A lot of people know about Nutaq’s hardware capabilities, a little less know about our Model-Based Design Kit (MBDK), and even less know that the MBDK is the main development tool for our customers. Today’s blog post tells an interesting story involving Pennsylvania State University. By using our MBDK, they were able to rapidly prototype and accurately simulate a software-defined radio (SDR) ground station for a nanosatellite.
The Local Ionospheric Measurements Satellite (LionSat) project at Pennsylvania State University provided students from diverse backgrounds and technical fields the opportunity to develop a small satellite. One of the objectives of this mission was to test uplink and downlink IP-based communication for a spacecraft in low Earth orbit.
Their main technical challenge was programming the FPGA card. Writing and debugging code for an FPGA in VHDL can be a time-consuming and error-prone process. To allow more time to focus on system design, they needed an approach that minimized low-level programming.
The team was lead by Dr. Sven G. Bilén, an Associate of Engineering Design and Electrical Engineering at Pennsylvania State University. Using MATLAB and Simulink along with related toolboxes and blocksets, they explored different algorithm options, optimized system parameters, and evaluated design trade-offs. “MathWorks tools helped us to move from floating point to fixed point to get an accurate representation of the hardware,” said Jerker Taudien, a student involved in the project.
Using Xilinx System Generator and our MBDK, they generated VHDL code from their Simulink models without any manual coding. “By using a high-level environment such as Simulink, I can focus the course on SDR architecture and applications, rather than the nitty-gritty of VHDL. This will also enable students who come with an electromagnetics and RF background, as I do, to get as much out of the course as those with a signal processing background,” said Bilén.
Simulink and System Generator enabled them to produce a prototype very fast. “With MathWorks tools and Nutaq technology, we have an integrated solution that allowed us to prototype very quickly without programming our FPGAs at a low level. This enabled us to immediately begin exploring how to implement an overall system”, said Bilén.
It also helped them to develop a complete simulation model, from inputs to outputs, including the hardware interfacing. “After generating the code, it worked the first time. I expected more debugging, but everything that worked in simulation worked in real time,” reported Taudien.
Finally, it gave students a single hardware test environment as they could design, develop, and simulate entire systems in Simulink. “With Simulink, you can switch easily between hardware-in-the-loop and computer-only simulation. You can use simulated signals, and then quickly change to using input from a real antenna or digitizer. That flexibility accelerates development,” said Bilén.
This success story does a very nice job at demonstrating the benefits of our model-based design approach to digital processing applications. This has been one of our key differentiators for many years. There is a reason behind the fact that we are still supporting and updating this technology: we believe in it and it just works.