Searcharxiv⌕ Search

arXiv subjects

Vladimir Poulkov

Publications and source records attributed to Vladimir Poulkov.

2 recordsLinked to original sources

Wireless AI Networks: Co-Adapting Communication Resources and Distributed Intelligence

Distributed artificial intelligence (AI) increasingly relies on wireless networks, where communication and computing resources shape how AI tasks are configured and executed. However, resource availability alone does not determine how effectively these tasks can be supported, since different information selections and transmission strategies may yield different AI performance under the same resource constraints. To address this challenge, this article presents Wireless AI Networks as a unified design perspective for bidirectional adaptation between wireless communication and distributed intelligence. Specifically, an AI-executable resource budget characterizes candidate AI configurations, while task-level information value guides their selection and wireless resource allocation. Building on these principles, a wireless-AI co-adaptation framework coordinates resource abstraction, joint configuration and allocation, and execution feedback. Two complementary experiments illustrate this perspective: update selection and statistical correction improve federated learning under identical model-update payload budgets, while sensitivity-aware power allocation reduces model accuracy degradation under the same average transmit-power constraint. Together, these results highlight the importance of coordinating resource feasibility and task-level information value in distributed AI execution.

cs.AI↗

Integrated Sensing and Communication via Channel Knowledge Maps Built from Beam Feedback Coefficients

Channel knowledge maps (CKMs) support wireless localization by linking channel features to user locations. Detailed channel fingerprints, however, can require substantial storage. This paper proposes a CKM-based localization framework that uses beam feedback coefficients collected during communication. These coefficients serve as compact location fingerprints within the beam management process, without requiring full channel feedback or additional sensing measurements. Using a coarse position prior, the base station searches nearby CKM entries and refines the location estimate through local interpolation. We analyze the relationship between spatial fingerprint variation and localization accuracy, and evaluate the effect of feedback quantization. Simulations show favorable storage--accuracy trade-offs relative to dense and sparsified spectral fingerprinting in the tested multipath environments.

eess.SP↗