We introduce the use of multidimensional logarithmic number system (MDLNS) as a generalization of the classical 1-D logarithmic number system (LNS) and analyze its use in DSP applications. The major drawback of the LN...
详细信息
We introduce the use of multidimensional logarithmic number system (MDLNS) as a generalization of the classical 1-D logarithmic number system (LNS) and analyze its use in DSP applications. The major drawback of the LNS is the requirement to use very large ROM arrays in implementing the additions and subtraction and it limits its use to low-precision applications. MDLNS allows exponential reduction of the size of the ROMs used without affecting the speed of the computational process: moreover, the calculations over different bases and digits are completely independent, which makes this particular representation perfectly suitable for massively parallel DSP architectures. The use of more than one base has at least two extra advantages. Firstly, the proposed architecture allows us to obtain the final result straightforwardly in binary form, thus, there is no need of the exponential amplifier, used in the known LNS architectures. Secondly, the second base can be optimized in accordance to the specific digital filter characteristics. This leads to dramatic reduction of the exponents used and, consequently, to large area savings. We offer many examples showing the computational advantages of the proposed approach.
This paper addresses the problem of designing signals for general group representations subject to constraints which are formulated as convex sets in the Hilbert space of the group states. In particular, the paper con...
详细信息
This paper addresses the problem of designing signals for general group representations subject to constraints which are formulated as convex sets in the Hilbert space of the group states. In particular, the paper considers irreducible representations in an infinite dimensional Hilbert space and derives an iterative procedure for proceeding from an arbitrary element of the Hilbert space to a state of the group subject to a priori imposed constraints with closed convex range. As examples, the paper focusses on narrowband and wideband radar ambiguity synthesis.
Current bilinear time-frequency representations apply a fixed kernel to smooth the Wigner distribution. However, the choice of a fixed kernel limits the class of signals that can be analyzed effectively. This paper pr...
详细信息
Current bilinear time-frequency representations apply a fixed kernel to smooth the Wigner distribution. However, the choice of a fixed kernel limits the class of signals that can be analyzed effectively. This paper presents optimality criteria for the design of signal-dependent kernels that suppress cross-components while passing as much auto-component energy as possible, irrespective of the form of the signal. A fast algorithm for the optimal kernel solution makes the procedure competitive computationally with fixed kernel methods. Examples demonstrate the superior performance of the optimal kernel for a frequency modulated signal.
Autocorrelation and spectra of linear random processes can be can be expressed in terms of cumulants and polyspectra, respectively. The insensitivity of the latter to additive Gaussian noise of unknown covariance, is ...
详细信息
Autocorrelation and spectra of linear random processes can be can be expressed in terms of cumulants and polyspectra, respectively. The insensitivity of the latter to additive Gaussian noise of unknown covariance, is exploited in this paper to develop spectral estimators of deterministic and linear non-Gaussian signals using polyspectra. In the time-domain, windowed projections of third-order cumulants are shown to yield consistent estimators of the autocorrelation sequence. Both batch and recursive algorithms are derived. In the frequency-domain, a Fourier-slice solution and a least-squares approach are described for performing spectral analysis through windowed bi-periodograms. Asymptotic variance expressions of the time- and frequency-domain estimators are also presented. Two-dimensional extensions are indicated, and potential applications are discussed. Simulations are provided to illustrate the performance of the proposed algorithms and compare them with conventional approaches.
The aim of this work is to contrast techniques used to estimate two instantaneous frequency parameters of the surface electromyographic (EMG) signal, the instantaneous median frequency and the instantaneous mean frequ...
详细信息
The aim of this work is to contrast techniques used to estimate two instantaneous frequency parameters of the surface electromyographic (EMG) signal, the instantaneous median frequency and the instantaneous mean frequency, based on their estimation error. Three methods are compared: Cohen class and Cohen-Posch class time-frequency representations are used to compute both the above-mentioned instantaneous frequency parameters, and a cross-time-frequency based technique is adopted to derive the instantaneous mean frequency. The results demonstrate that the algorithm based on Cohen-Posch class transformations leads to a standard deviation of the instantaneous frequency parameters that is smaller than that obtained using Cohen class representations. However, the cross-time-frequency estimation procedure for instantaneous mean frequency produced the smallest standard deviation compared to the other techniques. The algorithms based on Cohen class and Cohen-Posch class transformations often provided a lower bias than the cross-time-frequency based technique. This advantage was particularly evident when the instantaneous mean frequency varied non-linearly within the epochs used to derive the cross-time-frequency representation of the surface EMG signal.
Existing approaches to blind channel estimation and deconvolution (equalization) focus exclusively on channel or inverse-channel impulse response estimation. It is well-known that the quality of the deconvolved output...
详细信息
ISBN:
(纸本)0819416207
Existing approaches to blind channel estimation and deconvolution (equalization) focus exclusively on channel or inverse-channel impulse response estimation. It is well-known that the quality of the deconvolved output depends crucially upon the noise statistics also. Typically it is assumed that the noise is white and the signal-to-noise ratio is known. In this paper we remove these restrictions. Both the channel impulse response and the noise model are estimated from the higher-order (fourth, e.g.) cumulant function and the (second-order) correlation function of the received data via a least-squares cumulant/correlation matching criterion. It is assumed that the noise higher-order cumulant function vanishes (e.g., Gaussian noise, as is the case for digital communications). Consistency of the proposed approach is established under certain mild sufficient conditions. The approach is illustrated via simulation examples involving blind equalization of digital communications signals.
The scale dependent wavelet transform can be augmented by a rotation dependent version as well as other generalizations. Tomographic analysis and line segment transforms are special cases of rotation dependent wavelet...
详细信息
The scale dependent wavelet transform can be augmented by a rotation dependent version as well as other generalizations. Tomographic analysis and line segment transforms are special cases of rotation dependent wavelet analysis. Other cases suggested by biological analogy9 are a rotation dependent edge segment transform (using edge segments rather than line segments) and a binocular rotation dependent wavelet transform that introduces depth information into the reconstructed image. Applications to robot vision and synthetic aperture radar appear particularly promising.
The analysis of vehicle signals with methods derived from the theory of nonlinear dynamics is a potential tool to classify different vehicles. The nonlinear dynamical methodologies provide alternate system information...
详细信息
The analysis of vehicle signals with methods derived from the theory of nonlinear dynamics is a potential tool to classify different vehicles. The nonlinear dynamical methodologies provide alternate system information that the linear analysis tools have ignored. In order to observe the nonlinear dynamic phenomena more clearly, and estimate system invariants more robustly, we exploit the maximum power blind beamforming algorithm as a signal enhancement and noise reduction method when locations of a source and sensors are unknown. The dynamical behavior of an acoustic vehicle signal is studied with the use of correlation dimension D2 and Lyapunov exponents. To characterize the nonlinear dynamic behavior of the acoustic vehicle signal, Taken's embedded theory is used to form an attractor in phase space based on a single observed time series. The time series is obtained from the coherently enhanced output of a blind beamforming array. Then the Grassberger-Procaccia algorithm and Sano-Sawada method are exploited to compute the correlation dimension and Lyapunov exponents. In this paper, we also propose some efficient computational methods for evaluating these system invariants. Experimental classification results show that the maximum power blind beamforming processing improves the estimation of the invariants of the nonlinear dynamic system. Preliminary results show that the nonlinear dynamics is useful for classification applications.
This paper proposes a novel time-frequency maximum likelihood (t-f ML) method for direction-of-arrival (DOA) estimation for non-stationary signals, and compares this method with conventional maximum likelihood DOA est...
详细信息
This paper proposes a novel time-frequency maximum likelihood (t-f ML) method for direction-of-arrival (DOA) estimation for non-stationary signals, and compares this method with conventional maximum likelihood DOA estimation techniques. Time-frequency distributions localize the signal power in the time-frequency domain, and as such enhance the effective SNR, leading to improved DOA estimation. The localization of signals with different t-f signatures permits the division of the time-frequency domain into smaller regions, each contains fewer signals than those incident on the array. The reduction of the number of signals within different time-frequency regions not only reduces the required number of sensors, but also decreases the computational load in multi-dimensional optimizations. Compared to the recently proposed time-frequency MUSIC (t-f MUSIC), the proposed t-f ML method can be applied in coherent environments, without the need to perform any type of preprocessing that is subject to both array geometry and array aperture.
Focussing techniques have proven efficient in direction-of-arrival estimation of broadband signals. However, when used alone at high frequency operation, these techniques cannot accurately locate the sources, due to t...
详细信息
Focussing techniques have proven efficient in direction-of-arrival estimation of broadband signals. However, when used alone at high frequency operation, these techniques cannot accurately locate the sources, due to the numerous spurious peaks in the spectrum. Since the spurious peaks depend on the array manifold, their location varies from one focussing frequency to another. Improved jammer localization can, therefore, be achieved by focussing at different frequencies and then averaging the corresponding MUSIC spectra. The averaging smooths out the undesired peaks while boosting the common spectral peaks, allowing correct detection and location of the waveforms impinging on the array.
暂无评论