Noncoherent MIMO for satellite communications

Determine whether noncoherent MIMO based on unitary space-time modulation, Grassmannian signaling, or differential schemes could provide benefits for satellite communications, particularly given satellite links’ higher latency and potentially outdated channel-state information at the transmitter.

Background

The paper argues that satellite communications may require separate consideration because their latency is higher than that of terrestrial links. Although line-of-sight conditions and temporal channel correlation may facilitate channel tracking, the increased latency can make updated channel-state information at the transmitter difficult to acquire. The authors explicitly leave unresolved whether noncoherent approaches can address this challenge effectively.

References

Although the preceding general observations regarding LoS conditions broadly apply, the higher latency relative to terrestrial communications may severely limit the acquisition of updated CSIT, even when temporal channel correlation is present. Although beyond the scope of this work, whether noncoherent MIMO based on USTM, Grassmannian signaling or differential schemes could be beneficial for satellite communications is an interesting question for future investigation.

— Noncoherent MIMO in the FR3 Upper Mid-Band Using Grassmannian Signaling: Prototype and Experimental Evaluation  (2610.08056 - Pendás-Recondo et al., 6 Oct 2026) in Section II-B, “Motivation: Noncoherent MIMO in the FR3 Upper Mid-Band,” subsection “Input Parameters”

Finally, although in this paper we identify $f_{\mathrm{c} = 15$ GHz as a possible non-strict upper limit for the practical application of Grassmannian signaling under our working assumptions, the potential use of these or similar noncoherent schemes in satellite communications and at higher carrier frequencies remains an interesting open research question.

— Noncoherent MIMO in the FR3 Upper Mid-Band Using Grassmannian Signaling: Prototype and Experimental Evaluation  (2610.08056 - Pendás-Recondo et al., 6 Oct 2026) in Conclusion, final paragraph