When determining the error probability, the number of bits, characters or data packets transmitted with errors is related to the total number of units transmitted. The error probability is represented by a ratio of these two quantities.
For example, an error probability of `5*10^-6` means that there are 5 faulty units for every million (`10^6`) units transmitted. Since the error probability depends on the size of the data packets, it is possible to increase the error probability by reducing the size of the data packets. The error probability can be determined empirically from the data packet size or the payload.