If we generate this error signal, we get the
following (neglecting the noise terms):
(7.120)
(7.121)
(7.122)
In an AWGN channel, then ideally, h(k) would be a constant and not influence the error signal over
time. However, for time-varying signals, the channel estimate influence should be removed. The common
technique is to use a noncoherent DLL, where the late and early powers are determined to obtain
the error signal. In this case, we have the following error signal, neglecting noise cross terms as
(7.123)
The channel power can be removed or normalized by obtaining the channel power from the on-time
correlator output. The LPF can be replaced by an accumulator with appropriate decisions functions.
In Fig. 7.83, the top correlator is demodulating the late timing instance, while the lower correlator
is demodulating the early timing instance. The despread waveforms are squared to obtain the symbol
energies or powers to be used to calculate the error signal. The error signal computes the late
minus early correlation powers, which then become low-pass filtered to reduce the results of smaller
integration output values, if used. This low-pass filtered error signal will be used to either speed up
or slow down the PN generator clock.
Pages:
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678