“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

System Identification I

Thank you for reading statistical signal processing detection estimation and time series analysis. Maybe you have knowledge that, people have look hundreds times for their chosen novels like this statistical signal processing detection estimation and time series analysis, but end up in harmful downloads. Rather than reading a good book with a cup of tea in the afternoon, instead they juggled with some malicious bugs inside their laptop.

Statistical Signal Processing Detection Estimation And Time Series Analysis

This paper is devoted to the maximum likelihood estimation of multiple sources in the presence of unknown noise. With the spatial noise covariance modeled as a function of certain unknown parameters, e.g., an autoregressive (AR) model, a direct and systematic way is developed to find the exact maximum likelihood (ML) estimates of all parameters associated with the direction finding problem, including the direction-of-arrival (DOA) angles /spl Theta/, the noise parameters /spl alpha/, the signal covariance /spl Phi//sub s/, and the noise power /spl sigma//sup 2/. We show that the estimates of the linear part of the parameter set /spl Phi//sub s/ and /spl sigma//sup 2/ can be separated from the nonlinear parts /spl Theta/ and /spl alpha/. Thus, the estimates of /spl Phi//sub s/ and /spl sigma//sup 2/ become explicit functions of /spl Theta/ and /spl alpha/. This results in a significant reduction in the dimensionality of the nonlinear optimization problem. Asymptotic analysis is performed on the estimates of /spl Theta/ and /spl alpha/, and compact formulas are obtained for the Cramer-Rao bounds (CRB's). Finally, a Newton-type algorithm is designed to solve the nonlinear optimization problem, and simulations show that the asymptotic CRB agrees well with the results from Monte Carlo trials, even for small numbers of snapshots. >

Maximum likehood DOA estimation and asymptotic Cramer-Rao bounds for additive unknown colored noise

We propose a new subspace-based localization of far-field (FF) and near-field (NF) narrowband signals (LOFNS) without eigendecomposition impinging on a symmetrical uniform linear array, where the oblique projection operator is utilized to isolate the NF signals from the FF ones, and the procedures of computationally burdensome eigendecomposition are not required in the estimation of the NF and FF location parameters and the computation of oblique projection operator. As a measure against the impact of finite array data, an alternating iterative scheme is presented to improve the estimation accuracy of the oblique projection operator and, hence, that of the NF location parameters, where the “saturation behavior” encountered in most of localization methods is overcome. Furthermore, the statistical analysis of the proposed LOFNS is studied, and the asymptotic mean-squared-error expressions of the estimation errors are derived for the FF and NF location parameters. Finally, the effectiveness and the theoretical analysis of the proposed LOFNS are substantiated through numerical examples, and the simulation results demonstrate that the LOFNS provides remarkable and satisfactory estimation performance for both the NF and FF signals compared with some existing localization methods even with eigendecomposition.

Subspace-Based Localization of Far-Field and Near-Field Signals Without Eigendecomposition

Source localization for near-field narrowband signal is an important topic in array signal processing. Deep neural network (DNN) based methods are data-driven and free of pre-assumptions about data model and are expected to learn the intricate nonlinear structure in large data sets. This paper proposes a framework of DNN where a regression layer is utilized to address the problem of near-field source localization. Unlike previous studies in which DOA estimation is modeled as a classification problem and have a relatively low resolution, we exploit a regression model and aim to improve the estimation accuracy. In the training stage, we propose a novel form of feature representation to take full advantage of the convolution networks. In addition, the architecture of deep neural networks is well designed taking in to consideration the trade-off between the expression ability and under-training risks. The simulation results show that the proposed approach has a rather high validation accuracy with a high resolution, and also outperforms some conventional methods in adverse environments such as low signal to noise ratio (SNR) or small number of snapshots.

Deep Learning Based Localization of Near-Field Sources with Exact Spherical Wavefront Model

In this paper, we consider the source localization for a mixed near-field (NF) and far-field (FF) narrowband signals impinging on a uniform linear array (ULA) with the symmetrical geometric configuration. A computationally efficient direction-of-arrivals (DOAs) and range estimation method for the mixed NF and FF signals is proposed, where the DOAs of the NF and FF signals are estimated separately, and the computationally burdensome eigendecomposition is avoided. Comparing to some existent methods, the proposed method can separate the NF signals from the FF signals more efficiently, and consequently the estimation performance is improved. The effectiveness of the proposed method is verified though numerical examples.

A computationally efficient source localization method for a mixture of near-field and far-field narrowband signals

The invention discloses a mixed signal direction estimation method based on an even linear array, which can respectively estimate the directions of an incoherent signal and a coherent signal in two steps: 1. estimating the direction of arrive of the incoherent signal with a traditional subspace method; 2. removing the information of the incoherent signal from a receiving signal by the matrix difference technology; decorrelating by the submatrix average technology; and estimating the direction of arrive of the coherent signal. With the mixed signal direction estimation method, the defect that the number of solvable signals is far smaller than the number of sensors can be overcome. The performance of the method is improved by step-by-step estimation, and the direction of arrive can be quickly estimated by respectively estimating the directions of arrive of the coherent signal and the incoherent signal.

Mixed signal direction estimation method based on even linear array

The invention discloses a method for locating far field and near field mixed signal sources. Oblique projection operators are used, far field signals are effectively extracted from mixed signals, a sensor covariance matrix only including near field signals is obtained, and thus the one-dimensional azimuth information of the near field signals is calculated. In addition, a more effective estimation method is provided for the calculation of the oblique projection operators, the saturation phenomenon occurring in the locating of the near field signal source is eliminated by introducing an alternating iterative process, and the method for locating the far field and near field mixed signal sources lowers computing complexity and is simple and effective.

Method for locating far field and near field mixed signal sources

The invention discloses a partly-corrected symmetrical uniform linear array based narrow-band near-field signal source positioning method. The method includes: dividing an array into two overlapped and symmetrical subarrays, supposing that phase and amplitude gain of part of sensors of the array are given, and utilizing a spatial rotary invariant technology to estimate to obtain a direction-of-arrival angle of a near-field signal prior to utilizing the estimated direction-of-arrival angle and a multiple-subspace signal classification technology to solve distance information of the near-field signal. Calculated amount is lowered and estimated performance is improved from two aspects including firstly, a low-dimension cost function is constructed when the direction-of-arrival angle is estimated and secondly, a two-dimensional estimation problem is transformed into two one-dimensional estimation problems when the near-field direction-of-arrival angle and the distance are estimated.

Weiliang Zuo

Papers

Deep Learning Based Localization of Near-Field Sources with Exact Spherical Wavefront Model

A computationally efficient source localization method for a mixture of near-field and far-field narrowband signals

Mixed signal direction estimation method based on even linear array

Method for locating far field and near field mixed signal sources

Partly-corrected symmetrical uniform linear array based narrow-band near-field signal source positioning method