This is just straightforward arithmetic, so the conversion is pretty simple. I would forget about using single precision floats though unless you are trying to match a specific precision calculation. Just use doubles.
However, it appears that some of the inputs are top level variables that are not coming into the function via the argument list which is void. So you will have to figure out what those variables are and put them into the input argument list. And it appears that these top level variables are also outputs of the function, so they will need to be in the output argument list as well. E.g.,
E.g.,
% Put this in a file called DNLRX.m
function [uf,a,b,d1,...etc] = DNLRX(a,b,d1,...etc) % you fill this in for all the I/O variables
SIGN = @sign;
Min = @min;
B=0.0026;
r1=6.4527;r2=-4.6055;r3=-1.84030;r4=-1.8386;r5=-4.5868;r6=6.4215;
s1=1.0439;s2=14.3983;s3=0.0001;s4=-0.1792;s5=-14.3811;s6=-0.8453;s7=0.0018;
k1=SIGN(a-b+d1);
z1=abs(a-b+d1);
l1=Min(z1,0.0708);
d1=k1*l1;
k2=SIGN(a-b+d2);
z2=abs(a-b+d2);
l2=Min(z2,0.0737);
d2=k2*l2;
k3=SIGN(a-b+d3);
z3=abs(a-b+d3);
l3=Min(z3,54.1993);
d3=k3*l3;
k4=SIGN(a-b+d4);
z4=abs(a-b+d4);
l4=Min(z4,0.0353);
d4=k4*l4;
k5=SIGN(a-b+d5);
z5=abs(a-b+d5);
l5=Min(z5,0.0734);
d5=k5*l5;
k6=SIGN(a-b+d6);
z6=abs(a-b+d6);
l6=Min(z6,0.0779);
d6=k6*l6;
k7=SIGN(a-b+d7);
z7=abs(a-b+d7);
l7=Min(z7,1.0964);
d7=k7*l7;
x=s1*d1+s2*d2+s3*d3+s4*d4+s5*d5+s6*d6+s7*d7; //theta*delta
uf=r1*a+r2*b+r3*c+r4*d+r5*e+r6*f+x+B; //maxwell equation
f=e;
e=d;
d=c;
c=b;
b=a;
a=z; //z=new_postion
end
