Institution profile

KTH Royal Institute of Technology

Academic institutioneurope · se
Official website
Research library1,076linked papers
Opportunities0open roles
Selected work

Representative Papers

Joint Processing and Transmission Energy Optimization for ISAC in Cell-Free Massive MIMO with URLLC

Jan 18, 2024arXiv.org

This work addresses energy-efficient integrated sensing and communication (ISAC) in cell-free massive MIMO downlink systems under joint ultra-reliable low-latency communication (URLLC) and multistatic sensing constraints. Method: It proposes the first end-to-end energy-saving optimization framework jointly modeling both sensing processing energy consumption and communication transmission energy consumption. To tackle the non-convex joint optimization of transmit power and transmission blocklength, two efficient algorithms are developed—based on feasible point pursuit-successive convex approximation (FPP-SCA) and concave–convex procedure (CCP)—with fractional programming incorporated to handle the energy-efficiency ratio objective. Contribution/Results: The proposed joint design significantly reduces total energy consumption compared to conventional decoupled approaches. Numerical results show that increasing the number of access points raises sensing energy consumption; raising the sensing SINR threshold enlarges total energy consumption while narrowing the performance gap between the two algorithms.

10 citations1 influentialRead paper

Machine-Learning-Based Condition Monitoring of Power Electronics Modules in Modern Electric Drives

Mar 01, 2023IEEE Power Electronics Magazine

To address the reliance on external sensors for temperature monitoring in electric drive power modules and the difficulty of early detection of thermal management anomalies, this paper proposes a lightweight, data-driven thermal modeling approach leveraging only built-in electrical signals (e.g., current, voltage, and switching states). The method integrates linear regression, LSTM, and fully connected neural networks, with architecture and inference optimized for embedded platforms—enabling a hardware-modification-free thermal digital twin for real-time case-temperature estimation and autonomous anomaly diagnosis. Experimental results show prediction errors below 1.2 °C under both static and dynamic operating conditions, and 100% accuracy in early warning of typical cooling failures (e.g., air duct blockage). This work presents the first purely built-in-signal-driven inference of power module thermal states and precursor-level fault identification, significantly enhancing system reliability and practical deployability.

5 citations1 influentialRead paper

Solving partial differential equations with sampled neural networks

May 31, 2024arXiv.org

To address the gradient optimization difficulties and non-causal temporal treatment inherent in physics-informed neural networks (PINNs) for time-dependent partial differential equations (PDEs), this work proposes a gradient-free, causally structured stochastic neural basis function method. Spatially, it constructs neural basis functions with random weights in the hidden layer; temporally, it explicitly integrates time evolution via classical ODE solvers. We introduce a novel dual-mode weight sampling strategy—data-agnostic and data-aware—and establish its $L^2$ convergence in Barron space theoretically. The method combines mesh-free flexibility with spectral convergence accuracy. It enables long-time-domain simulation and inverse problem solving. Numerical experiments across diverse elliptic and time-dependent PDEs demonstrate 1–2 orders-of-magnitude improvements in training speed and accuracy over PINNs, alongside strong generalization capability and numerical stability.

5 citationsRead paper

Geometry of Lightning Self-Attention: Identifiability and Dimension

Aug 30, 2024arXiv.org

This work investigates the functional-space geometric structure of normalization-free self-attention networks, focusing on identifiability and dimensionality characterization. Methodologically, it introduces—within an algebraic geometry framework—the first polynomial mapping model for deep self-attention, rigorously deriving the universal fiber structure of parameter-to-function mappings for arbitrary depth, thereby fully resolving functional identifiability. It provides a closed-form formula for the dimension of the induced function space, completely characterizes the singularity set and boundary structure of single-layer models, theoretically proves the single-layer normalization conjecture, and numerically validates its plausibility in deeper architectures via computational algebraic techniques. The core contribution is the establishment of the first algebraic-geometric theoretical framework for self-attention, furnishing a rigorous geometric foundation for understanding the intrinsic representational capacity of deep attention mechanisms.

4 citationsRead paper

Ice-Breakers, Turn-Takers and Fun-Makers: Exploring Robots for Groups with Teenagers

Aug 29, 2022IEEE International Symposium on Robot and Human Interactive Communication

This study investigates how social robots can support adolescent group interactions to foster identity development and self-esteem. We conducted a two-week summer camp employing participatory methods—including focus groups, in-depth interviews, adolescent-led co-design sessions (10+ hours), and Wizard-of-Oz prototype testing—to systematically uncover dynamic interaction needs across ice-breaking, turn-taking, and engagement-fostering scenarios. To our knowledge, this is the first long-term, adolescent-centered co-design study of social robots for group settings. Findings reveal adolescents’ expectations of robot roles form a dynamic spectrum, necessitating adaptive functionality aligned with group developmental stages (forming → norming → performing). We identify three context-dependent core assistive functions, empirically demonstrate adolescents’ capacity to actively reinterpret and reconfigure robot roles, and propose a transferable “group–robot interaction stage model” alongside four evidence-based design principles.

4 citationsRead paper
Recent publications

Latest Papers