In this thesis, new statistical pro cedures for arra yand m ultic hannel signal pro cessing are dev elop ed. Nonlinear bayesian estimation of fmri bold signal under. Signal detection in nongaussian noise by a kurtosisbased probability density function model. The mmse can be regarded as a function of the signaltonoise ratio snr as well as a functional of the input distribution of the random variable to be estimated.
Filters with relatively many free filter coefficients are designed. Pdf detection of random signals in gaussian mixture noise. Signal detection in nongaussian noise springer texts in electrical engineering. The performance of the synthesized detection structure is assessed via monte carlo computer simulations, by assuming the generalized cauchy model for the univariate pdf of the noise. Frequency estimation of fm signals under nongaussian and. However,it requires the knowledge,but for a scale factor, of the noise correlation matrix. For this reason, the main goal of this dissertation is to develop statistical signal processing algorithms for the detection and modulation classi cation of signals in radio channels where the additive noise is. Detection of weak signals in nongaussian noise ning hsing lu on.
Mike wiegers department of electrical and computer engineering 3108 engineering hall 1701d platt st. Adaptive bayesian multiuser detection for synchronous cdma. This book contains a unified treatment of a class of problems of signal detection theory. Taking into account parameters of nongaussian distribution of random variables such as the moments of. Orthogonal polynomial approximation, signal detection and.
Signal detection in non gaussian noise by a kurtosisbased probability density function model. It is further shown in this paper that the mmse curve of a nongaussian input and that of a gaussian input cross each other. Dottorato di ricerca in ingegneria matematica, xvii ciclo alessandro foi anisotropic nonparametric image processing. It may enter the receiver through the antenna along with the desired signal or it may be generated within the receiver. Polynomial transformation method for nongaussian noise. Simulations for density click here conditions and probability. Gaussian pdf, the middleton class a pdf, and some such. Linear dynamics subject to thermal fluctuations and nongaussian. Pdf this paper has focused attention on the problem of optimizing signal detection in presence of additive independent. In this paper, we generate colored gaussian noise, colored nongaussian noise, and nongaussian noise types, these will then be added to singletone sinusoidal signals and fm signals. Pro auto system where pdf, fundamental analysis applied.
This thesis is concerned with detection of transient signal families and detectors in nonlinear static sensor systems. Detection and estimation of chirp signals in nongaussian. Newcomb absrractthe gramcharlier series representation of the noiseprobability density function is used to determine an optimum detector for signals in norrgauaaianbut neargaussian ngng noise. Distributed detection of a signal in generalized gaussian. Colored noise for signal detection 5660 is not adequately investigated in the context of stochastic resonance. The detection of a known deterministic signal in unknown nongaussian noise is a problem of great interest in many fields, such as communications and image processing. The obtained detection structure does not depend on the noise univariate probability density function pdf. Adaptive bayesian multiuser detection for synchronous cdma 2015 with and. The total noise variance under distribution 5 is given by 6 denote. It is shown that the mmse is concave in the input distribution at any given snr. The optimality of the proposed td is proved under the assumptions of white noise, small signal, and a large number of. A class this paper is based on a neural solution for signal detection in nongaussian noise, by d. It is wellknown that the mmse of a nongaussian input is dominated by the mmse of a gaussian input of the same variance.
In this paper, we propose a radial basis function rbf neural network for detecting a known signal in the presence of nongaussian and gaussian noise. A new moment quality criterion decision making is proposed based on a random process description using moments and a formation of polynomial decision rules. These processes encompass a large number of nongaussian distributions mentioned previously and include, of course, gaussian. Random signal detection in correlated nongaussian noise. Nongaussian process generation file exchange matlab. One of the most general and elegant nongaussian noise model is provided by the compoundgaussian process which includes the socalled spherically invariant random vectors sirvs. I joined caltech as an assistant professor of electrical engineering in the fall of 2014. Detection under gaussian and nongaussian environments and application to stap soutenue le 18 novembre 2011 devant les membres du jury.
Sengupta, estimation and detection for non gaussian processes. Signal detection and modulation classi cation in non. The problem of detecting the presence of a random signal embedded in additive correlated nongaussian noise modeled as a spherically invariant random. The majority of the signal detection and modulation classification algorithms available in the literature assume that the additive noise has a gaussian distribution. Detection in nongaussian noise university of washington. In this paper, we consider the mai mitigation problem in dscdma channels with nongaussian ambient noise.
Signal detection in nongaussian noise springerlink. Pdf signal detection is important in sonar and underwater digital communication. The principle diagram of multifrequency weak signal detection based on rfbsr under colored noise. For the most part the material developed here can be. T1 detection and estimation of chirp signals in nongaussian noise. New energy detector extensions with application in sound. N2 detection of chirps in non gaussian additive and multiplicative noise is explored via a novel cyclostationary approach. We will further assume that both xn and n have zero means and that they are statistically independent. A robust detector of known signal in nongaussian noise using. The authors discuss the need to provide a realistic model of a generic noise probability density function pdf, in order to optimize the signal detection in nongaussian environments. Here, the term represents the nominal ambient noise, and the term represents an impulsive component, with representing the probability that an impulse occurs. Robust signaltonoise ratio estimation based on waveform.
However, while this is a good model for thermal noise, various studies have shown that the noise experienced in most radio channels, due to a variety of manmade and natural. Pdf comparison of methodologies for signal detection in. The detection problems are treated within the framework of. However, underwater acoustic noise uwan influences the reliability of signal detection in applications, in which the noise is nonwhite and nongaussian. Pdf radar signal detection in nongaussian noise using. Optimum linear detectors, under the assumption of additive gaussian noise are suggested in 1. In the detection part, we analyze the behavior of the optimal detector by using the optimal decision regions framework. These comparison results demonstrate the potential capability of bistable systems for detecting weak signals in nongaussian noise. A neural solution for signal detection in nongaussian noise. Radar signal detection in nongaussian noise using rbf. Robust multiuser detection in nongaussian channels. Locally optimum bayes detection in nonadditive nongaussian noise. The authors of this paper study the synthesis of new models and methods for signal detection in additive correlated nongaussian noise. I completed my phd at princeton university in september 20.
Red i am attaching screens to be processed to noise. Optimum detection and signal design for channels with non. Viswanathan and arif ansari abstractthe problem of distributed detection of a signal in incom pletely specified noise is considered. Signal detection in correlated nongaussian noise using. Thomas ieee it 1975 gaussian noise shows few outliers impulsive noise is common in practice lightning, glitches, interference, pulses. Get your kindle here, or download a free kindle reading app.
For example, in watermark detection in discrete cosine transform dct domain, the signal is the watermark or a signature, which is usually known, while the dct coefficients of an image is the noise, whose. Study on multifrequency weak signal detection method. A robust detector of known signal in nongaussian noise. Dottorato di ricerca in ingegneria matematica, xvii. Taking into account parameters of nongaussian distribution of random variables such. Consequen tly, there has been gro wing in terest in estimation metho ds whic h w ork reliably in b oth gaussian and nongaussian noise. Pdf signal detection in nongaussian noise by a kurtosisbased. Detection of random signals in gaussian mixture noise. The corrupt speech signal zn is represented by the following equation. Schematic of the different detectors for known signal in nongaussian noise. Signal detection in nongaussian noise springer texts in electrical engineering kassam, saleem a. Signal processing 86 2006 34563465 noiseenhanced nonlinear detector to improve signal detection in nongaussian noise david rousseaua, g. Comparison of bistable systems and matched filters in non.
Bala natarajan electrical and computer engineering. In the present files, the method to transform of a gaussianprocess into a nongaussian one is based on the momentbased hermite transformation model mbhtm, and uses a cubic transform. Signal detection in nongaussian noise springerverlag, new york, 1988. Of course the focus is on noise which is not gaussian. The contents also form a bridge between the classical results of signal detection in gaussian noise and those of nonparametric and robust signal detection, which are not con sidered in this book. In the area of arra y signal pro cessing, the w ork concen trates on. Pdf distributed detection of a signal in generalized. The noise assumed belongs to the generalized gaussian family and the sensors in the distributed network employ the wilcoxon test. Previously, i worked as a postdoctoral researcher with prof.
Robust adaptive filtering algorithms for system identification and array signal processing in nongaussian environment by tiange shao a dissertation presented to the faculty of the graduate school of the missouri university of science and technology in partial ful. Radar signal detection in nongaussian noise using rbf neural network. Desai, which appeared in the proceedings of the fourth international. Optimum detection and signal design for channels with non but neargaussian additive noise adisai bodharamik, john b. Braham himed air force research laboratory rapporteur. Users may download and print one copy of any publication from the public portal for the. This is the detection of signals in addi tive noise which is not required to have. Particle filter tutorial file exchange matlab central. Pdf signal detection in nongaussian noise by a kurtosis. However, the computational complexity of ml detection is quite high, and therefore, effective nearoptimal multiuser detection techniques in nongaussian noise are needed. In this paper, we propose a thresholdsystembased detector td for detecting a known deterministic signal in independent nongaussian noise whose probability density function pdf is unknown but is symmetric and unimodal.
Kassam, optimum quantization for signal detection, ieee trans actions on. Distributed detection of a signal in generalized gaussian noise article pdf available in ieee transactions on acoustics speech and signal processing 375. To the best of our knowledge, there is no previous work on td for detecting an arbitrary signal in nongaussian noise with unknown pdf, which is the focus of this paper. Swedish postgraduate education leads to a doctors degree andor a licentiates degree.
Distributed detection of a signal in generalized gaussian noise r. I spent the spring of 2015 as a research fellow at simons institute for the theory of computing. I hold a bachelors degree from moscow institute of physics. Noiseenhanced nonlinear detector to improve signal. In this paper, a novel method is proposed that the multifrequency weak signal detection methods based on rfbsr under color noise background see fig. It has been described in 1, but i relies mainly on 2 for the implementation of the code. Receiver noise noise is the unwanted electromagnetic energy that interferes with the ability of the receiver to detect the wanted signal. Signal detection and modulation classification in non. Stochastic resonance with colored noise for neural signal detection.
1292 1187 903 933 160 133 1164 185 626 1 967 7 627 622 50 122 253 1113 446 572 1293 1567 236 1251 682 3 1265 97 1289 1219 439 1167 1211 994 1043 609 1002 1151 898