle-accurate descriptions in a single model. The feasibility and design efficiency of our approach is demonstrated with a number of contemporary, real-world processor architectures.

A novel approach for flexible and consistent ADL-driven ASIP design

In this paper, we examine code-aided synchronization in the presence of carrier frequency uncertainties and phase noise. As code-aided synchronization can only achieve high estimation accuracy if the initial parameter offset is sufficiently small, we employ a data-aided coarse synchronization unit comprising a feed-forward frequency estimator and a LMMSE estimator for pre-compensating frequency offsets and phase noise. Furthermore, we make use of known iterative synchronization techniques for frequency and phase offsets. We here extend the concept of iterative synchronization to iterative phase noise compensation and we show analytically under which circumstances our approach can achieve a worthwhile improvement in terms of estimation accuracy.

/pdf/iterative-code-aided-phase-noise-synchronization-based-on-yvyi5swudl.pdf

Iterative code-aided phase noise synchronization based on the LMMSE criterion

A new adaptive timing error detector (TED) embedded in a code-tracking loop for RAKE reception of direct sequence-CDMA signals is presented. The loop consists of a digital coherent TED and a loop filter with lowpass characteristic. In a multipath fading environment, one such loop can be allocated for each finger in a RAKE receiver and the interference from adjacent paths can be mitigated by adaptively prefiltering the signal prior to the correlation process. The filter coefficients are computed online in order to minimize an interference cost function. Multipaths whose delays differ by as little as one chip duration become resolvable and can be tracked, resulting in significant performance gains of the overall system, especially if the channel delay spread is small. The tracking performance of the proposed loop is assessed by computer simulation.

Low complexity adaptive code tracking with improved multipath resolution for DS-CDMA communications over fading channels

The authors propose an algorithm that makes maximum a posteriori (MAP) equalization feasible for severe frequency-selective fading channels. In the algorithm the fully blown up intersymbol interference trellis is transformed into a set of parallel trellises with few states by means of tentative decisions. MAP equalization is performed on each subtrellis. The authors study the full MAP algorithm as well as reduced-complexity variants operating on parallel subtrellises. The proposed equalizer gives soft decisions which can be exploited by a successive decoder. The performance of the proposed algorithm was compared to M-algorithm equalization by extensive computer simulations. Results are presented for uncoded and trellis-coded signals for typical HF fading channels. The proposed soft output equalizer offers an additional soft decision coding gain of about 2.5 dB for short channel impulse responses and an improvement in bit error rate of two orders of magnitude for long channel impulse responses with a simple trellis code. >

Heinrich Meyr

Papers

Maximum ratio combining of correlated diversity branches with imperfect channel state information and colored noise

A novel approach for flexible and consistent ADL-driven ASIP design

Iterative code-aided phase noise synchronization based on the LMMSE criterion

Low complexity adaptive code tracking with improved multipath resolution for DS-CDMA communications over fading channels

Soft output MAP equalization and trellis coded modulation for severe frequency-selective fading channels