I need to solve:
argmin||W*FT^-1(Ax)-W*p||
using lsqr. p is image, x is k-space matrix, A is a matrix and W is a weighting matrix. In order to pass them to matlab lsqr, I vectorized p, x and W.
This is my code:
b=W.*p;
x=lsqr(#(x1,modo)FUNC(x1,W_vector,A,modo),b,tol,maxit);
X_k_lsqr=reshape(x,dim(1),dim(2),dim(3));
X_lsqr=real(ifftn(X_k_lsqr)).*MASK(:,:,:,1);
%% Auxiliary function
function [result modo]=FUNC(x1,W_vector,A,modo)
%Computes y=A*x for modo='notransp'
%Computes y=A'*x for modo='transp'
switch modo
case 'notransp'
res=A*x1;
R1=reshape(res,norient,dim(1)*dim(2)*dim(3));
for co=1:norient
R2(:,:,:,co)=reshape(R1(co,:),dim(1),dim(2),dim(3));
FR(:,:,:,co)=ifftn(R2(:,:,:,co));
aux=FR(:,:,:,co);
R3(co,:)=aux(:).';
end
result=W.*R3(:);
case 'transp'
RR1=reshape(x1./(W+eps),norient,dim(1)*dim(2)*dim(3));
for co=1:norient
RR2(:,:,:,co)=reshape(RR1(co,:),dim(1),dim(2),dim(3));
FRR(:,:,:,co)=fftn(RR2(:,:,:,co));
aux=FRR(:,:,:,co);
RR3(co,:)=aux(:).';
end
result=A'*RR3(:);
end
end
As W appears in both terms of the minimization problem, I would have expected the image I obtain as a result to be almost independent on W values.
The image looks qualitatively the same if I change W, but its values strongly depend on W. I don't know if something is wrong with my code.
Should I actually obtain almost the same values for different W?
A question: Have you checked the flag? Are you sure that you are using the results while the flag is 0?
An argument: I can assume W also as a switch that determines which components of p I want to minimize. If W*(Ax-b) is really equal to 0, then you are right but here we know that there is not an exact solution and it becomes W*(Ax-b) < W*tolerance. So, you are depending the end point of the algorithm on the elements of W together with the tolerance.
Related
How do I solve the following system of equations on MATLAB when one of the elements of the variable vector is a constant? Please do give the code if possible.
More generally, if the solution is to use symbolic math, how will I go about generating large number of variables, say 12 (rather than just two) even before solving them?
For example, create a number of symbolic variables using syms, and then make the system of equations like below.
syms a1 a2
A = [matrix]
x = [1;a1;a2];
y = [1;0;0];
eqs = A*x == y
sol = solve(eqs,[a1, a2])
sol.a1
sol.a2
In case you have a system with many variables, you could define all the symbols using syms, and solve it like above.
You could also perform a parameter optimization with fminsearch. First you have to define a cost function, in a separate function file, in this example called cost_fcn.m.
function J = cost_fcn(p)
% make sure p is a vector
p = reshape(p, [length(p) 1]);
% system of equations, can be linear or nonlinear
A = magic(12); % your system, I took some arbitrary matrix
sol = A*p;
% the goal of the system of equations to reach, can be zero, or some other
% vector
goal = zeros(12,1);
% calculate the error
error = goal - sol;
% Use a cost criterion, e.g. sum of squares
J = sum(error.^2);
end
This cost function will contain your system of equations, and goal solution. This can be any kind of system. The vector p will contain the parameters that are being estimated, which will be optimized, starting from some initial guess. To do the optimization, you will have to create a script:
% initial guess, can be zeros, or some other starting point
p0 = zeros(12,1);
% do the parameter optimization
p = fminsearch(#cost_fcn, p0);
In this case p0 is the initial guess, which you provide to fminsearch. Then the values of this initial guess will be incremented, until a minimum to the cost function is found. When the parameter optimization is finished, p will contain the parameters that will result in the lowest error for your system of equations. It is however possible that this is a local minimum, if there is no exact solution to the problem.
Your system is over-constrained, meaning you have more equations than unknown, so you can't solve it. What you can do is find a least square solution, using mldivide. First re-arrange your equations so that you have all the constant terms on the right side of the equal sign, then use mldivide:
>> A = [0.0297 -1.7796; 2.2749 0.0297; 0.0297 2.2749]
A =
0.029700 -1.779600
2.274900 0.029700
0.029700 2.274900
>> b = [1-2.2749; -0.0297; 1.7796]
b =
-1.274900
-0.029700
1.779600
>> A\b
ans =
-0.022191
0.757299
Since the original problem is more complicated, the idea is described using a simple example below.
For example, suppose we want to put several router antennas somewhere in a room so that the cellphone get most signal strength on the table (received power > Pmax) while weakest signal strength on bed (received power < Pmin). What is the best (minimum) number of antennas that should be used, and where should they be placed, in order to achieve the goal.
Mathematically,
SIGNAL_STRENGTH is dependent on variable (x, y, z) and the number
of variables
. i.e. location and number of antennas.
Besides, assume
PREDICTION = f((x1, y1, z1), (x2, y2, z2), ... (xi, yi, zi), ... (xn,
yn, zn))
where n and (xi, yi, zi) are to be optimized. The goal is to minimize
cost function = ||SIGNAL_STRENGTH - PREDICTION||
I tried to use GA with mixed integer programming in Matlab to implement that. Two optimization functions are used, outer function is to optimize n, and inner optimization function optimizes (x, y, z) with given n. This method works slow and I haven't seen one result given by this method so far.
Does anyone have a more efficient way to solve this problem? Any suggestion is appreciated. Thanks in advance.
Terminology | Problem Definition
An antenna is sending at position a in R^3 with constant power. Its signal strength can be measured by some S: R^3 -> R where S has a single maximum S_0 at a and the set, constructed by S(x) > const, is simply connected, i.e. S(x) = S_0 * exp(-const * (x-a)^2).
Given a set of antennas A the resulting signal strength is the maximum of a single antenna
S_A(x) = max{S_a(x) : for all a in A} ,
which means we 'lock' on the strongest antenna, which is what cell phones do.
Let K = R^3 x R denote a space of points (position, intensity). Now concider two finite subsets POI_min and POI_max of K. We want to find the set A with the minimal amount of antennas (|A| -> min.), that satisfies
for all (x,w) in POI_min : S_A(x) < w and for all (x,w) in POI_max : S_A(x) > w .
Implication
As S(x) > const is simply connected there has to be an antenna in a sphere around the position of each element (x,w) in POI_max with radius r = max{||xi - x|| : for all xi in S(xi) = w}. Which means that if we would put an antenna at the position of (x,w), then the furthest we can go away from x and still have signal strength w is the radius r within which an actual antenna has to be positioned.
With a similar argumentation for POI_min it follows that there is no antenna within r = min{||xi - x|| : for all xi in S(xi) = w}.
Solution
Instead of solving a nonlinear optimization task we can intersect spheres to obtain the optimal solution. If k spheres around the POI_max positions intersect, we can place a single antenna in the intersection, reducing the amount of antennas needed by k-1.
However each antenna that is placed must satisfy all constraints given by the elements of POI_min. Assuming that antennas are omnidirectional and thus orientation of an antenna doesn't matter we can do (pseudocode):
min_sphere = {(x_i,r_i) : from POI_min},
spheres_to_cover = {(x_i,r_i) : from POI_max}
A = {}
while not is_empty(spheres_to_cover)
power_set_score = struct // holds score, k
PS <- costruct power set of sphere_to_cover
for i = 1:number_of_elements(PS)
k = PS[i]
if intersection(k) \ min_sphere is not empty
power_set_score[i].score = |k|
else
power_set_score[i].score = 0
end if
power_set_score[i].k = k
end for
sort(power_set_score) // sort by score, biggest first
A <- add arbitrary point in (intersection(power_set_score[1].k) \ min_sphere)
spheres_to_cover = spheres_to_cover \ power_set_score[1].k
end while
On the other hand you have just given an example problem and thus this solution may not be applicable or broad enough for your case. I did make a few assumptions. So being more specific in the question might give you an even better answer.
I am trying to solve
argmin||W_vector*FT^-1(Ax)-W_vector*B||
using lsqr with a function handle and without any initial guess (W_vector is a weighting vector). I thought that lsqr would have performed the first step computing x by C'b (call C=W*FT^-1(A)) using the vector b=W_vector.*B I passed to the function. This is the code:
b=W_vector.*B;
[x, flag]=lsqr(#(x1,modo)FUNC(x1,W_vector,A,modo),b,tol,maxit);
function [result, modo]=FUNC(x1,W_vector,A,modo)
%Computes y=A*x for modo='notransp'
%Computes y=A'*x for modo='transp'
switch modo
case 'notransp'
res=A*x1;
R1=reshape(res,norient,dim(1)*dim(2)*dim(3));
for co=1:norient
R2(:,:,:,co)=reshape(R1(co,:),dim(1),dim(2),dim(3));
FR(:,:,:,co)=ifftn(R2(:,:,:,co));
aux=FR(:,:,:,co);
R3(co,:)=aux(:).';
end
result=W_vector.*R3(:);
case 'transp'
R1=reshape(x1./(W_vector+eps),norient,dim(1)*dim(2)*dim(3));
for co=1:norient
R2(:,:,:,co)=reshape(R1(co,:),dim(1),dim(2),dim(3));
FR(:,:,:,co)=fftn(R2(:,:,:,co));
aux=FR(:,:,:,co);
R3(co,:)=aux(:).';
end
result=A'*R3(:);
end
end
When I checked the result of R1=reshape(x1./(W_vector+eps),norient,dim(1)*dim(2)*dim(3)); in case 'transp' in the first iteration, I found out that x1 is not equal to b (after reshape, they are images, so I can see that look qualitatively the same but with very different values). I would like to understand how lsqr works because I am having some problems with weightings. Thanks for your help.
having a problem with my "new love", matlab: I wrote a function to calculate an integral using the trapz-method: `
function [L]=bogenlaenge_innen(schwingungen)
R = 1500; %Ablegeradius
OA = 1; %Amplitude
S = schwingungen; %Schwingungszahl
B = 3.175; %Tapebreite
phi = 0:2.*pi./10000:2.*pi;
BL = sqrt((R-B).^2+2.*(R-B).*OA.*sin(S.*phi)+OA.^2.*(sin(S.*phi)).^2+OA.^2.*S.^2.*(cos(S.*phi)).^2);
L = trapz(phi,BL)`
this works fine when i start it with one specific number out of the command-window. Now I want to plot the values of "L" for several S.
I did the following in a new *.m-file:
W = (0:1:1500);
T = bogenlaenge_innen(W);
plot(W,T)
And there it is:
Error using .*
Matrix dimensions must agree.
What is wrong? Is it just a dot somewhere? I am using matlab for the second day now, so please be patient.... ;) Thank you so much in advance!
PS: just ignore the german part of the code, it does not really matter :)
In your code, the arrays S and phi in the expression sin(S.*phi) should have same size or one of them should be a constant in order the code works
The error is most likely because you have made it so that the number of elements in schwingungen, i.e. W in your code, must be equal to the number of elements in phi. Since size(W) gives you a different result from size(0:2.*pi./10000:2.*pi), you get the error.
The way .* works is that is multiplies each corresponding elements of two matrices provided that they either have the same dimensions or that one of them is a scalar. So your code will work when schwingungen is a scalar, but not when it's a vector as chances are it has a different number of elements from the way you hard coded phi.
The simplest course of action (not necessarily the most Matlabesque though) for you is to loop through the different values of S:
W = (0:1:1500);
T = zeros(size(W); %Preallocate for speed)
for ii = 1:length(W)
T(ii) = bogenlaenge_innen(W(ii));
end
plot(W,T)
In your function you define phi as a vector of 10001 elements.
In this same function you do S.*phi, so if S is not the same length as phi, you will get the "dimensions must agree" error.
In your call to the function you are doing it with a vector of length 1501, so there is your error.
Regards
How solve a system of ordinary differential equation ..an initial value problem ....with parameters dependent on time or independent variable?
say the equation I have
Dy(1)/dt=a(t)*y(1)+b(t)*y(2);
Dy(2)/dt=-a(t)*y(3)+b(t)*y(1);
Dy(3)/dt=a(t)*y(2);
where a(t) is a vector and b(t) =c*a(t); where the value of a and b are changing with time not in monotone way and each time step.
I tried to solve this using this post....but when I applied the same principle ...I got the error message
"Error using griddedInterpolant The point coordinates are not
sequenced in strict monotonic order."
Can someone please help me out?
Please read until the end to see whether the first part or second part of the answer is relevant to you:
Part 1:
First create an .m file with a function that describe your calculation and functions that will give a and b. For example: create a file called fun_name.m that will contain the following code:
function Dy = fun_name(t,y)
Dy=[ a(t)*y(1)+b(t)*y(2); ...
-a(t)*y(3)+b(t)*y(1); ...
a(t)*y(2)] ;
end
function fa=a(t);
fa=cos(t); % or place whatever you want to place for a(t)..
end
function fb=b(t);
fb=sin(t); % or place whatever you want to place for b(t)..
end
Then use a second file with the following code:
t_values=linspace(0,10,101); % the time vector you want to use, or use tspan type vector, [0 10]
initial_cond=[1 ; 0 ; 0];
[tv,Yv]=ode45('fun_name',t_values,initial_cond);
plot(tv,Yv(:,1),'+',tv,Yv(:,2),'x',tv,Yv(:,3),'o');
legend('y1','y2','y3');
Of course for the fun_name.m case I wrote you need not use sub functions for a(t) and b(t), you can just use the explicit functional form in Dy if that is possible (like cos(t) etc).
Part 2: If a(t) , b(t) are just vectors of numbers you happen to have that cannot be expressed as a function of t (as in part 1), then you'll need to have also a time vector for which each of them happens, this can be of course the same time you'll use for the ODE, but it need not be, as long as an interpolation will work. I'll treat the general case, when they have different time spans or resolutions. Then you can do something of the following, create the fun_name.m file:
function Dy = fun_name(t, y, at, a, bt, b)
a = interp1(at, a, t); % Interpolate the data set (at, a) at times t
b = interp1(at, b, t); % Interpolate the data set (bt, b) at times t
Dy=[ a*y(1)+b*y(2); ...
-a*y(3)+b*y(1); ...
a*y(2)] ;
In order to use it, see the following script:
%generate bogus `a` ad `b` function vectors with different time vectors `at` and `bt`
at= linspace(-1, 11, 74); % Generate t for a in a generic case where their time span and sampling can be different
bt= linspace(-3, 33, 122); % Generate t for b
a=rand(numel(at,1));
b=rand(numel(bt,1));
% or use those you have, but you also need to pass their time info...
t_values=linspace(0,10,101); % the time vector you want to use
initial_cond=[1 ; 0 ; 0];
[tv,Yv]= ode45(#(t,y) fun_name(t, y, at, a, bt, b), t_values, initial_cond); %
plot(tv,Yv(:,1),'+',tv,Yv(:,2),'x',tv,Yv(:,3),'o');
legend('y1','y2','y3');