Browsing by Author "Zakayo, Ndiku Morris"

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    Acoustical Direction Finding using a Bayesian Regularized Multilayer Perceptron Artificial Neural Networks on a Tri-Axial Velocity Sensor
    (IJMEC, 2020-01) Nnonyelu, Chibuzo Joseph; Zakayo, Ndiku Morris
    A two-dimensional direction-of-arrival estimation scheme based on Bayesian-regularized (BR) Multilayer Perceptron (MLP) artificial neural network (ANN) is developed around a unit acoustic vector sensor (AVS). The AVS basically consists of three collocated and orthogonally oriented velocity sensors, hence, senses acoustic waves in the three Cartesian directions while offering portability in size and simplicity in its array manifold. It is shown that the Bayesian regularized Multilayered Perceptron neural network performs well in terms of estimation’s root-mean-square error even when tested with data of different signal-to-noise ratio (SNR) after training. This is useful as it accounts for unexpected changes of received data SNR during field operation. The proposed system is ideal for applications in mobile systems such as robots for search-and-rescue operations or soldiers in the battle field to estimate the source of a sniper fire.
  • No Thumbnail Available
    Item
    A Centrosymmetric Array Comprising a Horizontal Uniform Circular Subarray and a Vertical Uniform Linear Subarray—Its Design in Reference to Its Direction-Finding Cramér–Rao Bound.
    (IEEE, 2021-06) Lin, Yang; Wong, Kainam Thomas; Zakayo, Ndiku Morris
    Azimuthal centrosymmetry in an array grid is typically associated with arrays that are circular, concentric, cylindrical, spherical, or hemispherical. However, a recently proposed alternative combines an azimuthal circular array with a linear vertical array. For this elegantly simple new array grid's use in the direction-of-arrival estimation, this article advances array-design insights to meet a given estimation-precision threshold, by examining the tradeoff between the azimuth-angle Cramér-Rao bound vis-a-vis the polar-angle Cramér-Rao bound in a proposed two-step design procedure.
  • No Thumbnail Available
    Item
    Comparing the “Rim” Versus the “Filled” Rectangular Array Grids—Their Direction-Finding Cramér-Rao Bounds
    (IEEE, 2018) Zakayo, Ndiku Morris; Wong, Kainam Thomas
    A rectangular array of sensors is often used in direction finding, due to the geometric regularity in its spatial rectangular grid. The sensor positions may be confined to the rectangle's perimeter (as in a “rim” array), or may span over the rectangle's entire interior as well (as in a “filled” array). This paper compares these two array grids by their precision in direction finding, by pioneering Cramér-Rao bound expressions for both array grids above, in closed forms and explicitly in terms of the array parameters.
  • No Thumbnail Available
    Item
    Hybrid Cramer-Rao bound for near-field source localization using a spatially spread acoustic vector sensor
    (Acoustical Society of America, 2019-03) Zakayo, Ndiku Morris; Yuting, Han
    This presentation characterizes the source-localization precision obtainable via "received signal strength indication" (RSSI) based on data from a tri-axial velocity sensor and a spatially separated pressure sensor. That scheme was proposed originally by Y. I. Wu and K. T. Wong in January 2012 in the IEEE Transactions on Aerospace and Electronic Systems. That source-localization scheme depends on the acoustic propagation path-loss exponent, which is typically not precisely known a priori but could be modeled stochastically. That exponent may also differ in value for the propagation path to the tri-axial velocity sensor and for the propagation path to the pressure sensor. This presentation accounts for these two practical considerations in characterizing the scheme's source-localization precision, through the metric of the "hybrid Cramer-Rao bound" (HCRB), the correctness of which is here validated by Monte Carlo simulations of the corresponding "maximum a posteriori" (MAP) estimator.
  • No Thumbnail Available
    Item
    On the Performance of L- and V-Shaped Arrays of Cardioid Microphones for Direction Finding
    (IEEE, 2021-01-15) Nnonyelu, Chibuzo Joseph; Zakayo, Ndiku Morris; Chinaza, Alice Madukwe
    The L-shaped and V-shaped arrays of first-order cardioid microphones for direction finding are presented in this paper. A comparative study of the direction of arrival estimation performance of the arrays was carried out by analytically deriving, in closed-form, and comparing the Cramér-Rao bounds of an incident signal's direction-of-arrival (DoA) azimuth and polar angles for these arrays. The maximum-likelihood estimator is used to verify the correctness of derived bounds. This investigation reveals that for direction finding, the L-shaped array of cardioid microphones would generally outperform the V-shaped array of cardioid microphones in more sub-regions of the DoA polar-azimuth angle space. However, in regions where the V-shaped array of cardioid microphones outperforms the L-shaped array of cardioid microphone, the variance ratios are usually higher in favor of the V-shaped array.