Jitka Pomenkova

Wireless device authentication through transmitter imperfections — measurement and classification

This paper is oriented in the area of wireless device identification through the analysis of emitted signals corrupted by radio frequency transmitter impairments. These impairments (such as power amplifier nonlinearity, IQ modulator imbalance, DC offset, phase noise, carrier frequency offset, shaping filter length and its shape) represent a unique characterization of any radio wireless transmitting device. We have performed a set of measurements on a sample of nine different transmitters (three baseboards plus three different front-ends) based on USRP software defined radio family in order to verify the potential of real applications of this authentication principle. After performing measurements in ideal conditions, we further proceeded in simulating a simple classifier based on gaussian mixture models and we briefly evaluated its performance in a real situation of a radio transmitting through air to the vector signal analyzer serving as the receiver. We also analyzed the influence of number of measurements on the classifier performance.

Comparative Study Of Time-Frequency Analysis Approaches With Application To Economic Indicators.

Presented paper deals with comparison of various methods for time-frequency representation of a signal with time-varying behavior. We choose methods such as wavelet analysis, multiple window method using Slepian sequences, time-frequency varying autoregressive process estimation and time-frequency Fourier transform representation (periodogram). We apply these methods first on the simple simulated artificial signal and we assess their performance. Then we proceed with application on the real data which is monthly data of the industry production index of European Union in the period 1990/M1-2011/M11. During the evaluation we focus on the results with respect to the time of global crisis. The results of the experiments are represented in the the graphical form and briefly discussed. INTRODUCTION The description of time-frequency structure of signal has wide range of usage. Its application can be seen in many scientific areas such as engineering (Xu et al., 2011), medicine (Xu et al., 1999), economy and many others. In last several years these techniques are in the front of economic researchers which analyze comovement of economic indicators. In this sense the papers of (Rua, 2010), (Yogo, 2008) or (Hallett and Richter, 2007) and many others were written. Estimation of spectrogram or scalogram of input signal or time series depends on used methods and their parameters. We investigate in this article four basic methods such as wavelet analysis (Jan, 2002), multiple window method using the slepian sequences (MWM) (Xu et al., 1999), time-frequency varying autoregressive (AR) process (Proakis et al., 2002) spectrum estimation and time-frequency Fourier transform estimation (periodogram) (Jan, 2002). On the basis of simulations on the artificial well-know signal we analyze behavior of each method and search for their advantages, disadvantages and recommendations for their usage. Consequently we compare obtained results with the aim to give recommendation for methods application. In order to practically demonstrate and evaluate the performance of the chosen methods we apply them to the analysis of the real data which is the monthly data of the industry production index of the European Union countries in the period 1990/M1-2011/M11. The paper is organized as follows: In the section Methodical Background we describe chosen methods of time-frequency analysis. Consequently, in the section Data, we briefly describe data used both for the simulation as well as for the practical application. After that, in the section Simulation, we show results of an application of chosen methods on simulated artificial data. The section Application presents results of real data analysis of the industry production index of the European Union and its brief economic interpretation. In both later sections results are graphically represented. The paper ends up with the conclusion and the list of used references. METHODICAL BACKGROUND Let us have a signal (a time series) y(n), n = 1, . . . , N . Under assumption that the time series contain a longterm trend, we can apply additive decomposition in the following form y(n) = g(n) + s(n) + c(n) + ε(n), n = 1, . . . , N, (1) where g(n) denotes a long-term trend, s(n) is the seasonal component, c(n) is the cyclical component and ε(n) is the irregular component (a random noise). Focusing on analysis of cyclical movements around its longterm trend it is necessary to remove the long-term trend applying some filtering methods. When the seasonally adjusted data are not available (in other words the analyzed series contains the seasonal component), the seasonality should be removed by applying some corresponding method. The spectrum of the signal (time series) y(n), n = 1, . . . , N can be written as a Fourier sum (Hamilton, 1994)

Wavelet significance testing with respect to GWN background: Monte Carlo simulation usage

The paper deals with significance testing of wavelet coefficients. We investigate the test of wavelet power spectrum with respect to the Gaussian white noise background spectrum from two perspectives of calculating significance level: with the use of χ2 distribution and with the use of Monte Carlo simulation. Our experiment is performed on a special kind of synthetic signal in which the frequency component is changing during time. We investigate the level of variance σ2 of the signal and the significance risk α. We describe the advantages and disadvantages of both approaches and formulate recommendations for using time-frequency testing.

Combination Of Time-Frequency Representations For Background Noise Suppression.

The aim of the paper is to propose approach for enhancement of time-frequency representation leading to the background noise suppression. The approach is based on combination of continuous wavelet analysis, timevarying autoregressive process and short time Fourier transform. By such combination we make the identification of important areas in the time-frequency representation easier. The proposed method is an alternative approach to significance tests which can be problematic in some cases. The performance of methods is presented on the gross domestic product of the United Kingdom and Group of 7. The results show that in the UK, oil crisis has a bigger impact compared to financial crisis, while from the perspectives of G7 countries, the impact of financial crisis was stronger. The obtained results can be also used for consequent econometric analysis which identify dependencies, relations, bilateral causalities or other economic aspects.

Confidence assessment of face recognition results

The paper deals with face recognition evaluation methodology. A complex performance comparison of face recognition algorithms has not been established for specific test databases. A basic fact discussed in the paper is limited statistical accuracy of the results. Face recognition results and reported performance increments need not be convincing due to test database properties. The confidence of results should be taken into account in order to reach convincing and reliable test conclusions. In order to express the confidence of the reached results, ROC curves enhanced by confidence intervals are proposed. Enhanced ROC curves are generally applicable without methodological requirements on a test databases.

The Modified Empirical Mode Decomposition Method For Analysing The Cyclical Behavior Of Time Series.

This paper is devoted to the analysis of time series using the Empirical Mode Decomposition (EMD) method. This method decomposes the analyzed time series into a small set of narrow-band components (modes) that fully represent the original time series. The modified EMD method that eliminates excessive changes of individual mode periods is proposed and evaluated on one example application of industrial production data. In contrast to other decomposition methods, like the singular value decomposition, the empirical mode decomposition can describe the time-variation of the period of individual components.

Evaluation of background noise for significance level identification

The paper deals with the identification of the significance level for testing the time-frequency transform of the data. The usual procedure of time-frequency significance testing is based on the knowledge of background spectrum. Very often, we have certain expectations about the character the background noise (White noise, Red noise, etc.). Our paper deals with the case when the character of the noise is unknown and may not be Gaussian despite our assumptions. Thus, we propose how to identify our own critical values for testing time-frequency transform significance with respect to the data character. We compare our findings with the critical quantile of χ22.

Learning of a robusted nearest neighbor classifier using multiple training data

This paper deals with the application of face recognition in surveillance CCTV systems and effective usage of so called recognition clues. These clues are enrollment of multiple training face images and their usages in classifier training and real-time management of template database. A survey on classifiers from perspective of practical application is given resulting in the defense of nearest neighbor based classifiers. They are competitive with state of the art classifiers and are suitable for practical application. Template creation using multiple training face images and enhancement of NN-based classifier performance is achieved by novel approach. It consist of quantile interval method for template creation and robusted NN-based classifier using spatial templates with soft boundaries. We evaluate proposed recognition framework on highly representative IFaViD dataset. Proposed framework outperforms state of the art approaches.

Optimization of time-frequency curve description via kernel smoothing

Presented paper deals with trend modeling of spectral coefficients represents material properties at very rapid load. We focus on identification and optimization of the curve describing dependence between frequencies and time in spectrogram. The spectrogram is firstly processed with the aim to specify significant spectral coefficients. Consequently for such coefficient we apply non-parametric kernel smoothing. We investigate the type of estimator, kernel and the bandwidth. Input data are given from aluminum metal plate acceleration by detonation products of brisant high explosive obtained using Photonic Doppler Velocimetry in time on the basis of frequency response.