AI & ML
Fisher Information Characterization in GNSS DPE under Multipath-Induced Channel Uncertainty
arXiv eess.SP0 views
arXiv:2611.23709v1 Announce Type: new Abstract: The availability and utilization of Fisher information in global navigation satellite system (GNSS) direct position estimation (DPE) under multipath-induced channel uncertainty are investigated. First, a unified estimation-theoretic framework is developed to characterize information mapping and utilization across direct and two-step receivers. Next, a scalar-equivalent statistical channel model for short-delay multipath is established via least-squares (LS) projection. Based on this model, receiver-state Cram\'er-Rao bounds (CRBs) are derived under perfect and statistical channel knowledge, while misspecified CRBs (MCRBs) are derived under a simplified deterministic complex-gain model that neglects channel uncertainty. The closed-form approximations and asymptotic analyses of the derived bounds further reveal how channel uncertainty affects different components of waveform-domain Fisher information. Our results reveal that: 1) DPE and fully coupled two-step estimation achieve comparable accuracy, whereas the performance loss caused by sequential decoupling substantially exceeds the additional loss from suboptimal weighting; 2) under statistical channel knowledge, improving received signal quality does not necessarily yield noticeable positioning gains; 3) positioning information remains available even as the coherent channel mean vanishes; and 4) neglecting channel randomness causes the MCRBs to diverge in this limit and can introduce positioning error floors that become more pronounced with longer observations.
Source:https://arxiv.org/abs/2611.23709PreviousDeep-Learning-Assisted Full Waveform Inversion for Attenuation Reconstruction in Ultrasound Computed TomographyNextTowards Environment-Aware 6G Network Planning: An Any-to-Any Channel Knowledge Map Approach
Related posts
My Wallet Introduces 1:1 Cross-Chain Stablecoin Swaps After Being Listed First on WalletConnect
0HackerNoon
Jensen Huang Chinese Revenue, Affairs, Resignation, and Case for CEO Reinstatement Timeline
0HackerNoon
International Men’s Day: Anthropic IPOs, Trump Meets Putin and Xi, and GTA 6 Is Released
0HackerNoon

What Happens When AI Makes So Many Decisions That We Forget How?
0HackerNoon
Elon Musk says SpaceX has a plan to eliminate cell phone dead zones
0TechCrunch
Freezing a model that caught 2% of real errors: OCR error detection for prewar Japanese print
0dev.to