{"paper":{"title":"Multi-Block Attention for Efficient Channel Estimation in IRS-Assisted mmWave MIMO","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A multi-block attention network recovers channel estimates from 87 percent fewer pilots in IRS-assisted mmWave MIMO systems.","cross_cats":["cs.LG"],"primary_cat":"eess.SP","authors_text":"Maryam Sabbaghian, Mehrdad Momen-Tayefeh, Mehrshad Momen-Tayefeh","submitted_at":"2026-05-14T16:27:38Z","abstract_excerpt":"Intelligent Reflecting Surfaces (IRSs) are a promising technology for enhancing the spectral and energy efficiency of millimeter-wave (mmWave) multiple-input multiple-output (MIMO) systems. In these systems, accurate channel estimation remains challenging due to the passive nature of IRS elements and the high pilot overhead in large-scale deployments. This paper presents a deep learning-based Multi-Block Attention (MBA) framework for efficient cascaded channel estimation in IRS-assisted mmWave MIMO systems that utilize orthogonal frequency division multiplexing (OFDM). First, we show the optim"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The MBA method reduces pilot overhead by up to 87% compared to the LS estimator and achieves approximately 51% lower NMSE at 10 dB SNR than leading methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the simulated propagation environments and noise models accurately represent real-world mmWave channels, and that the neural network trained on synthetic data generalizes to practical deployments.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A multi-block attention neural network reduces pilot overhead by 87% and NMSE by 51% at 10 dB SNR for cascaded channel estimation in IRS-assisted mmWave MIMO-OFDM systems.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A multi-block attention network recovers channel estimates from 87 percent fewer pilots in IRS-assisted mmWave MIMO systems.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"75ea12ab2241e2fc3433f6f6242c53375f0f04fa7aa15cb026ec3e0e301ffe31"},"source":{"id":"2605.15032","kind":"arxiv","version":1},"verdict":{"id":"853b99cb-69b2-4b86-849b-54b1a8423872","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T03:11:35.261610Z","strongest_claim":"The MBA method reduces pilot overhead by up to 87% compared to the LS estimator and achieves approximately 51% lower NMSE at 10 dB SNR than leading methods.","one_line_summary":"A multi-block attention neural network reduces pilot overhead by 87% and NMSE by 51% at 10 dB SNR for cascaded channel estimation in IRS-assisted mmWave MIMO-OFDM systems.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the simulated propagation environments and noise models accurately represent real-world mmWave channels, and that the neural network trained on synthetic data generalizes to practical deployments.","pith_extraction_headline":"A multi-block attention network recovers channel estimates from 87 percent fewer pilots in IRS-assisted mmWave MIMO systems."},"references":{"count":44,"sample":[{"doi":"","year":2021,"title":"Intellige nt reflecting surface-aided wireless communications: A tutorial,","work_id":"f5f5d282-6cde-449b-8323-7ccb79f2aadf","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Toward smart wireless communications via intellig ent reflecting surfaces: A contemporary survey,","work_id":"b3c7a705-4a09-4486-afb6-4c878aea14c4","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Reconfigur able in- telligent surfaces vs. relaying: Differences, similariti es, and performance comparison,","work_id":"49daeb34-0aa5-40e1-9e54-c4d351ab1aae","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Intelligent reflecting surface enhan ced wireless network via joint active and passive beamforming,","work_id":"5ae3a64a-05fc-4622-b603-c9f25e78f52b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Reconfigurable intelligent surfaces for wireless communica- tions: Principles, challenges, and opportunities,","work_id":"c3bba13d-dd2a-4082-b8ce-04ddfdbec85a","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":44,"snapshot_sha256":"a29b158fb371aec2499535174a5f6019b475949bd445fb0c499c6a502d087d50","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8f376f1c1b93246a8e6a05e8b5816d0681adfb4432364b096b6a41343c45a606"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}