Broadband Near-Field Range and Bearing Estimation Based on Fourth-Order Cumulants

Mismatch of Near-Field Bearing-Range Spatial Geometry in Source-Localization by a Uniform Linear Array

“…However, the th-order mismatched measurement-model of (21) gives (39) Hence, (38) becomes (40) The last equality holds by the definition of and by (39). Similarly,…”

“…No comprehensive mathematical analysis exists on how such model-mismatch degrades near-field bearing-range estimation: Much of the uniform linear array bearing-range estimators are assessed via limited Monte Carlo simulations, as in [3], [5], [7], [9], [11], [13], [15], [17]- [26], [28], [29], [31]- [34], [36], [39], [40]. Also, the performance analysis and/or Cramer-Rao bound in [10], [12], [16], [19], [20] are derived as if the mismatched measurement-model were the true measurement-model; hence, such analysis has not accounted for any model-mismatch effect.…”

Near‐field wideband source localisation from the sparse recovery perspective via the spatial‐only modelling of array data

“…Note that ∥ p r ∥ 1 = ∑ k = 1 K p k is equivalent to ∥ p ∥ 1 in the ideal situation, so the parameter ε is exactly suitable for (18). By working out (18) through SOC programming, the source localisation can be finally determined.…”

“…As far as near-field wideband cases are concerned, the traditional way has been to split the array outputs into several narrowband components by passing them through a bank of narrowband filters and then combine the results of acquired estimates at multiple frequencies independently. Or alternatively, exploiting the focusing technique to obtain the source localisation information integrated into these bands by aligning the signal subspaces of the narrowband components [16][17][18][19]. There are two apparent inadequacies in these methods.…”

Amplitude envelope analysis for feature extraction of direct-driven wind turbine bearing failure

“…Condition monitoring and fault diagnosis techniques based on Fourier spectral analysis of vibration signals have been widely used in industrial practice. In recent years, many new methods in signal analysis field are constantly used for bearing vibration monitoring and fault diagnosis areas, such as wavelet analysis [2], time-frequency analysis [3], high order spectrum analysis [1,5], etc. Moreover, many research achievements in pattern recognition field (such as support vector machine [4], etc.)…”