Nutaq’s PicoDigitizer and PicoSDR line of products offer great flexibility when prototyping wireless applications or capturing real-world wireless signals. The PicoDigitizer offers even more flexibility by giving a user the choice to use an external RF front-end with the device. To showcase this capability and address the issues most engineers encounter when capturing real-world wireless signals, Nutaq offers a bundle that includes an external RF receiver (the WR8G/20G) along with a PicoDigitizer 250-Series. The bundle is called the WD8G/20G RF Wideband Digitizer. The WR8G/20G front-end is a highly tunable receiver that can capture an entire 100 MHz wide communication band anywhere within a wireless frequency range of 100 kHz to 20 GHz. The PicoDigitizer 250-Series incorporates a dual 250 MSPS, 14-bit A/D converter and dual 1 GSPS 16-bit D/A channels on a large Virtex-6 field-programmable gate array (FPGA).
The following image shows the bundle.
For capturing real-world data, the RF Wideband Digitizer has the ability to record short bursts (in the 4 GB DDR3 memory) or longer bursts (on the internal 200 GB SDD or a larger external drive). The choice on where to store the recorded bursts typically depends on five factors:
- Bandwidth of the RF signal of interest
- Sampling rate used
- Whether or not real-time preprocessing (FPGA or PC) is applied on the signal before the samples are stored (e.g. filtering and down-sampling)
- Location(s) where real-time preprocessing is performed (PC and/or FPGA)
- Interfaces in which the signal travels through before being stored (e.g. DDR3, PCIe, SSD, etc) and their respective throughput limitations
The following block diagram shows the interfaces and their respective throughput limitations.
In subsequent posts in this blog series, we will demystify the different signal path options for capturing and/or processing real-world wireless signals based on these five factors. Each post will focus on one option, investigating what the data bottlenecks are and providing solutions to overcome those limitations with the introduction of real-time processing in the signal chain.