matlab_调频连续波雷达

classdef (Sealed) PhaseShiftBeamformer < phased.internal.AbstractNarrowbandBeamformer %PhaseShiftBeamformer Narrowband phase shift beamformer % H = phased.PhaseShiftBeamformer creates a conventional phase shift % beamformer System object, H. This object performs phase shift % beamforming on the received signal. % % H = phased.PhaseShiftBeamformer(Name,Value) creates a phase shift % beamformer object, H, with the specified property Name set to the % specified Value. You can specify additional name-value pair arguments % in any order as (Name1,Value1,...,NameN,ValueN). % % The phase shift beamformer uses the conventional delay and sum % beamforming algorithm. It assumes that the signal is narrowband, thus % using a phase shift to approximate the required delay. The beamformer % preserves the incoming signal power. % % Step method syntax: % % Y = step(H,X) performs phase shift beamforming on the input X, and % returns the beamformed output in Y. X is an MxN matrix where N is the % number of subarrays if SensorArray contains subarrays, or the number of % elements otherwise. Y is an MxL matrix where L is the number of % beamforming directions. % % [Y,W] = step(H,X) returns additional output W as the beamforming % weights when you set the WeightsOutputPort property to true. W is an % NxL matrix. % % Y = step(H,X,ANG) uses ANG as the beamforming direction, when you % set the DirectionSource property to 'Input port'. ANG is a 2-row % matrix whose columns are in the form of [AzimuthAngle; ElevationAngle] % (in degrees). The azimuth angle must be between [-180 180] and the % elevation angle must be between [-90 90]. % % You can combine optional input arguments when their enabling properties % are set. Optional inputs must be listed in the same order as the order % of the enabling properties. For example, % % [Y,W] = step(H,X,ANG) % % System objects may be called directly like a function instead of using % the step method. For example, y = step(H, x) and y = H(x) are % equivalent. % % PhaseShiftBeamformer methods: % % step - Perform phase shift beamforming (see above) % release - Allow property value and input characteristics changes % clone - Create phase shift beamformer object with same property % values % isLocked - Locked status (logical) % % PhaseShiftBeamformer properties: % % SensorArray - Sensor array % PropagationSpeed - Signal propagation speed % OperatingFrequency - Operating frequency % DirectionSource - Source of beamforming direction % Direction - Beamforming direction % NumPhaseShifterBits - Number of bits in phase shifters % WeightsNormalization - Weights normalizing method % WeightsOutputPort - Enable weights output % % % Example: % % Apply phase shift beamforming to the signal received by a % % 5-element ULA. The beamforming direction is 45 degrees azimuth and % % 0 degrees elevation. % % % simulate signal % t = (0:1000)'; % x = sin(2*pi*0.01*t); % c = 3e8; Fc = 3e8; % incidentAngle = [45; 0]; % ha = phased.ULA('NumElements',5); % x = collectPlaneWave(ha,x,incidentAngle,Fc,c); % noise = 0.1*(randn(size(x)) + 1j*randn(size(x))); % rx = x + noise; % % % beamforming % hbf = phased.PhaseShiftBeamformer('SensorArray',ha,... % 'OperatingFrequency',Fc,'PropagationSpeed',c,... % 'Direction',incidentAngle,'WeightsOutputPort',true); % [y,w] = step(hbf, rx); % plot(t,real(rx(:,3)),'r:',t,real(y)); % xlabel('Time'),ylabel('Amplitude'),legend('Original','Beamformed'); % % % plot response pattern % figure; pattern(ha,Fc,-180:180,0,'PropagationSpeed',c,'Weights',w); % % See also phased, phased.MVDRBeamformer, phased.LCMVBeamformer, % phased.SubbandPhaseShiftBeamformer. % Copyright 2009-2016 The MathWorks, Inc. % Reference % [1] Van Trees, Optimum Array Processing, Wiley, 2002 %#ok<*EMCLS> %#ok<*EMCA> %#codegen properties (Nontunable) %WeightsNormalization Weights normalizing method % Specify the method of normalizing the beamformer weights as one % of 'Distortionless' | 'Preserve power', where the default is % 'Distortionless'. If WeightsNormalization is 'Distortionless', % the gain toward the beamforming direction is 0 dB. If % WeightsNormalization is 'Preserve power', then the norm of the % weights is 1. WeightsNormalization = 'Distortionless' end properties(Access = private) %pre-calculated weights pWeights; end properties (Constant,Hidden) WeightsNormalizationSet = matlab.system.StringSet(... {'Distortionless','Preserve power'}); end methods function obj = PhaseShiftBeamformer(varargin) obj@phased.internal.AbstractNarrowbandBeamformer(varargin{:}); end end methods (Access = 'protected') function flag = isMultipleInputAnglesAllowed(obj) %#ok<MANU> flag = true; end function flag = isSubarraySupported(obj) %#ok<MANU> flag = true; end function setupImpl(obj,x,varargin) setupImpl@phased.internal.AbstractNarrowbandBeamformer(obj,x); if (obj.DirectionSource(1) == 'P') %Property obj.pWeights = calcWeights(obj,obj.Direction); end end function flag = isInputComplexityLockedImpl(obj,index) flag = isInputComplexityLockedImpl@phased.internal.AbstractNarrowbandBeamformer(obj,index); if (obj.DirectionSource(1) == 'I') && (index == 2) %Input flag = true; end end function flag = isInputSizeLockedImpl(obj,index) %#ok<INUSL> if index == 1 flag = false; else flag = true; end end function s = saveObjectImpl(obj) s = saveObjectImpl@phased.internal.AbstractNarrowbandBeamformer(obj); if isLocked(obj) s.pWeights = obj.pWeights; end end function s = loadSubObjects(obj,s) s = loadSubObjects@phased.internal.AbstractNarrowbandBeamformer(obj,s); if isfield(s,'isLocked') s = rmfield(s,'isLocked'); end end function loadObjectImpl(obj,s,wasLocked) %#ok<INUSD> s = loadSubObjects(obj,s); fn = fieldnames(s); for m = 1:numel(fn) obj.(fn{m}) = s.(fn{m}); end end function [y, w] = stepImpl(obj,x,varargin) if (obj.DirectionSource(1) == 'P') %Property w = obj.pWeights; else ang = varargin{1}; validateAngleRange(obj,ang); w = calcWeights(obj,ang); end y = x*conj(w); end function validateInputsImpl(obj,x,varargin) validateInputsImpl@phased.internal.AbstractNarrowbandBeamformer(obj,x); if strncmpi(obj.DirectionSource,'i',1) ang = varargin{1}; validateInputAngle(obj,ang); end end end methods (Access = private) function w = calcWeights(obj,ang) sv = step(obj.cSteeringVector,obj.OperatingFrequency,ang); w = complex(zeros(size(sv))); if (obj.WeightsNormalization(1) == 'D') %Distortionless for m = size(ang,2):-1:1 temp_sv = sv(:,m); w(:,m) = temp_sv/real(temp_sv'*temp_sv); % distortionless end else for m = size(ang,2):-1:1 temp_sv = sv(:,m); w(:,m) = temp_sv/norm(temp_sv); % preserve power