mirror of
https://github.com/saitohirga/WSJT-X.git
synced 2024-08-24 15:16:36 -04:00
svn+ssh://svn.berlios.de/svnroot/repos/wsjt/WSJT/branches/linux merged into svn+ssh://svn.berlios.de/svnroot/repos/wsjt/trunk git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/trunk@155 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
57 lines
1.6 KiB
Fortran
57 lines
1.6 KiB
Fortran
subroutine syncf0(data,jz,NFreeze,NTol,jstart,f0,smax)
|
|
|
|
C Does 512-pt FFTs of data with 256-pt step size.
|
|
C Finds sync tone and determines aproximate values for jstart and f0.
|
|
|
|
real data(jz) !Raw data
|
|
real s2(128,6) !Average spectra at half-symbol spacings
|
|
real x(512)
|
|
complex cx(0:511)
|
|
complex z
|
|
equivalence (x,cx)
|
|
|
|
ps(z)=real(z)**2 + aimag(z)**2 !Power spectrum function
|
|
|
|
call zero(s2,6*128) !Clear average
|
|
df=11025./512.
|
|
|
|
ia=(f0-400)/df
|
|
ib=(f0+400)/df + 0.999
|
|
if(NFreeze.eq.1) then
|
|
ia=(f0-NTol)/df
|
|
ib=(f0+Ntol)/df + 0.999
|
|
endif
|
|
|
|
C Most of the time in this routine is in this loop.
|
|
|
|
nblk=jz/256 - 6
|
|
do n=1,nblk !Accumulate avg spectrum for
|
|
j=256*(n-1)+1 !512-pt blocks, stepping by 256
|
|
call move(data(j),x,512)
|
|
call xfft(x,512)
|
|
do i=ia,ib
|
|
x(i)=ps(cx(i))
|
|
enddo
|
|
k=mod(n-1,6)+1
|
|
call add(s2(ia,k),x(ia),s2(ia,k),ib-ia+1) !Average at each step
|
|
enddo
|
|
|
|
C Look for best spectral peak, using the "sync off" phases as reference.
|
|
smax=0.
|
|
do i=ia,ib
|
|
do k=1,6
|
|
k1=mod(k+1,6)+1
|
|
k2=mod(k+3,6)+1
|
|
r=0.5*(s2(i,k1)+s2(i,k2))
|
|
s=s2(i,k)/r
|
|
if(s.gt.smax) then
|
|
smax=s
|
|
jstart=(k-1)*256 + 1 !Best starting place for sync
|
|
f0=i*df !Best sync frequency
|
|
endif
|
|
enddo
|
|
enddo
|
|
|
|
return
|
|
end
|