Documenti full-text disponibili:
Abstract
Integrated sensing and communication (ISAC) is a foundational paradigm for 6G and beyond wireless systems, enabling simultaneous high-rate communication and precise sensing within a unified radio framework. However, its practical deployment faces a fundamental challenge: the significant increase in computational and architectural complexity required to achieve high sensing performance. This dissertation addresses the complexity bottleneck of ISAC by proposing low-complexity signal processing algorithms and architectures across three complementary domains: ultra-wideband orthogonal frequency-division multiplexing (OFDM)-based ISAC at terahertz (THz) frequencies, wideband OFDM-based ISAC under near-field (NF) propagation, and orthogonal time frequency space (OTFS)-based ISAC. First, a bistatic multiple-input multiple-output (MIMO)-OFDM ISAC system operating at THz is proposed for precise localization and tracking of an extended target (ET) in cluttered indoor environments. To achieve high range resolution while controlling computational complexity, a two-stage transmission scheme with maximum likelihood (ML) estimation is introduced. Efficient low-complexity clutter suppression techniques and a particle filter (PF)-based ET tracking model are also developed. Second, a hybrid far-/near-field monostatic MIMO-OFDM ISAC architecture is introduced to exploit NF gains without incurring extremely large-scale MIMO (XL-MIMO) complexity. An extremely large antenna spacing (ELAS) design is proposed to reduce the number of antenna elements while avoiding grating lobes. The interplay between bandwidth and NF beam focusing is analyzed, identifying a NF super-resolution region where range resolution exceeds bandwidth limits. Finally, a low-complexity detector for OTFS sensing systems is proposed, based on an approximation criterion that leverages the properties of the Dirichlet kernel to retain the most informative elements of the sensing channel matrix. The method is extended to a cooperative multi-base station (BS) ISAC scenario, improving localization accuracy compared to single-BS setups. Overall, this dissertation demonstrates that high-resolution and accurate sensing performance can be achieved without prohibitive computational or hardware costs, providing practical and scalable design insights for low-complexity ISAC implementations.
Abstract
Integrated sensing and communication (ISAC) is a foundational paradigm for 6G and beyond wireless systems, enabling simultaneous high-rate communication and precise sensing within a unified radio framework. However, its practical deployment faces a fundamental challenge: the significant increase in computational and architectural complexity required to achieve high sensing performance. This dissertation addresses the complexity bottleneck of ISAC by proposing low-complexity signal processing algorithms and architectures across three complementary domains: ultra-wideband orthogonal frequency-division multiplexing (OFDM)-based ISAC at terahertz (THz) frequencies, wideband OFDM-based ISAC under near-field (NF) propagation, and orthogonal time frequency space (OTFS)-based ISAC. First, a bistatic multiple-input multiple-output (MIMO)-OFDM ISAC system operating at THz is proposed for precise localization and tracking of an extended target (ET) in cluttered indoor environments. To achieve high range resolution while controlling computational complexity, a two-stage transmission scheme with maximum likelihood (ML) estimation is introduced. Efficient low-complexity clutter suppression techniques and a particle filter (PF)-based ET tracking model are also developed. Second, a hybrid far-/near-field monostatic MIMO-OFDM ISAC architecture is introduced to exploit NF gains without incurring extremely large-scale MIMO (XL-MIMO) complexity. An extremely large antenna spacing (ELAS) design is proposed to reduce the number of antenna elements while avoiding grating lobes. The interplay between bandwidth and NF beam focusing is analyzed, identifying a NF super-resolution region where range resolution exceeds bandwidth limits. Finally, a low-complexity detector for OTFS sensing systems is proposed, based on an approximation criterion that leverages the properties of the Dirichlet kernel to retain the most informative elements of the sensing channel matrix. The method is extended to a cooperative multi-base station (BS) ISAC scenario, improving localization accuracy compared to single-BS setups. Overall, this dissertation demonstrates that high-resolution and accurate sensing performance can be achieved without prohibitive computational or hardware costs, providing practical and scalable design insights for low-complexity ISAC implementations.
Tipologia del documento
Tesi di dottorato
Autore
Bacchielli, Tommaso
Supervisore
Co-supervisore
Dottorato di ricerca
Ciclo
38
Coordinatore
Settore disciplinare
Settore concorsuale
Parole chiave
Low-Complexity Signal Processing, Integrated Sensing and Communication, Orthogonal Frequency-Division Multiplexing, Orthogonal Time Frequency Space, Ultra-Wideband THz Sensing, Wideband Near-Field Sensing
DOI
10.48676/unibo/amsdottorato/13307
Data di discussione
7 Luglio 2026
URI
Altri metadati
Tipologia del documento
Tesi di dottorato
Autore
Bacchielli, Tommaso
Supervisore
Co-supervisore
Dottorato di ricerca
Ciclo
38
Coordinatore
Settore disciplinare
Settore concorsuale
Parole chiave
Low-Complexity Signal Processing, Integrated Sensing and Communication, Orthogonal Frequency-Division Multiplexing, Orthogonal Time Frequency Space, Ultra-Wideband THz Sensing, Wideband Near-Field Sensing
DOI
10.48676/unibo/amsdottorato/13307
Data di discussione
7 Luglio 2026
URI
Statistica sui download
Gestione del documento: