arXiv subjects

Giovanni Di Gennaro

Publications and source records attributed to Giovanni Di Gennaro.

11 recordsLinked to original sources

arXiv2026

Exact Payload-Decoupling Conditions for Pilot-Only BEM Channel Estimation With Application to OTFS

In high-mobility doubly dispersive links, basis expansion models (BEMs) reduce channel dimensionality, yet unknown payload symbols generally contaminate conventional matched-pilot channel estimates. This paper establishes the exact conditions under which such estimates become payload-independent and derives a pilot, guard, and data-placement rule that guarantees these conditions. We prove a necessary-and-sufficient zero pilot--data interference (ZPDI) condition under which the matched-pilot least-squares (LS) solution coincides with the maximum-likelihood (ML) estimator for the reduced pilot statistic. When ZPDI holds, estimation requires a single precomputed projection. When it does not, the estimate contains a deterministic, channel-scaled payload bias that persists at high signal-to-noise ratio. A disjoint-support rule, independent of the selected basis, realizes ZPDI through pilot, guard, and data placement. We then specialize the framework to orthogonal time--frequency space (OTFS) and examine its structural and performance consequences. With the generalized complex-exponential BEM (GCE-BEM), the ZPDI estimator remains within about $2$~dB of the perfect channel state information benchmark in bit error rate at speeds up to 500~km/h.

cs.IT↗

arXiv2026

Unsupervised Learning for Scalable Downlink Power Control in Cell-Free Massive MIMO

In cell-free massive multiple-input multiple-output systems, downlink power control is essential to ensure uniformly high service quality across users. Existing methods range from centralized iterative approaches requiring global channel knowledge and supervised training, to simpler distributed strategies such as fractional power control that rely on local information but perform poorly in terms of fairness. This letter proposes an unsupervised, physics-informed framework that directly optimizes max-min fairness without requiring optimal labels or user position information. The method is inherently scalable in the number of user equipment, does not require retraining when the user population changes, and can be extended to achieve full scalability with respect to both access points and users. Numerical results show that it nearly doubles the worst-user spectral efficiency compared to existing scalable schemes.

eess.SP↗

arXiv2025

Probing Information Distribution in Transformer Architectures through Entropy Analysis

This work explores entropy analysis as a tool for probing information distribution within Transformer-based architectures. By quantifying token-level uncertainty and examining entropy patterns across different stages of processing, we aim to investigate how information is managed and transformed within these models. As a case study, we apply the methodology to a GPT-based large language model, illustrating its potential to reveal insights into model behavior and internal representations. This approach may offer insights into model behavior and contribute to the development of interpretability and evaluation frameworks for transformer-based models

cs.CL↗

arXiv2025

A General Framework for Scalable UE-AP Association in User-Centric Cell-Free Massive MIMO based on Recurrent Neural Networks

This study addresses the challenge of access point (AP) and user equipment (UE) association in cell-free massive MIMO networks. It introduces a deep learning algorithm leveraging Bidirectional Long Short-Term Memory cells and a hybrid probabilistic methodology for weight updating. This approach enhances scalability by adapting to variations in the number of UEs without requiring retraining. Additionally, the study presents a training methodology that improves scalability not only with respect to the number of UEs but also to the number of APs. Furthermore, a variant of the proposed AP-UE algorithm ensures robustness against pilot contamination effects, a critical issue arising from pilot reuse in channel estimation. Extensive numerical results validate the effectiveness and adaptability of the proposed methods, demonstrating their superiority over widely used heuristic alternatives.

cs.LG↗

arXiv2024

Approaching Massive MIMO Performance with Reconfigurable Intelligent Surfaces: We Do Not Need Many Antennas

This paper considers an antenna structure where a (non-large) array of radiating elements is placed at short distance in front of a reconfigurable intelligent surface (RIS), herein nicknamed reconfigurable intelligent base station (RIBS). We firstly derive a closed-form expression for the channel between the array of radiating elements and the RIS that captures the near-field effects, and give some considerations on the channel hardening and favorable propagation in this scenario. Focusing on both active and passive RIS, we describe channel estimation and downlink signal processing techniques suitable for the RIBS structure. Additionally, we formulate and solve an optimization problem aimed at maximizing the fairness among the users with respect to the downlink power coefficients and RIS configuration both in the cases of active and passive RIBS. Numerical results show that the proposed structure is effective and capable of outperforming conventional non-RIS aided MIMO systems, especially in the case of active RIBS. The proposed antenna structure is thus shown to be able to approach massive MIMO performance levels in a cost-effective way with reduced hardware resources.

cs.IT↗

arXiv2024

A Deep Learning Approach for User-Centric Clustering in Cell-Free Massive MIMO Systems

Contrary to conventional massive MIMO cellular configurations plagued by inter-cell interference, cell-free massive MIMO systems distribute network resources across the coverage area, enabling users to connect with multiple access points (APs) and boosting both system capacity and fairness across user. In such systems, one critical functionality is the association between APs and users: determining the optimal association is indeed a combinatorial problem of prohibitive complexity. In this paper, a solution based on deep learning is thus proposed to solve the user clustering problem aimed at maximizing the sum spectral efficiency while controlling the number of active connections. The proposed solution can scale effectively with the number of users, leveraging long short-term memory cells to operate without the need for retraining. Numerical results show the effectiveness of the proposed solution, even in the presence of imperfect channel state information due to pilot contamination.

eess.SP↗

arXiv2021

A Unified View of Algorithms for Path Planning Using Probabilistic Inference on Factor Graphs

Even if path planning can be solved using standard techniques from dynamic programming and control, the problem can also be approached using probabilistic inference. The algorithms that emerge using the latter framework bear some appealing characteristics that qualify the probabilistic approach as a powerful alternative to the more traditional control formulations. The idea of using estimation on stochastic models to solve control problems is not new and the inference approach considered here falls under the rubric of Active Inference (AI) and Control as Inference (CAI). In this work, we look at the specific recursions that arise from various cost functions that, although they may appear similar in scope, bear noticeable differences, at least when applied to typical path planning problems. We start by posing the path planning problem on a probabilistic factor graph, and show how the various algorithms translate into specific message composition rules. We then show how this unified approach, presented both in probability space and in log space, provides a very general framework that includes the Sum-product, the Max-product, Dynamic programming and mixed Reward/Entropy criteria-based algorithms. The framework also expands algorithmic design options for smoother or sharper policy distributions, including generalized Sum/Max-product algorithm, a Smooth Dynamic programming algorithm and modified versions of the Reward/Entropy recursions. We provide a comprehensive table of recursions and a comparison through simulations, first on a synthetic small grid with a single goal with obstacles, and then on a grid extrapolated from a real-world scene with multiple goals and a semantic map.

cs.LG↗

arXiv2020

Path Planning Using Probability Tensor Flows

Probability models have been proposed in the literature to account for "intelligent" behavior in many contexts. In this paper, probability propagation is applied to model agent's motion in potentially complex scenarios that include goals and obstacles. The backward flow provides precious background information to the agent's behavior, viz., inferences coming from the future determine the agent's actions. Probability tensors are layered in time in both directions in a manner similar to convolutional neural networks. The discussion is carried out with reference to a set of simulated grids where, despite the apparent task complexity, a solution, if feasible, is always found. The original model proposed by Attias has been extended to include non-absorbing obstacles, multiple goals and multiple agents. The emerging behaviors are very realistic and demonstrate great potentials of the application of this framework to real environments.

cs.LG↗

arXiv2020

Intent Classification in Question-Answering Using LSTM Architectures

Question-answering (QA) is certainly the best known and probably also one of the most complex problem within Natural Language Processing (NLP) and artificial intelligence (AI). Since the complete solution to the problem of finding a generic answer still seems far away, the wisest thing to do is to break down the problem by solving single simpler parts. Assuming a modular approach to the problem, we confine our research to intent classification for an answer, given a question. Through the use of an LSTM network, we show how this type of classification can be approached effectively and efficiently, and how it can be properly used within a basic prototype responder.

cs.CL↗

arXiv2020

An Analysis of Word2Vec for the Italian Language

Word representation is fundamental in NLP tasks, because it is precisely from the coding of semantic closeness between words that it is possible to think of teaching a machine to understand text. Despite the spread of word embedding concepts, still few are the achievements in linguistic contexts other than English. In this work, analysing the semantic capacity of the Word2Vec algorithm, an embedding for the Italian language is produced. Parameter setting such as the number of epochs, the size of the context window and the number of negatively backpropagated samples is explored.

cs.CL↗

arXiv2019

Optimized Realization of Bayesian Networks in Reduced Normal Form using Latent Variable Model

Bayesian networks in their Factor Graph Reduced Normal Form (FGrn) are a powerful paradigm for implementing inference graphs. Unfortunately, the computational and memory costs of these networks may be considerable, even for relatively small networks, and this is one of the main reasons why these structures have often been underused in practice. In this work, through a detailed algorithmic and structural analysis, various solutions for cost reduction are proposed. An online version of the classic batch learning algorithm is also analyzed, showing very similar results (in an unsupervised context); which is essential even if multilevel structures are to be built. The solutions proposed, together with the possible online learning algorithm, are included in a C++ library that is quite efficient, especially if compared to the direct use of the well-known sum-product and Maximum Likelihood (ML) algorithms. The results are discussed with particular reference to a Latent Variable Model (LVM) structure.

stat.ML↗