doaspc1.m

阵列信号处理的工具箱

function  spect = doaspc1(method, sigIn, specType, smplPoints, focusDist, rangeIx, cpiIx, in8, in9, in10, timeIx, snapshotDim)


%DOASPC1 Estimates a DOA-spectrum with different methods.
%
%  spect = doaspc1(method, sigIn, [], smplPoints, focusDist, rangeIx,
%    cpiIx, ...)
%
%More detailed help text by commands "help2 doaspc1" or "helpwin2 "doaspc1".


%DOASPC1 Estimates a DOA-spectrum with different methods.
%
%--------
%Synopsis:
%  spect = doaspc1(method, sigIn, [], smplPoints, focusDist, rangeIx,
%    cpiIx, ...)
%
%Description:
%  Estimates a DOA spectrum ((Direction of Arrival)) with different methods.
%
%  If the input parameters "rangeIx" or "cpiIx" are vectors, several DOA
%  spectra will be estimated, one for each specified range bin and CPI.
%
%  The input can either be a radar signal (RxCorrMatT) or a spatial
%  correlation matrix (RxCorrMatT). In the former case a correlation
%  matrix is estimated for all range bins and CPIs by calling
%     Rxx1  = ecorrm(sigIn,[],rangeIx, cpiIx);
%  This means that the samples used for the estimation are taken from all
%  pulses (see function "eccorrm").
%
%Output and Input:
%  spect (DoaSpecT): Output DOA-spectrum. It is a quadratic measure of the
%    presence of sources in different directions. It can be either a power
%    spectrum (e.g. cbf) or a pseudo spectrum (e.g. music).
%    All methods use the antenna signal correlation matrix, which is
%    estimated by the function "ecorrm" if it is not supplied as input.
%  method (StringT): Estimation method. See below for sections about each
%    method.
%    = 'cbf': Conventional beamforming.
%    = 'acbf': Adaptive conventional beamforming. This is beamforming with
%      adaptive sidelobe cancelation (ASLC). Not implemented.
%    = 'capon': Capons beamformer.
%    = 'capon2': An other implementation of Capons beamformer.
%    = 'music : MUSIC.
%    = 'music2': An other implementation of MUSIC.
%    = 'minnorm': Minimum norm.
%    = 'minnorm2': An other implementation of Minimum norm (freath).
%    = 'eigvec': Eigenvector method.
%    = 'pisa': Pisarenko Harmonic Decomposition.
%  sigIn (RxRadarSigT or RxCorrMatT): The input antenna signal or a
%    spatial correlation matrix.
%  smplPoints [D](Vector of DoaT): Calculate the spectrum at these angles.
%  focusDist [D](RealScalarT): Focus distance of the antenna [m]
%    (default = Inf). Not implemented.
%  rangeIx [D](IndexT): Indices of range bin (default = 1). Only used
%    when "sigIn" is of type RxRadarSigT. Then default is all range bins.
%  cpiIx [D](IndexT): CPI indices (default = 1). Only used
%    when "sigIn" is of type RxRadarSigT. Then default is all CPIs.
%  timeIx : Not Implemented.
%  snapshotDim [D](IntScalarT): Use snapshots from this dimension to
%    estimate the correlation matrix.  Not Implemented.
%
%--------
%Conventional Beamforming.
%Synopsis:
%  spect = doaspc1('cbf', sigIn)
%  spect = doaspc1('cbf', sigIn, [], smplPoints)
%  spect = doaspc1('cbf', sigIn, [], smplPoints, [], rangeIx, cpiIx)
%  spect = doaspc1('cbf', sigIn, [], [], [], [], [], taperType, taperParam)
%  spect = doaspc1('cbf', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    taperType, taperParam)
%
%Output and Input:
%   taperType (StringT): Tapering type.
%     = 'norm': Normalised beamforming ([2] p. 72-73).
%     = Otherwise the parameter "taperType" of function "gettap1".
%   taperParam (?): The third input parameter of function "gettap1".
%
%Description:
%  The conventional beamforming. It can be done either with any of the
%  taperings that are available with te function gettap1 or by a kind of
%  normalization.
%
%Algoritm:
%  taperType = 'norm':
%    P = (a'*Rxx*a)/(a'*a),
%    where P=P(doa) is the output spectrum as a function of direction,
%    a=a(doa) is the steering vector calculated by the spastemat function,
%    Rxx is an estimate of the input antenna signal correlation matrix.
%  taperType = Otherwise:
%     w = a .* taper
%     P = w'*Rxx*w,
%     where taper is a tapering from the function gettap1. Other variables
%     are as above.
%
%--------
%Adaptive Conventional Beamforming.
%Synopsis:
%  spect = doaspc1('acbf', sigIn, [], [], [], [], [], [], [], iNCorrMat)
%  spect = doaspc1('acbf', sigIn, [], [], [], [], [], taperType, taperParam,
%    iNCorrMat)
%  spect = doaspc1('acbf', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    taperType, taperParam, iNCorrMat)
%
%Output and Input:
%  taperType  (StringT): Tapering type.
%    = The parameter "taperType" of function "gettap1".
%  taperParam (?): The third input parameter of function "gettap1".
%  iNCorrMat (RxCorrMatT): The input antenna signal interference-plus-noise
%    correlation matrix. This should be estimated on a radar signal without
%    targets.
%
%Description:
%  Not implemented.
%  Adaptive (conventional) beamforming. Reference [4] p. 57-58 contains a
%  description of STAP which is a generalization of adaptive beamformning.
%  Possibility to choose a desired weight vector ([4] p. 57-58).
%
%Algoritm:
%
%Examples:
%  ant = defant('isotropULA',[noElem,D,lambda]);
%  sigJam = compsim2(ant, 1, T, jamModel, [thetaJam, phiJam, SNRJam, ...
%    alphaJam, dalphaJam, dist, ones(size(theta,1))], 'rndn', eye(noElem));
%  iNCorrMat = ecorrm(sigJam);
%  sigTgt = compsim2(ant, 1, T, tgtModel, [thetaTgt, phiTgt, SNRTgt, ...
%    alphaTgt, dalphaTgt, distTgt, ones(size(theta,1))], 'rndn', eye(noElem));
%  splot2(doaspc1('acbf', sigTgt, [], 'taylor', 40, iNCorrMat))
%
%--------
%Capons Beamformer.
%Synopsis:
%  spect = doaspc1('capon', sigIn)
%  spect = doaspc1('capon', sigIn, [], smplPoints)
%  spect = doaspc1('capon', sigIn, [], smplPoints, [], rangeIx, cpiIx)
%
%Description:
%  This is another adaptive conventional beamformer (see 'acbf') but there
%  is no possibility to choose a desired weight vector.
%  This method is the solution to a minimization that attempts to minimize
%  signals from other direction other than the current look direction, in
%  which a constant gain is maintained, [2] p73-74.
%  This method is also known as Minimum Variance Distorionless Response.
%
%Algoritm:
%  1/(a'*inv(Rxx)*a),
%  where P=P(doa) is the output spectrum as a function of direction,
%  a=a(doa) is the steering vector calculated by the spastemat function,
%  Rxx is an estimate of the input antenna signal correlation matrix.
%
%--------
%MUSIC.
%Synopsis:
%  spect = doaspc1('music', sigIn, [], [], [], [], [], noSrc)
%  spect = doaspc1('music', sigIn, [], smplPoints, [], [], [], noSrc)
%  spect = doaspc1('music', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    noSrc)
%
%Output and Input:
%   noSrc [D](IntScalarT): Hypotesis for the number of targets or signals.
%
%Description:
%  Multiple Signal Classification. Probably the most known model based DOA
%  spectrum estimation method. See for example [2] p. 74-75.
%
%Algoritm:
%  See [3].
%
%--------
%Minimum  Norm.
%Synopsis:
%  spect = doaspc1('minnorm', sigIn, [], [], [], [], [], noSrc)
%  spect = doaspc1('minnorm', sigIn, [], smplPoints, [], [], [], noSrc)
%  spect = doaspc1('minnorm', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    noSrc)
%
%Output and Input:
%   noSrc (IntScalarT): Hypotesis for the number of targets or signals.
%
%Description:
%  A modification of the MUSIC method. See [2] p. 75.
%Algoritm:
%
%--------
%Minimum  Norm 2.
%Synopsis:
%  spect = doaspc1('minnorm2', sigIn, [], [], [], [], [], noSrc)
%  spect = doaspc1('minnorm2', sigIn, [], smplPoints, [], [], [], noSrc)
%  spect = doaspc1('minnorm2', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    noSrc)
%
%Output and Input:
%   noSrc (IntScalarT): Hypotesis for the number of targets or signals.
%
%Description:
%  An other implementation of Minimum norm (by Fredrik Athley).
%
%Algoritm:
%  See [3].
%
%--------
%Eigenvector Method.
%Synopsis:
%  spect = doaspc1('eigvec', sigIn, [], [], [], [], [], noSrc)
%  spect = doaspc1('eigvec', sigIn, [], smplPoints, [], [], [], noSrc)
%  spect = doaspc1('eigvec', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    noSrc)
%
%Output and Input:
%   noSrc (IntScalarT): Hypotesis for the number of targets or signals.
%
%Description:
%  See [6] p. 9 and [7] p. 384.
%
%Algoritm:
%
%--------
%Pisarenko Harmonic Decomposition.
%Synopsis:
%  spect = doaspc1('pisa', sigIn, [], [], [], [], [], noSrc)
%  spect = doaspc1('pisa', sigIn, [], smplPoints, [], [], [], noSrc)
%  spect = doaspc1('pisa', sigIn, [], smplPoints, [], rangeIx, cpiIx,
%    noSrc)
%
%Output and Input:
%   noSrc (IntScalarT): Hypotesis for the number of targets or signals.
%
%Description:
%  See [6] p. 10. and [7] p. 371.
%  This method does not seem to work.
%
%Algoritm:
%
%--------
%Notations:
%  Data type names are shown in parentheses and they start with a capital
%  letter and end with a capital T. Data type definitions can be found in [1]
%  or by "help dbtdata".
%  [D] = This parameter can be omitted and then a default value is used.
%  When the [D]-input parameter is not the last used in the call, it must be
%  given the value [], i.e. an empty matrix.
%  ... = There can be more parameters. They are explained under respective
%  metod or choice.
%
%Examples:
%  ant = defant('isotropULA',[noElem,D,lambda]);
%  sig = compsim2(ant, 1, T, tgtModel, [theta, phi, SNR, alpha, dalpha, ...
%    dist, ones(size(theta,1))], 'rndn', eye(noElem));
%  splot2(sdoaspc('music',sig,[],MMu))
%  figure;
%  splot2(doaspc1('cbf', sig, [], d2r(-30:0.1:30), [], 3, 2, 'taylor', 40));
%
%Software Quality:
%  (About what is done to ascertain software quality. What tests are done.)
%  The methods 'cbf', 'capon', 'music' and 'minnorm' were used in the work
%  done by A. Heydarkhan [5].
%
%Known Bugs:
%  The wavelength is acquired from the antenna definition. This is not
%  logical but works as the wavelength happens to be stored there.
%
%References:
%  [1]: Bj鰎klund S.: "DBT, A MATLAB Toolbox for Radar Signal Processing.
%    Reference Guide", FOA-D--9x-00xxx-408--SE, To be published.
%  [2]: Krim H., Viberg M.:"Two Decades of Array Signal Processing Research",
%    IEEE Signal Processing Magazine, July 1996, pp. 67-94.
%  [3]: Fredrik Athley, Ericsson Microwave Systems, M鰈ndal, Sweden.
%  [4]: Ward J.: "Space-Time Adaptive Processing for Airborne Radar",
%    Technical report 1015, MIT Lincoln Laboratory 1994.
%  [5]: Heydarkhan A.: "Model Based Direction of Arrival Estimation Methods
%    Applied to Experimental Antenna Data", Master's Thesis, Chalmers
%    University of Technology 1997. FOA-R--97-00631-408--SE, FOA November 1997.
%  [6]: Pettersson L.: "Inverkan av fel p