A Model-Based Approach for Dual-Burst Processing of Laser Profile Velocimetry Signals

In order to characterize turbulent flows, correlation measurements are a common procedure. Laser Doppler sensors are often applied for this task, but in the case of high seeding concentrations, the so-called dual-burst signals occur. These are caused by multiple particles passing the measurement volume at the same time. In this paper, a model-based signal processing approach is presented which is able to evaluate such distorted signals. In order to assess the performance of dual-burst algorithms, the Cramér-Rao lower bound is calculated for the first time. A comparison of this new technique with conventional dual-burst algorithms reveals a systematic error which is at least seven times lower and even makes an evaluation of the complete signal in some cases possible in the first place. As a result, the total frequency estimation uncertainty of dual-burst signals, which is directly linked to the velocity uncertainty, is improved by a factor of five and more. Hence, this new technique enables the precise evaluation of correlation functions and instantaneous velocity gradients of complex flows.