A Methodological Approach for Estimating Protocol Analyzer Instrumental Measurement Uncertainty in Packet Jitter Evaluation

Packet jitter is a critical parameter to be accurately measured and monitored in communication and computer networks for quality-of-service assessment, particularly in Voice over Internet Protocol (VoIP) and real-time video applications. To this aim, suitable measurement systems, also known as “protocol analyzers,” are used. In particular, special-purpose-architecture-based protocol analyzer (SPAPA) and general-purpose-architecture-based protocol analyzer (GPAPA) are the most widespread measurement solutions. In both cases, the measurement chain involves a reference data traffic generator and one or more nodes, suitably displaced at the measurement points of interest in the network. With special regard to the instrumental measurement uncertainty of both SPAPAs and GPAPAs already present on the market, poor information is given by manufacturers. Moreover, the analysis in the literature concerning communication and computer network test and measurement clearly evidences the lack of important details on this key matter and, consequently, the need of further scientific investigation. This paper proposes a methodological approach to evaluate SPAPA and GPAPA instrumental uncertainty in packet jitter measurements. A general model accounting for the effect of the main quantities of influence is proposed, which allows quantifying the associated uncertainty components through a statistical analysis. A number of experiments on an actual GPAPA show the suitability and effectiveness of the proposal.