Abstract: A procedure for information transmission utilizing a system for reception of data signals using clock signals in which the phase shift between clock signals and data signals is calculated using a formulation of the gradient t.sub.n+1 =t.sub.n +p/q-.alpha.e.sub.n (t.sub.n)e'.sub.n (t.sub.n). The phase shift signal (t.sub.n+1) is calculated for an (n+1) data signal, the phase shift signal (t.sub.n) is calculated for a (.sub.n) data signal, and (p/q) in the ratio of the clock period to the data period. In addition, e.sub.n (t.sub.n) is the error between a real sample and an assumed theoretical value and e'.sub.n (t.sub.n) is the derivative of e.sub.n (t.sub.n) with respect to (t.sub.n).