Hardware Accelerator and Neural Network Co-Optimization for Ultra-Low-Power Audio Processing Devices.
Christoph GerumAdrian FrischknechtTobias HaldPaul Palomero BernardoKonstantin LübeckOliver BringmannPublished in: DSD (2022)
Keyphrases
- neural network
- processing capabilities
- real time
- reconfigurable hardware
- embedded systems
- low cost
- field programmable gate array
- ultra low power
- embedded devices
- multimedia
- hardware implementation
- parallel architecture
- low end
- processing elements
- high end
- image reconstruction from projections
- data processing
- data acquisition
- self organizing maps
- genetic algorithm
- fuzzy logic
- low power
- computational power
- optimization algorithm
- mobile devices
- central processor
- computer systems
- low latency
- hardware platforms
- personal computer
- optimization method
- neural network is trained
- control system
- highly non linear
- artificial neural networks
- high volume
- high speed
- massively parallel
- associative memory
- optimization problems