Page 385 - Complete Wireless Design

Communications System Design

384 Chapter Nine

oscillator from the mixer ports to minimize VSWR, and thus mixer-generat-
ed IMD, and to increase the oscillator’s output power to the nominal levels
to maintain the mixer’s NF and conversion losses. In design of this LO chain,
the LO amplifiers can be overdriven for maximum power output and flat-
ness, since the LO signal involves only a single frequency. This means that
no IMD will be generated, only harmonics, which are easily filtered out.
Both BPF3 and BPF4 are employed to reduce wideband noise, harmonics,
and possibly any subharmonics that will be present in the oscillator’s outputs.
Reduction of generated wideband noise improves the mixer’s NF and the
receiver’s sensitivity, while harmonic suppression prevents a decrease in the
mixer second-order intercept point (IP2) and the subsequent generation of
increased LO second harmonics.
The first IF chain of the receiver will furnish some gain and filtering (cer-
tain first IFs may supply only filtering), along with delayed AGC of any ampli-
fiers present, before injection into MIXER2. If desired, this first IF filtering of
the DIPLEXER and BPF6 will also remove the second image frequency from
reaching MIXER2, but the filtering itself must not be so tight as to introduce
excessive group delay variations, causing increased BER. The filtering is pres-
ent mainly to provide some channel selectivity and the rejection of spurious
signals, since image noise into MIXER2 is not a concern if relatively high gain
stages are adopted in the receiver’s front end, which is where the NF is set (the
noise floor at MIXER2 will be mainly noise amplified by the LNA).
The second IF amplifiers and filter stages supply most of the receiver’s gain
and selectivity, since lower-frequency circuits are cheaper and more stable.
However, any such high-gain IF stages should be shielded against EMI that may
have been emitted from other internal stages, or from external emissions, in
order to prevent amplifier oscillations and receiver interference. The second IF
strip will also increase the receiver’s total dynamic range by the use of variable-
gain amplifiers (VGAs) within an AGC loop. This automatic gain control will not
only control the gain of the second IF stages but, with many high-end receivers,
it will also set the gain of the first IF strip and the LNA. The second IF strip will
output f into an external modem, or into an internal detector stage.
OUT
Considerations. When deciding on the number of filter poles required for a
receiver’s (or transmitter’s) RF or IF, calculate the amplitude of all undesired
signals (such as adjacent channels, mixer products, and LO feedthrough) and
how much they must be attenuated to meet specs. A program such as Kirt
Blattenberger’s RF Workbench, and to a certain extent The Engineer’s Club’s
MxrSpur, can be invaluable in this regard. Run simulations for the chosen fil-
ter type (Butterworth, Chebychev, etc.), along with the preliminary number of
poles (3, 5, 7, etc.). Add some extra attenuation margin (approximately 10 per-
cent) to this figure. Confirm that the filters will now properly attenuate all
undesired signals, without adding excessive group delay variation, insertion
loss, sideband cutting, cost, or ripple to the design.
To discover the receiver’s AGC range needed for the IF strip, and in some
cases for the RF stages, find the difference in dB between the lowest RF signal
expected that will still be able to supply the desired amplitude at the receiver’s

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.