{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:F4NMZ6HYTC62RAOBM4Z3Q35RNF","short_pith_number":"pith:F4NMZ6HY","schema_version":"1.0","canonical_sha256":"2f1accf8f898bda881c16733b86fb1697bcbff69af75dfdd4bcb0c772024b159","source":{"kind":"arxiv","id":"2502.17472","version":1},"attestation_state":"computed","paper":{"title":"In-sensor 24 classes HAR under 850 Bytes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.SP","authors_text":"Ahmed.S Benmessaoud, Farida Medjani, Khalid Bouaita, Tahar Kezai, Wassim Kezai","submitted_at":"2025-02-13T19:00:44Z","abstract_excerpt":"The year 2023 was a key year for tinyML unleashing a new age of intelligent sensors pushing intelligence from the MCU into the source of the data at the sensor level, enabling them to perform sophisticated algorithms and machine learning models in real-time. This study presents an innovative approach to Human Activity Recognition (HAR) using Intelligent Sensor Processing Units (ISPUs), demonstrating the feasibility of deploying complex machine learning models directly on ultra-constrained sensor hardware. We developed a 24-class HAR model achieving 85\\% accuracy while operating within an 850-b"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2502.17472","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2025-02-13T19:00:44Z","cross_cats_sorted":[],"title_canon_sha256":"7b34810731b826c1abd12995829dc96312d6a342319f36a2d9766af20928a3f0","abstract_canon_sha256":"55a886d6fc22d433eca42d77e46aa47d0a0998e99f15c7b78dc1b27c2155ccc9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:11.305419Z","signature_b64":"yCCPZkQwpvS4g9VFI+luMAr8Bes341m+F+0EinXl0cyyrtqDtoT1e3zCjyLz2o5y+xmHzV0wo+aSDwK1w774Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f1accf8f898bda881c16733b86fb1697bcbff69af75dfdd4bcb0c772024b159","last_reissued_at":"2026-07-05T10:19:11.304923Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:11.304923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In-sensor 24 classes HAR under 850 Bytes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"eess.SP","authors_text":"Ahmed.S Benmessaoud, Farida Medjani, Khalid Bouaita, Tahar Kezai, Wassim Kezai","submitted_at":"2025-02-13T19:00:44Z","abstract_excerpt":"The year 2023 was a key year for tinyML unleashing a new age of intelligent sensors pushing intelligence from the MCU into the source of the data at the sensor level, enabling them to perform sophisticated algorithms and machine learning models in real-time. This study presents an innovative approach to Human Activity Recognition (HAR) using Intelligent Sensor Processing Units (ISPUs), demonstrating the feasibility of deploying complex machine learning models directly on ultra-constrained sensor hardware. We developed a 24-class HAR model achieving 85\\% accuracy while operating within an 850-b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17472","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2502.17472/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2502.17472","created_at":"2026-07-05T10:19:11.304972+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.17472v1","created_at":"2026-07-05T10:19:11.304972+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17472","created_at":"2026-07-05T10:19:11.304972+00:00"},{"alias_kind":"pith_short_12","alias_value":"F4NMZ6HYTC62","created_at":"2026-07-05T10:19:11.304972+00:00"},{"alias_kind":"pith_short_16","alias_value":"F4NMZ6HYTC62RAOB","created_at":"2026-07-05T10:19:11.304972+00:00"},{"alias_kind":"pith_short_8","alias_value":"F4NMZ6HY","created_at":"2026-07-05T10:19:11.304972+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.02827","citing_title":"USAD: End-to-End Human Activity Recognition via Diffusion Model with Spatiotemporal Attention","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF","json":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF.json","graph_json":"https://pith.science/api/pith-number/F4NMZ6HYTC62RAOBM4Z3Q35RNF/graph.json","events_json":"https://pith.science/api/pith-number/F4NMZ6HYTC62RAOBM4Z3Q35RNF/events.json","paper":"https://pith.science/paper/F4NMZ6HY"},"agent_actions":{"view_html":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF","download_json":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF.json","view_paper":"https://pith.science/paper/F4NMZ6HY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.17472&json=true","fetch_graph":"https://pith.science/api/pith-number/F4NMZ6HYTC62RAOBM4Z3Q35RNF/graph.json","fetch_events":"https://pith.science/api/pith-number/F4NMZ6HYTC62RAOBM4Z3Q35RNF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF/action/storage_attestation","attest_author":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF/action/author_attestation","sign_citation":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF/action/citation_signature","submit_replication":"https://pith.science/pith/F4NMZ6HYTC62RAOBM4Z3Q35RNF/action/replication_record"}},"created_at":"2026-07-05T10:19:11.304972+00:00","updated_at":"2026-07-05T10:19:11.304972+00:00"}