Page 107 - Antennas for Base Stations in Wireless Communications

80 Chapter Two

the received C/I ratio, beam-steering is augmented by the use of null-
forming methods by which unwanted signals are suppressed and inter-
ference radiated toward the sources of interfering signals is reduced at
the same time. In these systems, beams can be switched, and radiation
patterns adapted on a user-by-user, burst-by-burst basis. Currently,
this adaptivity is available only in the azimuth patterns, the elevation
patterns being those of a conventional fixed array. The application of
adaptive beam-shaping in the elevation plane adds considerable cost
and complexity and may not be justifiable, but RET techniques could
easily be applied.
The use of a single-carrier TDD system greatly reduces the specifica-
tion requirements for PIM levels, so some lower-cost methods, materials,
and components can be used without penalty; for example, premium
PCB materials are not needed and Type-N connectors are entirely ade-
quate for this application.

2.3.11 Measurement Techniques
for Base Station Antennas
Base station antennas present a number of measurement challenges
because of their closely specified radiation patterns and PIM perfor-
mance, as well as the very large volumes of product for which conform-
ing performance must be assured.

2.3.11.1 Radiation Pattern Measurements Far-field measurements will
generally be made within the service area of one or more local networks.
For this reason the rotating antenna under test (AUT) should always
receive the signal radiated by a fixed illuminating antenna that should
be aimed in a direction in which it is least likely to cause interference
to any network user.
Accurate far-field measurements depend on the availability of a clear
and open test range. Because we are interested in the levels of nulls
and sidelobes at levels around −20 dB relative to the main-beam maxi-
mum, the traditional criterion for the necessary range length is inad-
2
equate, and a range length of at least 4d /l is needed. (At a range of
2
2d /l, there is a parabolic phase error of 45° at the ends of the array
relative to the center—if you doubt that this is too short, try applying
that phase error to a set of typical array currents.) The aperture occu-
pied by the AUT should be probed to establish that the amplitude and
phase of the illuminating signal are constant to acceptable limits. Time
gating methods can be used to offset the effects of site reflections, but
implementing these tends to slow down the measurement process. The
azimuth pattern of a base station antenna must be measured at the
elevation angle equal to the nominal beamtilt. This requires the use of