Open-source SoC-FPGA Platform for Signal Processing
Systems known as SoC-FPGAs have experienced a growing popularity in recent years. This devices integrate field programmable gate arrays with elements such as microprocessors, PLLs and embedded memory blocks. The advantages of this type of systems are clear: great reconfigurability, performance, and...
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Objeto de conferencia |
| Lenguaje: | Inglés |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | http://sedici.unlp.edu.ar/handle/10915/160854 |
| Aporte de: |
| Sumario: | Systems known as SoC-FPGAs have experienced a growing popularity in recent years. This devices integrate field programmable gate arrays with elements such as microprocessors, PLLs and embedded memory blocks. The advantages of this type of systems are clear: great reconfigurability, performance, and energy efficiency, but they come with an negative side:
programming and optimizing the applications that use them remains a long and complicated process. In particular, realtime signal processing at high frequencies is an application that can clearly benefit from the advantages of SoC-FPGAs, but the complex workflow assosiated with them usually prevents the designers from taking advantage of its capabilities. In this work, an open source SoC-FPGA platform, specifically intended for signal processing is presented, with the aim of alleviating this workflow. The platform structure is described, specifying the places where the designer may implement their algorithms, and then its operation is demonstrated by acquiring a signal at a maximum sampling frequency of 65 MHz and passing it through a 32th order FIR filter, verifying that the it meets it’s expected theoretical response. The whole system can operate at a maximum frequency of 85 Mhz, has a latency of 16 clock cycles, and uses less than half of the resources of a Cyclone V device. |
|---|