A low-voltage energy-sampling IR-UWB digital baseband employing quadratic correlation
This paper describes a digital baseband designed for use in a non-coherent IR-UWB system. Owing to the nonlinear statistics introduced by the energy-sampling RF front-end, the baseband employs a new quadratic correlation technique that achieves comparable performance to a matched filter classifier,...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers
2010
|
| Online Access: | http://hdl.handle.net/1721.1/59998 https://orcid.org/0000-0002-5977-2748 |
| Summary: | This paper describes a digital baseband designed for use in a non-coherent IR-UWB system. Owing to the nonlinear statistics introduced by the energy-sampling RF front-end, the baseband employs a new quadratic correlation technique that achieves comparable performance to a matched filter classifier, with the added benefit of being robust to SNR estimation errors. Additionally, “alias-free codes” are introduced that allow for pulse-level synchronization accuracy without requiring any increase in front-end complexity. Fabricated in a 90 nm CMOS process, the digital baseband utilizes significant parallelism in addition to clock and data gating to achieve low-power operation, with supply voltages as low as 0.55 V. At a clock frequency of 32 MHz, the baseband requires 14-to-79 μs to process a preamble during which it consumes an average power of 1.6 mW, while payload demodulation requires 12 pJ/bit. |
|---|