MOVE: a framework for high-performance processor design
Proceedings of the 1991 ACM/IEEE conference on Supercomputing
The TigerSHARC DSP Architecture
IEEE Micro
SODA: A Low-power Architecture For Software Radio
Proceedings of the 33rd annual international symposium on Computer Architecture
Silicon CMOS devices beyond scaling
IBM Journal of Research and Development - Advanced silicon technology
Vector processing as an enabler for software-defined radio in handheld devices
EURASIP Journal on Applied Signal Processing
ASAP '08 Proceedings of the 2008 International Conference on Application-Specific Systems, Architectures and Processors
The next generation challenge for software defined radio
SAMOS'07 Proceedings of the 7th international conference on Embedded computer systems: architectures, modeling, and simulation
Register File Power Reduction Using Bypass Sensitive Compiler
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
A simple transmit diversity technique for wireless communications
IEEE Journal on Selected Areas in Communications
Integration of GPU Computing in a Software Radio Environment
Journal of Signal Processing Systems
Hi-index | 0.00 |
Commercial and research work in the field of software defined radio (SDR) has produced designs which have been able to deliver the efficiency and computational power needed to process 3G wireless technologies. Though efficient 3G processing has been achieved by these designs, next generation 4G SDR technology requires 10---1000x more computational performance but limits the power budget increase to 2---5x. In this paper, we present a breakdown of the major 4G kernels and analyze two methods of increasing performance and reducing power consumption. Specifically, we consider the effect of SIMD width and reduction in number of register file accesses on the performance and energy consumption of a SDR architecture, SODA. We show that by increasing SIMD width we can gain almost 2---8x performance increase while increasing total energy used by 1---2x for different SIMD widths. We also show that by reducing SIMD register accesses we can reduce the total energy used by 5---20% for the 4G kernels.