Chuong 11.html

ADSL, VDSL, and Multicarrier Modulation . John A. C. Bingham Copyright # 2000 John Wiley & Sons, Inc. Print ISBN 0-471-29099-8 Electronic ISBN 0-471-20072-7

11

FUTURE IMPROVEMENTS

Until now we have discussed mostly what can be done and what has been done. In, this chapter we discuss what might be done. If some of the ideas are proven to be useful, it should be possible to incorporate them seamlessly¹into G.992 and G.VDSLvia the option exchange de®ned in G.994 (handshake protocol for DSL modems).

CAVEAT: The reader is advised that most of the ideas in this chapter have, for several reasonsÐlack of time being the main oneÐnot been subjected to peer review. They are offered therefore only as food for thought, with no claims made as to their nutritional value.

11.1 FREQUENCY-DOMAIN PARTIAL RESPONSE

In Section 7.1.1 we saw how frequency-domain partial response (FDPR) with a (1 À Á) correlation as de®ned by equation (7.1) reduces the sidelobes of the associated (I)FFT; it may also have other useful effects, depending on whether it is introduced in the transmitter or the receiver.

11.1.1 FDPR in the Transmitter

With a (1 À Á) correlation, each symbol of the transmit signal has a sinusoidal envelope as shown by equation (7.3) and has much reduced sidelobes as shown in Figure 7.2. It can be used with a quiet guard period (not a cyclic pre®x) of samples, and just as for an unshaped envelope, only IR terms beyond the range of the guard period (i.e., h_ifor i>) contribute to the distortion.

Effects of Distortion. The calculations in Section 6.2.1 and particularly equation (6.11) are based on the assumption that all transmit samples have equal average energy. If, however, the envelope is shaped as shown by (7.3), the same

¹See [Krechmer, 1996] for a discussion of etiquettes and protocols to accommodate innovations. 239
method of analysis can be used to show that the general form of (6.11) is modi®ed to
X jH₀j²‡jH₁j² ˆ 2 c²h² …11:1†
iˆ1
where, as before, c_i ˆ 0 for i4 but now
X₂