Page 435 - Complete Wireless Design

Wireless Issues

434 Chapter Ten

a receiver with an “IF” after the first mixer stage, that is fixed at or near zero
frequency. The RF input signal has been mixed down immediately to base-
band, with the middle of the desired band translated to this zero frequency.
DCRs have a much lower parts count, and are thus cheaper to build, than
the competing superheterodyne designs. DCRs do not require an image filter
or high-frequency IF filters and amplifiers since no image frequency is seen by
the DC receiver, and it has no IF. However, DCRs have multiple problems that
make a discrete DCR almost impossible, while even RFIC designs are rife with
difficulties. Nonetheless, many of the DCR’s limitations can, and have been,
addressed relatively successfully within the domain of some of the newer
RFICs, especially by companies such as Analog Devices and Maxim.

10.6.2 Direct-conversion issues
Most design choices have tradeoffs, and a DCR is no different.
Superheterodyne receivers will have more selectivity and sensitivity than an
equivalent direct-conversion receiver, with most DC receivers barely able to
function up to 900 MHz while attaining only 95 dBm sensitivity ( 105 dBm
is a requirement in many systems). And selectivity naturally suffers in most
RFIC designs because of the adoption of active low-pass filtering at baseband,
so DCRs have less interference rejection than superheterodynes. Many zero-
IF receiver RFICs, as well, have the following problems:
1. An incidental offset voltage is caused by the self-mixing in the direct con-
version receiver; the LO leakage actually mixes with the original LO signal,
creating a DC voltage that can contaminate the signal of interest, lowering
the SNR, and can even saturate the baseband amplifier stages.
2. LO leakage through the RF sections is a large consideration in design, since
the LO is very close to the frequency of the incoming RF and can be radiated
by the receiver’s antenna, causing in-band interference.
3. Flicker-effect (1/f) noise from the mixer output can be a problem, as the
down-converted signal is usually of low amplitude and low frequency (near
0 Hz), decreasing SNR.
4. In very wideband and high frequency DCR receivers, the difficulty of main-
taining equal amplitude and phase in both the I and Q legs, called I and Q
mismatch, has been difficult to solve. This will cause an increase in the BER.
5. Any circuit that employs phase shifts of the incoming signal by 90 degrees,
as all modern zero-IF receivers must do, undergoes compromises with
noise, linearity, and power.
6. Zero-IF receivers permit distortion, caused by strong signals at the mixer,
to reduce sensitivity more quickly than with superheterodynes.

Up until quite recently, the above problems have relegated zero-IF
receivers to FSK pagers and a few amateur SSB receivers. But companies
have attempted to get around many of these difficulties by some ingenious

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.