Physical layer capture (PLC) in 802.11b refers to the successful reception
of the stronger (higher signal strength at receiver) frame in a collision.
PLC causes significant imbalance in the throughputs of sources. Existing
802.11b simulators, including NS-2 and Qualnet, assume that PLC occurs
only if the reception of stronger frame starts first at the receiver.  We
show empirically that in reality PLC occurs even if the stronger frame
arrives later (but within the physical layer preamble of the first frame).
We have modified the NS-2 simulator to account for this and Qualnet will
be incorporating a fix in their next release.  Simulations using the
latter PLC model generate significantly (up to 15\%) higher, and more
realistic, throughput unfairness than with the former PLC model.  We also 
show that time-instants of start of colliding frames routinely differ by  
as much as 20 $\mu s$ due to inherent uncertainties of 802.11b firmware 
clock synchronization and rx/tx turnaround delays, and that the frame to
arrive first can be either the stronger or the weaker with equal
likelihood. To identify which frames were involved in collisions, when
their transmissions started, and which of them were retrieved, we have
devised a novel technique using multiple sniffers and instrumented device
drivers to reconstruct from the air interface all tx/rx events in a WLAN
to within 4 $\mu s$ accuracy.  This allows us to quantify the causal links
from the PHY layer through the MAC layer to the observed application layer
imbalance, which may be as high as \textbf{25 \%} with two sources and    
\textbf{ 75 \%} with four sources as seen in experiments spanning ad-hoc
and infrastructure modes with both UDP and TCP workloads.
UMIACS-TR-2004-26