Anders Buvarp
Foto:
Adam Troy
Om
Jeg driver forskning innenfor kvantefeilrettingskoding, robust kvantesignalbehandling, robust/sikker føderert læring, omkonfigurerbare intelligente overflater, etc. Jeg er også interessert i å anvende kompleksverdige og kvaternion-neurale nettverk (QNN) i trådløse systemer. I det siste har jeg begynt å se nærmere på feltet kvantesignalbehandling (QSP) og blind utjevning med omkonfigurerbare intelligente overflater (RIS).
Kompetanseord
- Complex-valued signal processing
- Federated learning
- Geometric signal processing
- Information theory
- Machine learning
- Quantum communications
- Quantum control
- Quantum error correction
- Quantum information
- Quantum state geometry
- Quaternion-valued signal processing
- Reconfigurable intelligent surfaces
- Robust estimation
- Statistical signal processing
- Wireless communications
Forskning
Jeg driver forskning innenfor kvantefeilrettingskoding, robust kvantesignalbehandling, robust/sikker føderert læring, omkonfigurerbare intelligente overflater, etc. Jeg er også interessert i å anvende kompleksverdige og kvaternion-neurale nettverk (QNN) i trådløse systemer. I det siste har jeg begynt å se nærmere på feltet kvantesignalbehandling (QSP) og blind utjevning med omkonfigurerbare intelligente overflater (RIS).
CORRECT — Continuously Operating Geometric Quantum Error Correction for Bosonic Codes
CORRECT develops geometric and robust methods for continuous quantum error correction in bosonic systems, combining structured quantum-state representations, real-time inference, and geometry-aware decoding under realistic noise and hardware constraints.
QUARTZ
QUARTZ develops robust quaternion-valued receivers for polarization-diverse quantum communications, including dual-polarization CV-QKD, with emphasis on structured signal representations, blind equalization, robustness, and learning-based detection.
KVIST
KVIST develops assurance methods for autonomous systems, with emphasis on robust estimation, validation, uncertainty, and trustworthy decision support under model mismatch, environmental variation, and other non-nominal conditions.
BERG
BERG studies structured quantum-state manifolds, intrinsic quantum geometry, and local versus global distinguishability, with robust trajectory design and manifold-aware decoding under realistic quantum noise and open-system dynamics.
TORQ
TORQ develops robust geometric methods for quantum control, combining Hamiltonian identification, noise-aware estimation, and constrained pulse optimization to improve calibration and gate performance under model uncertainty and non-Gaussian disturbances.
VISTA
VISTA develops variational and structure-aware inference methods for robust signal estimation, combining latent-variable models, uncertainty quantification, and divergence-based diagnostics under noise, model mismatch, and non-nominal conditions.
PRIOR
PRIOR studies physics-informed variational estimation for quantum SATCOM, focusing on prior mismatch in polarization channels and how Rician, Gaussian, and other latent priors affect robust channel estimation under fading and polarization mixing.
PRISM
PRISM develops identifiable blind and semi-blind channel inference for RIS-assisted TDD Massive MIMO using structured phase probing, latent-data estimation, and geometry-aware models under partial observation and model uncertainty.
DISAC
DISAC studies disintegrated sensing and communications in adversarial electromagnetic environments, including Eisenstein-lattice repeater deployment, resilient Massive MIMO, interference-aware inference, and robust operation under hostile sensing.
Publikasjoner
Toward robust real-time bosonic decoding with geometry-aware neural networks
Geometry-aware neural decoding for bosonic quantum error correction, combining Riemannian structure, robust real-time inference, and structure-aware machine learning under realistic quantum noise and hardware constraints.
Constant Curvature Curve Tube Codes for Low-Latency Analog Error Correction
Introduces constant-curvature curve tube codes for low-latency analog error correction, using geometric signal representations to balance local sensitivity, global separation, and robustness under channel noise.
Undervisning
Robust estimering, livslang læring.
Formidling
Communications with Quaternions
Animation for the IEEE Transactions on Antennas and Propagation article “Polarization-Space Modulation with Reconfigurable Intelligent Surfaces, Robust Estimation, and Quaternion Neural Networks,” featured on the IEEE TAP homepage in Fall 2025.
Introduction to Reconfigurable Intelligent Surfaces
Invited IEEE Pikes Peak Section presentation: “Introduction to Reconfigurable Intelligent Surfaces” (2023), covering RIS fundamentals, applications, and emerging research directions in wireless communications.