Improving Detection of Changepoints in Short and Noisy Time Series with Local Correlations: Connecting the Events in Pixel Neighbourhoods

This paper determines how to define a discretely implemented Fourier transform when analysing an observed spatial point process. To develop this we answer four questions; first what is the natural definition of transform, and are its spectral moments, second calculate fourth order moments using Campbell's theorem. Third determine implement tapering, important component for analysis other stochastic processes. Fourth question produce isotropic representation basic properties