The near-infrared system is usually operated with a high speed detector and preamplifier to search for pulsed laser signals. Laser pulses can be transmitted in a constant stream like a beacon, or can be modulated using a technique like pulse-position modulation, for example. The figure below represents typical output from the detection system.
The bandpass of the electronic system suggests the use of correlation techniques to detect pulses in the time-domain over analysis in the frequency domain. In the figure below, a small pulsed signal (intensity = 8.5% of the peak-to-peak noise) was injected into telescope noise to demonstarte pulse detection using correlation techniques.
The pulsed reference signal that appears in the figure below is not really visible in the noise above.
However, the autocorrelation spectrum shown below does show at least a hint of the original signal, at sample numbers 500, 600, 900 and 1000, and perhaps at 100 and 200 as well. These data were obtained near the detection limit for the pulsed signal by autocorrelation.
Cross-correlation outperforms autocorrelation by a wide margin (see figure below), but requires knowledge of the waveform sought in the noise. The detection limit for the pulsed signal by cross-correlation is less than 1% of the p-p noise in the example on this page. In actual use, however, the efficiency with which cross-correlation detects signals in noise depends upon the accuracy of the estimate of the original pulsed signal (e.g., pulse shape, duration, frequency and phase). Algorithms and models for creating this estimated signal are currently being researched by members of the ASRG.
Return to the Near-IR Telescope Page