A Fast DVM Algorithm for Wideband Time-Delay Multi-Beam Beamformers

This paper presents a sparse factorization for the delay Vandermonde matrix (DVM) along with fast, exact, radix-2, and recursive algorithms to compute the DVM-vector product for wideband multi-beam antenna arrays. The proposed algorithms enable low-complexity wideband beamformers in emerging millime...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on signal processing 2022-01, Vol.70, p.1-13
Hauptverfasser: Perera, Sirani M., Lingsch, Levi, Madanayake, Arjuna, Mandal, Soumyajit, Mastronardi, Nicola
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper presents a sparse factorization for the delay Vandermonde matrix (DVM) along with fast, exact, radix-2, and recursive algorithms to compute the DVM-vector product for wideband multi-beam antenna arrays. The proposed algorithms enable low-complexity wideband beamformers in emerging millimeter-wave wireless communication networks by reducing the complexity of N-beam wideband beamforming from \mathcal {O}(N^{2}) to \mathcal {O}(N \mathrm{\: log\:} N), where N=2^{r}(r \geq 1). As a result, the algorithms are faster than the brute-force computation of the DVM-vector product and more efficient than the direct realization of true-time-delay-based multi-beam beamformers. The proposed low-complexity algorithms' signal flow graph (SFG) is also presented to highlight their suitability for hardware implementations. The 2-D frequency responses of DVM-based beamformers are explained through an array signal processing example. Simulation results suggest that integrated circuit (IC) implementations of the SFG significantly reduce chip area and power consumption.
ISSN:1053-587X
1941-0476
DOI:10.1109/TSP.2022.3231182