SMIQ
Bit Error Rate Test
1125.5555.03
2.375
E-9
BER measurement
with interrupted
random sequence -
integrating BER
measurement
Due to the type of data, oversampling and the finite memory length of the SMIQ
(option SMIQB60) or the AMIQ, it may happen that the generated random
sequence is not cyclically repeated at the memory wrap-around but has a break
at this point. In a normal BER measurement operating exclusively with CLOCK
and DATA signals, a break of this kind would lead to approx. 50% faulty bits due
to the loss of synchronization.
A random sequence with a break can be processed by means of the
integrating
BER measurement,
it is switched on in the menu under EXTERNAL RESTART
(or by the remote-control command
BERT:SETup:RESTart EXTern)
. The
BER measurement is to be stopped in time and then restarted at the beginning
of the data sequence. The measurement is stopped and started via a signal at
the RES input (pin 9 of SUB-D connector): A transition from logic 0 to 1 causes
a partial result to be generated and the measurement to be stopped. A transition
from 1 to 0 starts the measurement for the next sub-interval. This measurement
is synchronized anew.
This input is best connected to an AMIQ marker channel or the SMIQ trigger
output, with a single 1 encoded both at the start and at the end of the data
sequence. This causes the BER measurement to stop briefly at memory wrap-
around (the break) and start again. The partial results are integrated.
When the signals are interrupted by other data (eg preambles), these data will
normally cause bit errors. The BER measurement for these data can then be
interrupted using the data enable input .
In the integrating BER measurement, partial BER results are added up
(controlled by a signal at the RES input) until the predefined total number of data
or error bits is attained or exceeded.
Note:
Since the test hardware can be programmed flexibly, other bit error
measurement techniques such as comparison with the output pattern, masking
of certain time and data ranges, are also possible. In this matter, please contact
your nearest R&S representative.
PRBS Polynomials
For generating and testing the PRBS, a shift register with feedback is used. The feedback depends on
the type of polynomial used. The sequence length of a generator is the result of
2
n
- 1,
n being the
degree of the polynomial.
EXOR
EXOR
EXOR
Fig. 2-236
PRBS polynomials
