National Research & Development Laboratory Applications

Atmospheric Remote Sensing

Customer Case

The customer requires a combination of hardware and software (“data system”) to do live update sampling, processing, display, and archival of atmospheric measurements using an airborne Lidar system. This system, which continuously emits laser pulses, measures line-of-sight velocity and backscattered signals at a particular range from the aircraft. Real-time performance is necessary to observe the state of the atmosphere and to assess the performance of their instrument. Their requirement is similar to that of laser Doppler anemometry.

The signal from the Lidar system will vary continuously with time and the signal waveform will be non-repeating.

They are searching for a PC-based, PCI-compatible analog-to-digital converter (ADC) with a minimum of 8-bit precision, at least 100 MHz sampling rate and at least 65 MHz analog input bandwidth. A separate Digital Signal Processing (DSP) board for signal processing is desirable but not required depending on availability of Commercial Off-The-Shelf (COTS) software to accomplish the signal analysis and display on the host PC.

The PC will also be tasked with on-line acquisition of ancillary aircraft data via asynchronous RS-232. The A/D board and software must not interfere with this data acquisition requirement.

An existing, user-friendly, graphical user interface is strongly preferred, as the customer wants to minimize code development.

Ideally, the data system should operate as follows. After being amplified and low-pass filtered below 65 MHz, the output signal from the Lidar system will be sent to the ADC.

  • Step 1. Software issues a trigger to the A/D board –> acquire 256-point time series.
  • Step 2. Do 256-point FFT –> obtain power spectrum (consisting of 128 frequency bins including the DC zero-frequency coefficient).
  • Step 3. Sum this power spectrum with previous power spectra on a bin-by-bin basis.
  • Step 4. Has one second of time elapsed? = YES, Go to Step 5 = NO, Go to Step 1.
  • Step 5. Transfer summed power spectrum to PC for processing (into scientifically-meaningful units), visually display the spectrum, and archive the results.
  • Step 6. Acquire aircraft housekeeping data via asynchronous RS-232 interface.
  • Step 7. Go back to Step 1.

The software or DSP board will read data from the ADC in sets of 256 samples, apply a Hamming window, and generate a power spectrum using a radix-2 Fast Fourier Transform (FFT). Successive power spectra will be integrated or summed on a bin-by-bin basis (64 bins total) to reduce noise. The integration period will be terminated by the PC at one-second intervals, allowing some number N of spectra to be accumulated. In the absence of on-board processing capability on the A/D, the integrated spectrum must be analyzed, displayed, and archived on the PC while the ADC continues to collect data. It is preferred that the software driving the ADC board be capable of mastering the PCI bus, because the customer wishes to maximize the number of spectra that can be acquired each second. Alternatively, a DSP board (linked to the ADC via a dedicated high-speed bus) could be used for data analysis, provided this option requires minimal programming expertise to implement. Ideally, they want to be able to accumulate literally tens of thousands of spectra per second.

GaGe Case Solution

The optimal GaGe solution is our CompuScope 12100 digitizer card in a GaGe Industrial Mainframe 2000 Computer. This 12-bit PCI digitizer card has high dynamic range, can sample at up to 100 MegaSamples/second and can offload data using PCI bus-mastering at sustained rates of up to 100 MegaBytes/second.

The objective is to capture 256-point Lidar signals at 100 MS/s, perform a Fourier power spectrum calculation and continuously sum the result. Summing will continue for about one second at which point the resulting grand average Fourier spectrum will be displayed. They want to be able to average as many spectra as possible during the one second refresh cycle.

The CompuScope 12100 will provide the fastest available capture of repetitive Lidar signals. We have an application note describing the use of our CompuScope 8500 in a demanding medical ultrasound application. The note demonstrates that our CompuScope 8500 can capture 1000-point records that occur at a rate in excess of 20,000 per second without missing triggers. Built on the same baseboard as the CS8500, the CS12100 has identical data transfer characteristics. The CS12100, therefore, is fully capable of capturing their 256 point records at rates in excess of 20,000 per second.

How many Fourier spectra that they can acquire and average will also depend on calculation speed. The best available calculation speed will be provided by the CPU in the GaGe Industrial Mainframe. The powerful GaGeScope PC oscilloscope software is available with an FFT option. As an example, GaGe engineers used GaGeScope to capture 100,000-point records and then calculated and displayed the resulting Fourier Power Spectrum. We observed the data to refresh at about a 1 Hz rate.

The accumulation and summing of multiple Fourier Spectra is an available feature under our GaGeScope software. We visited an engineer who happened to have a CS12100 set up under GaGeScope in his PC. We continuously captured 256 point records on one channel at 100 MS/s. We set up the FFT option to average 1,000 separate FFTs and display the result. We found the refresh time of the FFT display to be about five seconds. Using our easy-to-use CompuScope Software Development Kit, a custom C application could be written that provides optimal performance for their application. The refresh time could be improved in such a custom application.

GaGe Case Recommended Products

Research & Development Application Request

We encourage you to contact us and discuss your research & development application in more detail with our engineering team. GaGe can provide tailored custom data acquisition hardware and software solutions to meet specific application requirements.