What are some examples of how customers are using the VST right now?
At Mobile World Congress, National Instruments showcased NI vector signal transceivers in several demonstrations throughout the mobile ecosystem. From design and validation testing of the latest RF standards, such as LTE-Advanced and 802.11ac, to the parallel testing of mobile devices in manufacturing to help drive test time and test costs down using NI LabVIEW system design software and NI TestStand parallel programming software.
Specific to the user-programmable FPGA, customers have made small tweaks to the triggering functionality to better integrate with their application, enabling tighter, hardware-based timing that delivers faster test times and more consistent results. Customers have implemented digital DUT control, which reduces the need for additional instrumentation and allows the vector signal transceiver to respond more quickly to changes in their DUT state.
Others have performed more substantial processing, as in the case of RF power amplifier test, using the vector signal transceiver FPGA to measure the power amplifier output and adjust the vector signal transceiver's generated signal to reach a specific, calibrated output power, dramatically reducing test time. Some have even implemented closed-loop bit error rate test, synthesizing pseudorandom waveform data on the vector signal transceiver FPGA, and measuring the resulting error rate after the signal passes through their DUT. Finally, some have completely re-architected the vector signal transceiver FPGA logic, implementing real-time channel emulation by inserting channel models on the vector signal transceiver FPGA, between the RF input and RF output ports.
How will the new IP enable customers to get to testing faster?
While programming FPGAs in a more abstract environment such as LabVIEW FPGA is certainly more productive than using traditional languages like VHDL, developing new code in any language can be time consuming, making it inefficient for multiple engineers to develop the same algorithm independently. This is where reusable IP comes in.
NI is facilitating an ecosystem of FPGA IP for software-designed instruments, such as the vector signal transceiver, so that users who are less familiar with FPGA development can employ LabVIEW FPGA to easily integrate this IP into their own custom vector signal transceiver personality, optimized for their application. Or they can download pre-built examples/personalities for more common use cases.
For instance, let's look at something simple such as serial peripheral interface (SPI) DUT control. The NI IP ecosystem has a pre-built example for controlling a DUT using SPI through the vector signal transceiver DIO port, which might be sufficient for many applications. But if a user wants to also perform certain measurements on their vector signal transceiver FPGA, customize the triggering to respond to specific DUT events, and export signals to trigger other instruments in the test system, all in hardware time, they might need to integrate these particular pieces of reusable IP to build a vector signal transceiver personality completely optimized for their specific use case.
How many application engineers/feet on the ground do you have to help customers develop applications with the VST?
NI vector signal transceivers are sold into many different application areas in a variety of industries across the globe, so the purchase and support of the sale can vary depending on the customer's requirements. Some customers will purchase a vector signal transceiver and begin customizing their instrument on their own, starting from one of the LabVIEW Sample Projects and free IP available online. Other customers may prefer to leverage support from NI systems engineering and one or more of NI's 700 Alliance Partners across the globe to help develop a solution specifically to the customer's needs.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.