Millimeter wave (mmWave) radios offer attractive high bandwidths and small antenna sizes to help enabled modern miniaturized wireless systems. However, generation and distribution of a local oscillator (LO) at 10s of GHz consumes significant amounts of power. This problem is compounded when utilizing high-order QAM modulation schemes which demand very low jitter: for example, 256-QAM requires <350fs, which is very difficult and demands a high-power LO (e.g., 36mW in [1], which is even higher power than the LNA). On the other hand, a similar jitter spec can be achieved at low frequencies with much lower power consumption (e.g., 1mW [2]). This trade-off is illustrated in Fig. 1 (left). One approach to reduce the frequency and power requirements of LO generation is to perform sub-sampling (SS): as illustrated in Fig. 1 (right), down conversion in a SS receiver occurs by directly sampling the mmWave waveform at a lower frequency and extracting the desired signal's alias at a low-IF or zero IF. However, signals around integer multiples of half the sampling frequency also alias to the same IF, and thus strong antialiasing filtering is required (e.g., a rejection of 2BdB is required for 256-QAM at an SIR of OdB per Fig. 1 bottom right); this can potentially be combined with non-uniform sampling techniques [3] to further reduce the impacts of aliasing.
No takes yet. Share an insight, caveat, or question.
Gadelkarim et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: