KF UD decomposition pseudo code implementation advanced [2]

I don't say much nonsense , Directly on pseudo code ,matlab edition , Easy to read .
be based on UD Of KF In two parts ：
The first part realizes the residual sum v Update ;
The second part is about D\U\b Update of three parts .

%%  be based on UD Disintegrated KF Simulation code 
% ------ Part One Start ------
%  This part implements  J>1  Residual calculation and v Update 
for J=n:-1:2
    z=z-a(J)*x(J); % residual=delta_z
    for K=1:J-1
        a(J)=a(J)+U(K,J)*a(K); % f=U^T*a,  After the assignment a(J) Namely f
    end
    v(J)=U(J,J)*a(J);% vi=di*fi
end
%  This part implements  J=1  Residual calculation and v Update 
z=z-a(1)*x(1);
v(1)=U(1,1)*a(1);
% ------ Part One Last ------

% ------ Part Two Start------
%  This part implements  J=1  matrix D、 matrix U、 Unweighted KF gain b Three part update 
alpha=r+v(1)*a(1);
gamma=1.0/alpha;
U(1,1)=U(1,1)*r*gamma;
b(1)=v(1);

%  This part implements  J>1  matrix D、 matrix U、 Unweighted KF gain b Three part update 
for J=2:N
    beta=alpha; %  take alpha(J-1) Plug in 
    alpha=alpha+v(J)*a(J); %  to update alpha(J)
    p=-a(J)*gamma; %  take gamma(J-1) Plug in 
    gamma=1.0/alpha; %  to update gamma(J)
    U(J,J)=U(J,J)*beta*gamma; %  to update D
    b(J)=v(J);
    for I=1:J-1
        beta=U(I,J);
        U(I,J)=beta+b(I)*p; %  to update U
        b(I)=b(I)+v(J)*beta;%  to update b
    end
end

z=z/alpha; % delta_z=delta_z/alpha, namely delta_z Is the weighted residual 
for J=1:N
    x(J)=x(J)+b(J)*z; % x_=x+(k)*delta_z=x+(b/alpha)*delta_z, namely k=b/alpha, Realized with b A weighted 
end
% ------ Part Two Last ------