function micp


Loc_n=4;
Sen_n=3;
Nvar=Loc_n*Sen_n;  %Variable number

%Constants
C=[1 1;1 0;0 1;0 1];
s1=5.0693;
s4=3.8259;
Cov_M=[1.501 0.523 0.978 0.978; 3.002 1.046 1.956 1.956; 4.503 1.569 2.934 2.934]';
Z=[2500 1500 800]; %$,cost

%variable claim
X=binvar(Nvar,1);    % [q11 q12 q13 q21 q22 q23 q31 q32 q33 q41 q42 q43]

%Constraints (LMIs)

for i=1:Loc_n
    k=3*i-2;
    Xigma_inv(i,i)=1./Cov_M(i,:).^2*X(k:k+2);
end

P1(2:3,2:3)=C'*Xigma_inv*C;
P1(1,1)=s1;
P1(1,2:3)=C(1,:);
P1(2:3,1)=C(1,:)';

P2=P1;
P2(1,1)=s4;
P2(1,2:3)=C(4,:);
P2(2:3,1)=C(4,:)';

P3(1,1)=1-sum(X(1:3));
P3(2,2)=1-sum(X(4:6));
P3(3,3)=1-sum(X(7:9));
P3(4,4)=1-sum(X(10:12));


F=set(P1>0)+set(P2>0)+set(diag(P3)>=0);

%Objective
cc=zeros(1,Nvar);
cc(1:Sen_n)=Z;
cc(Sen_n+1:2*Sen_n)=Z;
cc(2*Sen_n+1:3*Sen_n)=Z;
cc(3*Sen_n+1:end)=Z;

obj=cc*X;

sol = solvesdp(F,obj);

mbg_asserttolequal(sol.problem,0);
mbg_asserttolequal(double(obj), 3000, 1e-5);