06.06.2022 Views

B. P. Lathi, Zhi Ding - Modern Digital and Analog Communication Systems-Oxford University Press (2009)

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

7.3 Pulse Shaping 353

In fact, many physical channels such as magnetic recording have a zero gain at de. Therefore,

it makes no sense for the baseband signal to have any de component in its PSD. Modified

partial-response signaling is often adopted to force a null at de. One notable example is the

so-called modified duobinary signaling that requires

n = - 1

n = l

for all other integers n

(7.41)

A similar argument indicates that P c (t) can be generated from any pulse P a (t) satisfying the

first Nyquist criterion via

Equivalently, in the frequency domain, the duobinary pulse is

which uses sin (2nfI' b ) to force a null at de to comply with the physical channel constraint.

7.3.6 Detection of Duobinary Signaling

and Differential Encoding

For the controlled ISI method of duobinary signaling, Fig. 7. 17 shows the basic transmitter

diagram. We now take a closer look at the relationship of all the data symbols at the baseband

and the detection procedure. For binary message bit h = 0, or 1, the polar symbols are simply

Under the controlled ISI, the samples of the transmission signal y(t) are

(7.42)

The question for the receiver is how to detect h from y(kTb) or b k . This question can be

answered by first considering all the possible values of b k or y(kTb), Because a k = ±I, then

b k = 0, ±2. From Eq. (7.42), it is evident that

bk = 2 =} ak = I or h = 1

bk = -2 =} ak = - 1 or h = 0 (7.43)

bk = 0 =} ak = -ak- 1 or h = 1 - h-1

Fi g ure 7.17

Equivalent

duobinary

signaling.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!