SUDMOFKCW - converted-wave DMO via F-K domain (log-stretch) method for
 		common-offset gathers					

 sudmofkcw <stdin >stdout cdpmin= cdpmax= dxcdp= noffmix= [...]	

 Required Parameters:							
 cdpmin		  minimum cdp (integer number) for which to apply DMO
 cdpmax		  maximum cdp (integer number) for which to apply DMO
 dxcdp		   distance between adjacent cdp bins (m)		
 noffmix		 number of offsets to mix (see notes)		

 Optional Parameters:							
 tdmo=0.0		times corresponding to rms velocities in vdmo (s)
 vdmo=1500.0		rms velocities corresponding to times in tdmo (m/s)
 gamma=0.5		 velocity ratio, upgoing/downgoing		
 ntable=1000		 number of tabulated z/h and b/h (see notes)	
 sdmo=1.0		DMO stretch factor; try 0.6 for typical v(z)	
 flip=0		 =1 for negative shifts and exchanging s1 and s2
 			 (see notes below)				
 fmax=0.5/dt		maximum frequency in input traces (Hz)		
 verbose=0		=1 for diagnostic print				

 Notes:								
 Input traces should be sorted into common-offset gathers.  One common-
 offset gather ends and another begins when the offset field of the trace
 headers changes.							

 The cdp field of the input trace headers must be the cdp bin NUMBER, NOT
 the cdp location expressed in units of meters or feet.		

 The number of offsets to mix (noffmix) should typically equal the ratio of
 the shotpoint spacing to the cdp spacing.  This choice ensures that every
 cdp will be represented in each offset mix.  Traces in each mix will	
 contribute through DMO to other traces in adjacent cdps within that mix.

 The tdmo and vdmo arrays specify a velocity function of time that is	
 used to implement a first-order correction for depth-variable velocity.
 The times in tdmo must be monotonically increasing. The velocity function
 is assumed to have been gotten by traditional NMO. 			

 For each offset, the minimum time at which a non-zero sample exists is
 used to determine a mute time.  Output samples for times earlier than this", 
 mute time will be zeroed.  Computation time may be significantly reduced
 if the input traces are zeroed (muted) for early times at large offsets.

 z/h is horizontal-reflector depth normalized to half source-reciver offset
 h.  Normalized shift of conversion point is b/h.  The code now does not
 support signed offsets, therefore it is recommended that only end-on data,
 not split-spread, be used as input (of course after being sorted into	
 common-offset gathers).  z/h vs b/h depends on gamma (see Alfaraj's Ph.D.
 thesis, 1993).							

 Flip factor = 1 implies positive shift of traces (in the increasing CDP
 bin number direction).  When processing split-spread data, for example,
 if one side of the spread is processed with flip=0, then the other side
 of the spread should be processed with flip=1.  The flip factor also	
 determines the actions of the factors s1 and s2, i.e., stretching or	
 squeezing.								

 Trace header fields accessed:  nt, dt, delrt, offset, cdp.		


 Credits:
	CWP: Mohamed Alfaraj
		Dave Hale

 Technical Reference:
	Transformation to zero offset for mode-converted waves
	Mohammed Alfaraj, Ph.D. thesis, 1993, Colorado School of Mines

	Dip-Moveout Processing - SEG Course Notes
	Dave Hale, 1988