System and method for detection of noise in sparse data sets with edge-corrected measurements

摘要

A signal processing system and method includes a transducer for receiving an analog signal having a random component and possibly an information component. The analog signal is converted into a digital signal having sample points. A nearest-neighbor calculation component calculates the expected average nearest neighbor distance between the sample points, the actual average distance and an error value. These values are corrected for edge effects. A first randomness assessment compares the actual average distance against the expected average distance with the standard error value. A second randomness assessment compares actual repeated values in the digital signal against expected repeated values. An information processing system continues processing the signal if the assessments indicate the possibility of an information component.

主权项

1. A signal processing system comprising:
transducer means for receiving an analog signal, the analog signal including a noise component and possibly also an information component, and generating in response a digital signal which is represented by a plurality of sample points distributed over a selected region; a nearest-neighbor distance calculation component joined to receive the digital signal and capable of calculating an actual average nearest neighbor distance between said plurality of sample points without edge effects, said nearest-neighbor distance calculation component being capable of calculating an expected average nearest neighbor distance between said plurality of sample points without edge effects; a standard error value calculation component joined to receive the digital signal and capable of calculating a standard error value without edge effects for the expected average nearest neighbor distance; a nearest-neighbor test component joined to receive the actual average nearest neighbor distance, the expected average nearest neighbor distance and the standard error value, said nearest-neighbor test component capable of generating a first random noise assessment that the digital signal comprises solely random noise when a comparison of the actual average nearest neighbor distance against the expected average nearest neighbor distance and the standard error value indicates that the actual average nearest neighbor distance is within the standard error value of the expected average nearest neighbor distance; a runs test component joined to receive the digital signal and capable of generating a second randomness assessment based on a comparison of actual repeated values in said digital signal against expected values in said digital signal, a similarity in actual repeated values to expected values providing a second random noise assessment that the digital signal comprises solely random noise; and an information processing sub-system joined to receive the digital signal, the first random noise assessment and the second random noise assessment, said information processing subsystem processing the data signal to extract said information component if at least one of the first random noise assessment and the second random noise assessment indicates that the digital signal does not comprise solely random noise.