5 MC simulation system.
This method requires knowledge of the transmitted bit stream. This method makes no ???a priori???
assumptions of the received signal PDF and, as such, is the most general of all the estimation techniques.
The price we must pay for this luxury is computational complexity. This is related to the number
of observations (N) that must be processed to get an acceptable level of confidence in the resulting
BER. Recall the earlier confidence interval guidelines, where a 10v BER for a 90% confidence interval
and N 10v3 produced low spread in the estimated values. For example, consider a target of
107 BER; this means we need to observe 1010 bits to obtain such statistical confidence. This is, in
general, a highly undesirable drawback with this technique. An additional point worth mentioning is
the actual complexity of the system simulation. Depending on the level of algorithmic abstraction, the
simulation can contain very low level and complete details about the target implementation, receiver
algorithms, technology assumptions, and so forth. If the system complexity is large, then the simulation
can take a considerable amount of time to completely analyze the number of bits desired to obtain
statistical confidence in the results.
Pages:
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754