A technique has been developed whereby an adaptive receiver may employ decision
updating in a manner insensitive to RBS. One realization achieves nearly-ideal
training of an adaptive equalizer of a modem during TRN1d training despite
potential RBS in a digital portion of the PSTN. Updates in the exemplary realization
are based upon the true value of the corresponding equalizer output (the decision)
and insensitive to RBS. Adaptive equalizer realizations improve equalizer training
by treating the received signal as a sequence of blocks of 24 symbols with 24 corresponding
separate decision values. In an exemplary variation, equalizer training begins
using a single pair of decision points for coefficient updating. After gross convergence
of the equalizer coefficients is achieved the update broadens to include updating
each of 24 decision points and the equalizer coefficients.