Example Application

Many signal processing techniques apply the batch-mode FFT to a signal, such as an audio signal, manipulate it in relatively simple ways, and then reverse the FFT

One shortcoming of this method is that much of the information from short-lived, transient signals and long, complex signals is lost because the input signal is divided into batches, and the batch boundaries are not usually at the best position with respect to the signal. Sliding signal processing manipulates transient signals with no loss of fidelity and without the signal artifacts characteristic of batch processing. With long, extended signals, there is never an abrupt break between two adjacent pieces of the signal. These special characteristics of sliding signal processing have many applications.

For example, a sliding system in an airborne radar measures aircraft maneuvers more accurately. The continual stream of signal processor updates improves the predictive capability of the radar computer. Integrating these updates also provides some extra measure of noise cancellation.

The sliding processor has immediate application in processing extended radar signals, sending encrypted messages over a communications channel, MPEG compression and speech processing, to name a few. In short, it removes many of the processing bottlenecks involved in transmitting and receiving high-fidelity, information-carrying signals.