6.1, where an attempt to provide an outline has been made.
6.1.1 Linear Equalizer
The equalizer structure shown in Fig. 6.2 is a Fractionally Spaced Equalizer (FSE), where the delay
elements between the filter coefficients are a fraction of a baud time, T, apart [1]. This type of equalizer
has been shown to be more robust against symbol-timing recovery errors. If the delay elements
spacing are T seconds apart, then the equalizer is called a Symbol-Spaced Equalizer (SSE) or commonly
called T-spaced equalization. The equalizer shown spans for T seconds, depending on the
amount of ISI you wish to reduce; the equalizer temporal duration can increase.
The received signal, r(k), enters the adaptive filter (shown below in FIR form) whose responsibility
is to remove the ISI present in the signal. The adaptive filter output, x(k), enters a decision device
to produce a detected symbol x??™(k). There are two modes of operation: the first mode is called
293
decision directed mode while the second is called the training mode. In most wireless communication
systems there is some information known a priori to the receiver such as pilot bits or a pilot channel
in the case of WCDMA. This information is typically used to train the coefficients of the adaptive
filter.
Pages:
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488