Spectrum Instrumentation's new package uses SCAPP (Spectrum's CUDA Access for Parallel Processing) together with the company's latest digitizers to harness the parallel structure of a CUDA graphics card for data processing. The data is sent directly to a GPU using Remote Direct Memory Access (RDMA) transmission, where fast time and frequency domain signal averaging is possible and limitations associated with other approaches are avoided. The new averaging package is interesting for all users who work with weak signals. These include applications such as mass spectrometry, LIDAR, radio astronomy, automation, radar, biomedicine, sonar and many more. The package is available now.
Until now, users who wanted to perform signal averaging had three basic options: First, they could buy a digital oscilloscope that included signal averaging. Another possibility was the purchase of a digitizer card, where the digitized data is sent to a PC and the host processor performs the calculations. As a third option, a special digitizer could be purchased, which takes over the signal averaging itself with FPGA technology (Field Programmable Gate Array). However, all these three techniques have severe limitations when averaging for long signals has to be performed.