(RP11) HPC Oriented Algorithm for Computation of Recurrence Quantitative Analysis
TimeTuesday, June 18th8:30am - 10am
DescriptionRecurrence quantitative analysis (RQA) is a quantification of the dynamical properties of time series. This method is used in many medicine, biology, speech and vocalization research. The main drawback of this method is its computational complexity. The first step for the RQA is the computation of the distance matrix of subsequences of the time series. Afterwards, this distance matrix is thresholded, setting 1 if the distance is less than a parameter ε and 0 otherwise. These are called recurrence plots. The histogram of consecutive 1s on the diagonals is computed. This histogram is then used to compute RQA. We present an algorithm for the computation of RQA directly from the input data. This algorithm allows easy parallelization of the computation with minimal spatial complexity.