Modern wireless communication systems are designed for worst-case channel noise and interference conditions. This results in circuits that over-perform and consume more power than necessary most of the time when channel conditions are not worst-case. In this paper, we propose a power savings methodology that allows "graceful degradation" of baseband system performance when channel conditions are good without compromising overall bit-error rate. This adaptation is achieved by "dynamically modulating the active circuit critical paths via signal scaling" and by simultaneously modulating the supply voltage of the baseband signal processing circuitry. The proposed approach uses continuous monitoring of the error vector magnitude (EVM) of the demodulated signal to drive the adaptation procedure. The proposed architecture can reduce power consumption in the baseband processor of an OFDM receiver by as much as 30%. As opposed to other schemes, the proposed technique requires no modification of existing signal processing hardware.
Citation:
Muhammad Mudassar Nisar, Rajarajan Senguttuvan, Abhijit Chatterjee, "Adaptive Signal Scaling Driven Critical Path Modulation for Low Power Baseband OFDM Processors," vlsid, pp.71-76, 21st International Conference on VLSI Design (VLSI Design 2008), 2008