Sequential detection of a temporary change in multivariate time series

In this work, we aim to provide a new and efficient recursive detection method for temporarily monitored signals. Motivated by the case of propagation an event over field sensors, assumed that change in statistical properties signals can only be temporary. Unfortunately, our best knowledge, existing simple techniques such as ones based on cumulative sum (CUSUM) do not consider temporary aspect multivariate time series. paper, propose novel sequential algorithm, named Temporary-Event-CUSUM (TE-CUSUM). By combining with adaptive way aggregate local CUSUM variables from each data stream, empirically show TE-CUSUM has very good rate passing through sensors noisy environment.