RIVIERAA Programmable RISC V Edge Architecture
summary
NXP develops RIVIERA, a RISC-V edge architecture for NFC that moves DSP tasks to a microprocessor plus tightly coupled hardware accelerator using the ISA extension interface. Two variants exist: a conservative MPU for auxiliary DSP and an ambitious design that replaces RX/TX paths. Iterations cut required clock from GHz to ~100 MHz, making integration into existing NFC SoCs feasible. Implementation runs on silicon-level SoC prototypes and FPGA-stimulus tests, decodes frames correctly and tolerates added noise with a remaining performance gap to hand-tuned hardware.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 1:06NFC background and challenges
- 4:37Riviera architecture concept
- 6:45Conservative architecture details
- 7:08Riviera ambitious architecture
- 9:43Performance results and feasibility
- 12:28Conclusion and future outlook
related talks
02mRISC-V Welcome
24mAndes RISC-V Processors for Control and Data Paths
19mAccelerating Edge AI with the Help of RISC-V
15mRISC-V processor IP product line
24mRISC-V FutureWatch: Introducing a New Software-defined Silicon Capability to the RIS...
15mSmallest RISC-V Device for Next-Generation Edge Computing
10mRVfpga: Using A Commercial RISC-V Processor to Teac...
14mUse of RISC-V on Pixel Visual Core
