Higher order crossing analysis applied to signal detection and evaluation of radio channels

A real-time channel evaluation (RTCE) method developed by the authors has been presented in previous papers, which relied on the counting of the zerocrossings of a sampled waveform. They now look at an extension of this work to higher order crossings (HOX) with the aim of increasing the accuracy of the channel evaluation and provide a means of signal detection. This work has been developed as a result of some papers by Kedem (1984, 1986) in which he discusses the uses of HOX to characterise a signal. In these papers, Kedem takes a sampled waveform, or time-series as he terms it, and counts the number of zerocrossings. He then takes the differences and sums of successive samples to generate two new time series and counts the zerocrossings of them. This procedure is repeated many times in order to generate many series and many counts of HOX. These sets of HOX are then used to characterise the original series in terms of sinusoidal components and noise. In this paper, the extra information that can be gained from looking at these different orders of zerocrossings is used to provide a measure of the channel signal-to-noise ratio (SNR) and hence expected error rates