Tone Detection Via Incoherent Averaging of Fourier Transforms to Support the Automated Spacecraft-Monitoring Concept

An algorithm is presented for detection of very weak spacecraft tones generated by an onboard auxiliary oscillator with stability speciflcations similar to those of the small deep space transponder. Signal power is evaluated via discrete Fourier transforms, and detection is determined by comparing the measured power with a predetermined threshold. Limited oscillator stability precludes coherent integration of the signal over time scales longer than a few seconds; thus, the Fourier transform is performed independently on successive segments of data. Resulting power spectra are then averaged; potential frequency drift is accounted for by shifting the spectra when forming the average. The detection scheme is well suited to the proposed automated spacecraft monitoring system, in which a spacecraft sends a carrier signal modulated with one of four subcarrier frequencies to indicate the spacecraft’s current state. Analysis based on theory and measurements indicates that reliable detection in a 1000-s interval can be achieved at a power signal-to-noise-density level, P=N0 ,a s low asi1 dB-Hz.