engleski

Practical Consistency between Bit-Error and Block-Error Performance Metrics up to Application Layer QoS

Although the 3G (UMTS) systems performance is still described by both Bit-Error-Rate (BER) and Block-Error-Rate (BLER), the 4G (LTE) uses exclusively the BLER – also commonly referred to as Frame-Error-Rate (FER) or Packet-Error-Rate (PER) of the input data streams for both downlink and uplink, which is effectively the subframe error rate (where a subframe is considered to be in error if at least one of the bits in the subframe is detected incorrectly). By this way, BLER corresponds to the number of exhibited negative link-layer protocol acknowledgements relative to the total count of acknowledgements. However, such BLER estimation scheme heavily depends on the success rate of negative acknowledgements transport via return channel, and so does not deliver very trustful low values (e.g. around 10-5), which are requested by higher protocols (mostly TCP). Therefore, BER and/or the Symbol-Error-Rate (SER) are and will certainly be used by network equipment manufacturers during tests in R&D and production of network elements. With this regard, the so-called residual channel is determined by the intensity of errors that could not be corrected by physical layer, so delegating this task to higher protocol layers who therefore must be provided with a robust real-time prediction of BER over the residual channel. Fulfilling this requirement is important not only in R&D and manufacturing, but also in network operator's environment. However, the relation between bit and block error oriented performance metrics is not always a simple one, especially in a non-standard environment. Specifically, as a practical example, in this paper, both performance metrics of an operational microwave point-to-point link (as a residual channel) was investigated by comparing the obtained „native“ residual BER values to the ones estimated by means of the analogous block-oriented performance metrics – the so called „background“ BLER. High consistency that was found in this particular example was complemented at the application layer by conducting end-to-end perceptual speech quality measurements (PSQM).

BER; BLER; residual channel;; performance metrics

nije evidentirano