Diluka Galappaththige, Ph.D.

Diluka Galappaththige, Ph.D.

Post-Doctoral Research Fellow

Current Research

Set Transformer-Based Beamforming Design for Cell-Free Integrated Sensing and Communication

This work introduces STCIB, the first Set Transformer-based beamforming framework for cell-free integrated sensing and communication (CF-ISAC). Instead of relying on computationally heavy iterative optimization or locally constrained learning architectures, the method models the CF-ISAC network as an unordered set and uses global self-attention to capture interactions among distributed access points, users, and sensing targets.

The resulting architecture is permutation-invariant, operates in an unsupervised manner, and produces beamforming solutions through a single forward pass. In this view, the method transforms estimated channels into set-structured features, propagates information through attention blocks, and outputs a power-constrained beamforming matrix suitable for sensing-centric, communication-centric, and joint ISAC regimes.

Paper: Set Transformer-Based Beamforming Design for Cell-Free Integrated Sensing and Communication.

STCIB conceptual view from unordered channels to attention-based beamforming AP₁ AP₂ AP₃ U₁ U₂ U₃ target channel estimates after I/Q transform
set-structured input, no preferred order iterative solver repeat update high online complexity local interaction bias nearby links are easy; global coupling is weak global self-attention
set transformer pipeline input ISAB encoder PMA / decoder
single-pass beamforming beamforming matrix
AP beam user rate sensing
Current step

Optimization on the Manifold for ISAC

This work develops a Riemannian manifold approach for constrained resource allocation in integrated sensing and communication (ISAC). Rather than optimizing beamforming directly in a difficult non-convex Euclidean space, the method reformulates the beamforming variable on a structured manifold and updates it through geometric operations such as the Riemannian gradient, vector transport, and retraction.

An augmented Lagrangian mechanism is then used to enforce sensing beampattern targets, user SINR constraints, and transmit power limits during the iterative process. The resulting view is both algorithmic and geometric: the iterate moves along tangent directions, retracts back to the manifold, and updates its multipliers until the communication and sensing constraints are satisfied.

Paper: A Riemannian Manifold Approach to Constrained Resource Allocation in ISAC.

beamforming matrix append auxiliary vector z
constrained on
vector transport
retraction
constraint feedback iterate until convergence IALMO manifold view
Current step

Integrated Sensing and Backscatter Communication

This research line introduces integrated sensing and backscatter communication (ISABC), where a base station simultaneously supports communication and sensing through the same transmitted waveform. Instead of treating the reflected signal as a by-product, the reflected signal from passive tags becomes a shared resource: it carries data to the user while also revealing environmental or tag-state information to the base station.

Across the letter, journal, and magazine works, the idea evolves from a first system model to optimized multi-tag beamforming and then to a broader ambient-IoT vision. The animation below highlights that progression through a single visual story.

Papers:
- Integrated Sensing and Backscatter Communication.
- Transmit Power-Efficient Beamforming Design for Integrated Sensing and Backscatter Communication.
- Optimization of Rate-Splitting Multiple Access with Integrated Sensing and Backscatter Communication.
- Dual Function of Sensing and Backscatter Communication in Cellular Networks.

From ISAC and Backscatter to ISABC FD BS joint transmitter / sensor User / Reader data decoding Target sensing only Backscatter Tag passive sensing + data T₁ T₂ Tₖ same waveform drives both tasks communication to user + sensing at the BS joint beamforming + sensing + reflection design multi-tag ISABC optimization smart home healthcare warehouse IoT direct communication backscatter path sensing path
Current step