Energy Consumption of Channel Decoders for OFDM-based UWB Systems

Ultra Wide Band (UWB) systems are designed for short range transmission with very high throughput demands (480 Mbit/s). Current UWB systems specify convolutional codes (CC) for channel coding. However, it was already shown that more sophisticated Low-Density Parity-Check Codes (LDPC) can outperform CC. This gain of communications performance comes with the penalty of a higher implementation complexity. In this paper, the currently specified convolutional code and an LDPC code are compared concerning their decoders' implementation complexity with special emphasis on its energy consumption. In this context, we evaluate an advanced iteration control for LDPC decoders in the environment of the future WIMEDIA UWB industry standard. The number of iterations is directly related to the LDPC decoder's energy consumption. Both decoders are implemented on a rapid prototyping platform to evaluate the energy consumption of LDPC decoding compared to CC decoding. Measurements show that the energy consumption of both decoders are nearly identical while LDPC codes come along with a more than 3 dB better communications performance.