Closed-Form PARAFAC (CANDECOMP): Development of a fast and robust algorithm for tensor
factorization.
Tensor decompositions for biomedical data: Processing of biomedical signals
(such as EEG, MEG, EMG, EKG) using tensor decompositions with the goal to separate
components within the signals.
High-resolution parameter estimation:
Tensor-based subspace estimation: Enhancing subspace-based parameter estimation
methods by applying the HOSVD (Higher-Order Singular Value Decomposition) for improved subspace estimation.
Tensor-ESPRIT: Reformulation of ESPRIT-type methods using tensors which led to
standard Tensor-ESPRIT, Unitary Tensor-ESPRIT and Tensor-Structure Structured Least Squares (TS-SLS).
ESPRIT-type algorithms for non-circular sources: NC Unitary ESPRIT.
Cramér-Rao Bounds: Derivation of a deterministic Cramér-Rao Bound for strict-sense non-circular sources.
Near-field localization: Near-field localization based on Unitary ESPRIT.
Multi-User MIMO communication for Local Area scenarios: 4G system design for an indoor
(enterprise) communication scenario using distributed antennas; including:
linear precoding (regularized block diagonalization or Successive MMSE) in the downlink, linear decoding
(regularized block diagonalization or Successive MMSE) and non-linear decoding
(successive interference cancellation) in the uplink, pilots and channel estimation (spatial pilot reuse),
scheduling and resource allocation (opportunistic, Proportional Fair). This work was carried
out within the European research project WINNER-2.
MIMO strategies for voluntary infrastructure sharing between operators in the multi-pair two-way relaying
and the multi-pair interference channel settings. This work was carried
out within the European research project SAPHYRE.
Two-Way Relaying:
Tensor-based signal processing for two-way relaying: Channel estimation for two-way MIMO relaying (TENCE).
Beamforming-Design for MIMO-Amplify-and-Forward-Relays in Two-Way Relaying:
Algebraic Norm Maximizing (ANOMAX) transmit strategy, Rank-Restored ANOMAX (RR-ANOMAX),
Rate-Maximization via Generalized Eigenvectors for Single-Antenna Terminals in Two-Way Relaying (RAGES).
Compressed Sensing
Optimized measurement matrix design (in particular for DOA and TDOA estimation)
Grid offset estimation for gridded sparse recovery algorithms
Tensor-based dictionary learning for separable manifolds
Sparsity order estimation from compressed measurements
Applications: DOA/TDOA estimation (Radar, Sounding), Spectrum sensing, Synchronization, Channel estimation,
X-ray tomography and ultrasound (in particular for non-destructive testing),
Sheet-of-Light Surface Scanning
