Multichannel reproduction of sound has been and continues to be proposed with different approaches, there are for example systems in which loudspeakers are placed all around the listeners, systems based only on a pair of headphones which rely totally on how the signal is processed and filtered, and systems which aim at trying to give the impression of sound sources placed all around the listener, which feels immersed. But one thing is clear when it comes to consumer’s products, which have to find space in domestic environments: simplicity and compactness are important characteristics, that’s why for example soundbars are so popular these days. This thesis aims at finding the best configuration to convey a convincing surround listening experience but with the lowest possible number of loudspeakers, and without the need of placing loudspeakers all around the environment. It does so by exploring possible innovative uses of the dipole steering technique, which basically consists in reproducing some channels by means of pairs of loudspeakers used in dipole configuration, so that one plays in opposite phase with respect to the other, and additionally the inversed pole is delayed by some fractions of milliseconds to steer the direction of maximum destructive interference (zero-sound direction of the dipole) of a controllable angle. The research is performed by theoretical simulations made with MATLAB, by hybrid simulations made with EASE SpeakerLab, in which the sources are measurements of real loudspeakers, and with actual measurements on the performance of a prototype in a controllable reflective environment. The investigated approaches are the possibility to create virtual sources on lateral walls by creating reflections which can be perceived as prevalent with respect to the direct sound, or the enhancing of the sensation of spatialization by increasing the difference between the perceived level of the signals at the two ears, which is the performing of crosstalk cancellation in a transaural approach. The optimal solution will merge the two approaches in an innovative double dipole configuration, in which two different delays are applied.
La riproduzione sonora multicanale è stata e continua ad essere proposta attraverso molteplici approcci, ci sono per esempio sistemi in cui gli altoparlanti sono posizionati tutti intorno agli ascoltatori, sistemi basati solo su un paio di cuffie che si affidano totalmente sul processamento e il filtraggio del segnale, e sistemi che puntano a dare l’impressione che le sorgenti sonore siano poste tutte intorno all’ascoltatore, che si sentirà immerso. Ma una cosa è chiara quando si parla di prodotti per consumatori, che devono trovare spazio in ambienti domestici: la semplicità e la compattezza sono caratteristiche importanti, ecco perché ad esempio le soundbar sono così popolari al giorno d’oggi. Lo scopo di questa tesi è trovare la miglior configurazione per trasmettere una convincente esperienza di surround ma col numero minore possibile di altoparlanti, e senza la necessità di posizionare gli altoparlanti per tutto l’ambiente. Ciò è fatto esplorando possibili usi innovativi della tecnica dello steering del dipolo, che consiste nel riprodurre dei canali sonori attraverso coppie di altoparlanti usati in configurazione a dipolo, cosicché uno suoni a fase opposta rispetto all’altro, e in aggiunta il polo invertito è ritardato di qualche frazione di millisecondo per virare la direzione di massima interferenza distruttiva (lo zero del dipolo) di un angolo controllabile. La ricerca è svolta attraverso simulazioni teoriche fatte in MATLAB, attraverso simulazioni ibride fatte con EASE SpeakerLab, in cui le sorgenti sono misurazioni di veri altoparlanti, e con effettive misurazioni sulle performance di un prototipo in un ambiente riflettente controllabile. Gli approcci sondati sono la possibilità di creare sorgenti virtuali sui muri laterali creando riflessioni che possano essere percepite come prevalenti rispetto al suono diretto, o l’arricchimento della sensazione di spazializzazione attraverso l’aumento della differenza tra i livelli dei segnali percepiti dalle due orecchie, che è l’attuazione della crosstalk cancellation in un approccio transaurale. La soluzione ottima unirà i due approcci in un’innovativa configurazione a doppio dipolo, in cui sono applicati due diversi valori di delay.
Multichannel signal reproduction optimization by means of dipole steering technique in a domestic non-treated environment
TALONE, LORENZO
2020/2021
Abstract
Multichannel reproduction of sound has been and continues to be proposed with different approaches, there are for example systems in which loudspeakers are placed all around the listeners, systems based only on a pair of headphones which rely totally on how the signal is processed and filtered, and systems which aim at trying to give the impression of sound sources placed all around the listener, which feels immersed. But one thing is clear when it comes to consumer’s products, which have to find space in domestic environments: simplicity and compactness are important characteristics, that’s why for example soundbars are so popular these days. This thesis aims at finding the best configuration to convey a convincing surround listening experience but with the lowest possible number of loudspeakers, and without the need of placing loudspeakers all around the environment. It does so by exploring possible innovative uses of the dipole steering technique, which basically consists in reproducing some channels by means of pairs of loudspeakers used in dipole configuration, so that one plays in opposite phase with respect to the other, and additionally the inversed pole is delayed by some fractions of milliseconds to steer the direction of maximum destructive interference (zero-sound direction of the dipole) of a controllable angle. The research is performed by theoretical simulations made with MATLAB, by hybrid simulations made with EASE SpeakerLab, in which the sources are measurements of real loudspeakers, and with actual measurements on the performance of a prototype in a controllable reflective environment. The investigated approaches are the possibility to create virtual sources on lateral walls by creating reflections which can be perceived as prevalent with respect to the direct sound, or the enhancing of the sensation of spatialization by increasing the difference between the perceived level of the signals at the two ears, which is the performing of crosstalk cancellation in a transaural approach. The optimal solution will merge the two approaches in an innovative double dipole configuration, in which two different delays are applied.File | Dimensione | Formato | |
---|---|---|---|
2021_12_Talone.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
10.68 MB
Formato
Adobe PDF
|
10.68 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/183970