Digital Signal Processing Reference
In-Depth Information
Fig. 6.5
Typical indoor
pathloss model
frequency. This corresponds to a coherence time of 166 ms for 802.11a networks.
A set of 1000 channel frequency response realizations (sampled every 2 ms over
one minute) were generated and normalized in power. Data was encoded using a
turbo coder model [77] and the bit stream was modulated using 802.11a OFDM
specifications. For a given back-off
b
and transmit power
P
Tx
, the SINAD at the
receiver antenna was computed for the channel realization
A(t, f )
by Eq.
6.2
.
We assume an average path-loss
A
of 80 dB at a distance of 10 m, which consists
both of the transmit antenna gain
G
t
, receive antenna gain
G
r
and the propagation
loss. This path-loss is from an empirical model developed at IMEC (Fig.
6.5
). It
is representative for indoor channels and results in pathlosses in between the
Free
space
model and
Two ray ground
model, which is commonly used for outdoor ap-
plications.
From the channel realization database, a one-to-one mapping of SINAD to re-
ceive block error rate was determined for each modulation and code rate. The chan-
nel is then classified into 8 classes, determined by a 2 dB difference in the receive
SINAD that is measured for a turbo code BlER of 10
−
3
(Fig.
6.6
(a)). We use a simi-
lar 2 dB discrete step for the PA knobs (Table
6.1
). In order to derive a time-varying
link-layer error model, we associate each channel class to a Markov state, each with
a probability of occurrence based on the channel realizations database (Fig.
6.6
(b)).
Given this eight-state error model, we are able to model the PER, expressing the
performance per MAC layer packet
L
frag
, for different configurations. The PER is
obtained in Eq.
6.4
by assuming the block errors are uncorrelated for a packet size
of
L
frag
bits and a block size of 288 bits:
P
e
=
1
.
L
frag
288
−
(
1
−
BlER
)
(6.4)
This calibration is not a fundamentally new characterization of the hardware.
This BlER-SINAD information just has to be combined with a set of channel states
that are extracted on-line. The combination of this run-time channel with hardware
properties into metrics for higher layer metrics of the database is a key enabler.
