A Note about Link Budgets and Power

We have talked about the need to deliver link budget gain by using various techniques
(for example, smart antennas). Another way of improving link budget is to move
nearer to the base station.
Note that in GSM, the biggest cell radius supported is 35 km. In W-CDMA (3GPP1),
the maximum cell radius is 20 km, and the maximum data rate is 144 kbps. Amicrocell
is defined as having a coverage radius of 300 meters, a picocell, a coverage radius of
100 meters.
In either a macrocell, microcell, or picocell, as you move closer to the Node B (base
station), you have more power available both on the uplink and downlink to support
higher data rates. The dynamic range in a macrocell is typically 70 dB (the difference
between being very close to the cell and right at the cell edge). However, you are also
trying to follow the fast fading envelope, which means tracking short-term fades of 20
dB or more, so it could be argued the usable dynamic range is typically 50 dB. Nevertheless,
this is still a huge difference in link budget. As we move closer to the base station,
we can either reduce coding overhead (that is, increase user data rate) or increase
modulation complexity—for example, move to 8-level PSK or 16-level QAM in a
3GPP2 air interface (CDMA 2000) to take advantage of the increase in available power.
The problem—which is pretty unavoidable—is that real throughput will vary
depending on the user’s distance from the base station, which makes it difficult to
deliver a consistent user experience. The general principle, then, is just to provide as
good a link budget as we can, given the power and cost constraints we have to meet.
In Chapter 4, which dealt with handset hardware, we pointed out that battery bandwidth
determines the amount of offered uplink delivery bandwidth available. We have
a certain amount of instantaneous RF power available, for instance, either 125 mW for
a Class 4 handset and 250 mW for a Class 3 handset (determined by regulatory/standards
groups), and a certain amount of peak power deliverable from the battery.

We then have a certain amount of battery capacity (600, 800, 1200, 1800 milliamp/
hours for example), which determines how much data we can send from the handset
before the battery goes flat.
The closer we are to the base station, the higher the peak data rate available to us
and the more data we can send. (This means we can use the power available to send
data rather than to overcome the 60 or 70 dB of path loss if we are at the cell edge.)
Uplink offered traffic loading is therefore a function of user geometry—how close
users are to the base station/Node B, which means user proximity to the Node B determines
uplink bandwidth. Another way of looking at this is that revenues increase as
network density increases.
Link gain products can be justified on a similar basis. By delivering downlink directivity,
more downlink bandwidth can be delivered. Billable bandwidth increases in
both directions. Sometimes, investing in link gain products, such as sectored or smart
antennas, also gives us parallel side benefits in terms of other services. An example is
being able to use sectored antennas or smart antennas to provide location and positioning
information. 310