Detailed Evaluation of Centroid Analysis for Extracting Brillouin Frequency Shift of Fiber Distributed Sensors

Performances of centroid analysis (CA) used for extracting Brillouin frequency shift (BFS) from noisy signals are studied in this paper. The variance of extracted BFS, i.e., the minimum detectable BFS, is deduced as a function of signal-to-noise ratio, frequency step, Brillouin linewidth, and the data window used in the analysis. It is found theoretically that both of the averaged BFS and its variance are susceptible to the deviation of data window center from real-Brillouin central frequency, termed data window deviation (DWD). The theoretically analyzed results are verified by experiments and simulation, showing good agreement with each other. Since the DWD occurs often for noisy signals, an iterative CA (ICA) is proposed and demonstrated to reduce this impact and improve the accuracy of CA. Compared with curve fitting methods (CFM) the CA has attractive features, such as extremely shorter time of processing. The CA is especially suitable for extracting BFS from complicated spectra observed often in sensors paved in fields, for which the usual CFM may yield wrong results.