MambaCSP: Hybrid-Attention State Space Models for Hardware-Efficient Channel State Prediction
2026-04-27T08:51:49Z•3d4150dd2a6a6b87237ad00b7c0a1a15bfd740dac8d96cf146324504c85288dd
CSI predictionGamma-distributed constellationISACMIMOMambaCSPPCRBattentionbeamformingchannel-state-informationconstellation designdiffusiongenerative modelsgrouped patternintegrated sensing and communicationslow-latencynull-space flow matchingrange estimationsecrecy outage probability','wireless secrecy','reconfigurable-@semantic HARQsemantic communicationsemantic error correctionshort block codesstate-space modelstransformerswireless
What happened
Collection of April 27, 2026 arXiv submissions focused on wireless communications, sensing, and related ML methods. Key contributions: MambaCSP — a hybrid attention + linear-time state-space model (Mamba) for hardware-efficient channel state prediction with lower latency, memory and higher throughput vs. LLM-style predictors; Null-Space Flow Matching — a latency-optimized generative framework for pilot-limited MIMO channel estimation using flow-matching to refine null-space components; Semantic Error Correction/Decoding — language-model-based segment-level semantic reconstruction, list-decoder
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_math_it
- Record identifier
- 3d4150dd2a6a6b87237ad00b7c0a1a15bfd740dac8d96cf146324504c85288dd
- Enrichment time
- 2026-04-27T08:51:49Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.