In the DDC process, RF or microwave signals are converted to baseband, which contains the signal of interest. This conversion dramatically reduces the resulting dataset while improving signal quality and measurement accuracy. Spectrum Instrumentation offers 48 different PCIe-based digitizers (with sampling rates from 5 MS/s to 10 GS/s) that can perform the new DDC function, allowing customers to select the "Perfect-Fit" model for their application.

Integrated FPGA DDC or External GPU Card?
Most DDC digitizer implementations use integrated FPGA (Field Programmable Gate Array) technology. Incoming analog signals are converted to digital data before being passed to the FPGA for downconversion. This method is fast and efficient, but it has its limitations. It requires large and expensive FPGA technology, as well as custom-built firmware. Customizing the firmware is also challenging, requiring specialized firmware development skills and costly software tools. Using the company's SCAPP (Spectrum's CUDA Access for Parallel Processing) software development kit, the data acquired by the digitizer can be streamed over the digitizer's PCIe bus directly to a CUDA-based GPU. The GPU, which can have thousands of cores working in parallel, then allows the processing software to be created using C/C++. This greatly simplifies DDC implementation, as customization can be accomplished with standard programming skills. Starting with a proven DDC example delivers immediate results and provides a platform from which further software optimization is possible.

A working example with transmission speeds of 12.8 GB/s.
Spectrum Instrumentation's digitizer range includes PCIe cards on three different platforms (M2p, M4i, and M5i). They offer sampling rates from 5 MS/s to an ultrafast 10 GS/s, with resolutions from 8 to 16 bits and bandwidths from 2.5 MHz to over 3 GHz. The M5i series, shown in Figure 1, represents the top of the range, offering the fastest sampling rates and highest bandwidths, all at 12-bit resolution. Another key feature of the M5i series is its ability to transmit data over the PCIe bus at a market-leading speed of 12.8 GB/s.

The figure shows frequency domain screenshots of a signal before and after downsampling. In this case, an M5i.3337-x16 digitizer card was used to sample a 702 MHz input signal at a rate of 6.4 GS/s. The acquired data was continuously streamed directly to an Nvidia RTX A4000 GPU with 6144 cores at a maximum transfer rate of 12.8 GB/s. Once there, the working example has access to the various processing blocks necessary to perform the downsampling function. This includes a direct digital synthesizer (DDS), low-pass filtering, and downsampling. The GPU performs all the necessary downsampling tasks. In this case, it mixes the data with a complex sinusoid (generated by the DDS), applies a moving average, decimates the result (in this case by a factor of 512), passes the decimated data through a finite impulse response (FIR) filter, rescales it, and then transfers the processed data to the PC's memory for storage (or further processing).

The two frequency domain screenshots show that the signal-to-noise ratio of the reduced signal (green) has improved by almost 10 dB compared to the original (yellow). Furthermore, the data file has been drastically reduced by a factor of 512, yet it still retains all the important information within the frequency range of the signal of interest. The smaller data file is now much easier to store, analyze, and visualize on a standard PC.

From now on, the new DDC function is part of the low-cost SCAPP package. This software package is required to combine new or existing Spectrum digitizers with a GPU card.

More information