Showing 1–45 of 45 results for author: Petrov, V

Do Sequential Recommendation Benchmarks Really Require Higher-Order Sequence Modelling?

Authors: Aleksandr V. Petrov, Praveen Chandar, Paul N. Bennett, Hugues Bouchard, Mounia Lalmas

Abstract: Sequential recommenders increasingly use language-model architectures designed to capture complex, context-dependent interactions. Yet it remains unclear whether widely used benchmarks actually require this modelling capacity. We investigate this question using two simple, recency-weighted pairwise probes that do not learn higher-order sequence representations: Sequential Rules (SeqRules) and our… ▽ More Sequential recommenders increasingly use language-model architectures designed to capture complex, context-dependent interactions. Yet it remains unclear whether widely used benchmarks actually require this modelling capacity. We investigate this question using two simple, recency-weighted pairwise probes that do not learn higher-order sequence representations: Sequential Rules (SeqRules) and our Probabilistic Collaborative Transition Model (PCTM). Using the evaluation protocol of eSASRec, at least one probe exceeds our eSASRec reproduction by 15-38% on three Amazon datasets and by 4.4% on MovieLens-1M, but trails it by 27.3% on MovieLens-20M. On the four remaining datasets, at least one probe also outperforms our sampled-softmax SASRec reproduction by 9-28%, suggesting that these widely used benchmarks are poorly suited to measuring gains from higher-order sequence modelling. More broadly, comparing Transformer-based models against strong recency-weighted pairwise probes provides a concrete test of whether a benchmark can meaningfully measure gains from higher-order sequence modelling. △ Less

Submitted 20 August, 2026; originally announced August 2026.

Comments: Accepted at the 20th ACM Conference on Recommender Systems (RecSys 2026)

Hypothesis-Driven Shelf Generation for Personalised Recommendation

Authors: Aleksandr V. Petrov, Tarun Chillara, Matthew D. Moellman, Lucas de Haas, Yabai Song, Alina Susoykina, Melissa Crawford, Gabriel Negash, Erik Franco, Tasnim Rahman, Binal Jhaveri, Shubham Bansal, Hugues Bouchard, Roberto Mirizzi, Mounia Lalmas, Aloïs Gruson

Abstract: Modern recommendation interfaces organise content into shelves: themed rows such as "More of What You Like" or "New Releases for You." In production systems, these shelves are typically defined through hand-crafted templates coupled with dedicated retrieval logic. While effective for broad recommendation intents, this approach does not scale to the long tail of individual taste. We present a conte… ▽ More Modern recommendation interfaces organise content into shelves: themed rows such as "More of What You Like" or "New Releases for You." In production systems, these shelves are typically defined through hand-crafted templates coupled with dedicated retrieval logic. While effective for broad recommendation intents, this approach does not scale to the long tail of individual taste. We present a content-hypothesis-driven shelf generation system for Spotify Home that replaces fixed templates with natural-language hypotheses describing what a personalised shelf should contain. The system has four stages hypothesis generation, catalogue fulfilment, shelf alignment, and offline serving. This decomposition decouples shelf planning from catalogue fulfilment, supports independent optimisation of planning and retrieval, and enables both constrained generative retrieval over catalogue entities and distillation of frontier LLM behaviour into compact models. Our production pipeline combines hypothesis generation, generative retrieval, candidate selection and shelf alignment, offline LLM-as-a-judge evaluation, and precomputed serving. We describe the end-to-end architecture and evaluate it through offline analyses and an early online evaluation under uniform random exposure on Spotify Home. Results show that hypothesis-driven shelves substantially expand personalised recommendation supply with engagement that varies by content type and is competitive with strong existing shelves in some settings. △ Less

Submitted 28 July, 2026; originally announced July 2026.

Comments: Accepted at ACM RecSys '26 (Industry Track)

Bessel Beam Optimization for Near-Field THz Communications under UE Location Uncertainty

Authors: Aditya Jolly, Vitaly Petrov, Gábor Fodor, Emil Björnson

Abstract: To achieve the desired coverage and capacity levels, future terahertz (THz) wireless systems are envisioned to utilize extremely large antenna arrays. At THz frequencies, the combination of short wavelengths and large array apertures often makes many of the conventional far-field assumptions invalid in practice. As a result, many UEs operate in the radiative near-field zone, where novel near-field… ▽ More To achieve the desired coverage and capacity levels, future terahertz (THz) wireless systems are envisioned to utilize extremely large antenna arrays. At THz frequencies, the combination of short wavelengths and large array apertures often makes many of the conventional far-field assumptions invalid in practice. As a result, many UEs operate in the radiative near-field zone, where novel near-field beam synthesis methods become viable. This paper studies phase-only Bessel-like near-field beam configurations for downlink THz multiple-input multiple-output links under imperfect UE location knowledge. We first formulate a spectral efficiency maximization problem with respect to the "Bessel cone angle''. We then derive low-complexity closed-form approximations for the optimal Bessel beam configuration for: (i)deterministic UE location; (ii)Gaussian and (iii)uniform error in the UE location. Finally, through extensive simulations across multiple signal frequencies, UE locations, and array sizes, we show that our proposed simple closed-form approximations closely match (under 0.1% difference) the best performance achieved via exhaustive search, while simultaneously reducing the configuration complexity down to as low as O(1). △ Less

Submitted 9 July, 2026; v1 submitted 8 July, 2026; originally announced July 2026.

Comments: Accepted to IEEE VTC Fall, 2026. Copyright 2026 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material, creating new works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

t-STEP: An interpretable model for Total Electron Content predictions and irregularities estimations

Authors: Stephen Tete, Carl Shneider, Maxime Cordy, Claudio Cesaroni, Andreas Hein, Vasily Petrov

Abstract: Earth system infrastructures relying on satellite-based technologies, such as Global Positioning System (GPS) communications, are affected by ionospheric Total Electron Content (TEC) gradients. Modeling these gradients under physical constraints remains challenging due to their dynamic and transient nature. While existing machine learning (ML) models can predict hourly TEC variations, it remains u… ▽ More Earth system infrastructures relying on satellite-based technologies, such as Global Positioning System (GPS) communications, are affected by ionospheric Total Electron Content (TEC) gradients. Modeling these gradients under physical constraints remains challenging due to their dynamic and transient nature. While existing machine learning (ML) models can predict hourly TEC variations, it remains unclear whether their temporal resolution is sufficient to preserve small-scale TEC irregularities within predicted signals. To address this gap, we introduce an interpretable ML-based model, t-STEP, designed to predict TEC at a 30-second resolution and estimate irregularity signatures from the modeled signals. This high cadence enables the derivation of Rate of TEC changes (ROT) and the ROT Index (ROTI) as diagnostic indicators of ionospheric variability. The model is developed using GPS observations from solar cycle 24 at a station located at 5.49°S, 47.49°W. A multi-metric evaluation framework, including dynamic time warping, is used for robustness assessment, while SHAP (SHapley Additive exPlanations) provides insight into feature contributions. The 30-second TEC predictions achieve 91% accuracy with a mean absolute error (MAE) of 4.38 TECU during high solar activity (2015). Compared with the International Reference Ionosphere (IRI-2020), the hourly model improves accuracy by 35%, reduces absolute errors by 57%, and increases prediction skill by 54%. More importantly, the 30-second model captures TEC irregularity dynamics and morphologies during geomagnetic storms of different intensities, outperforming an attention-based Long Short-Term Memory model under the same experimental conditions. This study demonstrates the potential of a single TEC prediction framework for scalable irregularity monitoring without requiring separate models for individual transient events. △ Less

Submitted 28 June, 2026; originally announced June 2026.

Comments: 40 pages, 15 figures. Note that the article has been published in the Earth Science Informatics Journal

MSC Class: 68T07; 62M10; 86A32 ACM Class: I.2.6; I.2.8; I.2.1

Journal ref: Earth Sci Inform 19, 121 (2026)

Sub Terahertz LEO Satellite Communication: Vision, Opportunities, and Challenges toward the First Prototype in Space

Authors: Sergi Aliaga, Vitaly Petrov, Andrew Benincasa, Albert Diez Comas, Ali J. Alqaraghuli, Jose V. Siles, Ken R. Duffy, Marc Sanchez Net, Tommaso Melodia, Josep M. Jornet

Abstract: The landscape of sub-terahertz (sub-THz, 100GHz - 300GHz) wireless technology evolved drastically over the last two decades - from only a few niche use cases in sensing and ultra-short-range communications in early 2000s toward operational multi-kilometer range 100GBbit/s+ wireless backhaul links demonstrated recently. Building on this momentum, this article explores the feasibility of extending s… ▽ More The landscape of sub-terahertz (sub-THz, 100GHz - 300GHz) wireless technology evolved drastically over the last two decades - from only a few niche use cases in sensing and ultra-short-range communications in early 2000s toward operational multi-kilometer range 100GBbit/s+ wireless backhaul links demonstrated recently. Building on this momentum, this article explores the feasibility of extending sub-THz communications to 100-km-scale satellite links. We first assess the technological readiness of emerging sub-THz hardware and signal-processing techniques, highlighting their potential to support long-range operation in low-Earth-orbit (LEO) systems. We then outline the unique role that sub-THz links can play as a complementary solution to existing millimeter-wave and optical (``laser'') satellite technologies, offering additional capacity, improved resilience, and new architectural flexibility. We further discuss open research and engineering challenges toward implementing such sub-THz satellite communication systems in practice. We finally outline the key state-of-the-art solutions and the roadmap of TeraLink, an ongoing international R&D project aiming to build and launch, through an approved NASA CSLI space mission, the first hardware prototype of sub-THz LEO satellite communications in space. △ Less

Submitted 9 July, 2026; v1 submitted 13 June, 2026; originally announced June 2026.

Comments: 6 Pages, 5 Figures, 1 Tables. The work has been submitted to IEEE Wireless Communications Magazine

Fast Adversarial Attacks with Gradient Prediction

Authors: Kamil Ciosek, Aleksandr V. Petrov, Nicolò Felicioni, Konstantina Palla

Abstract: Generating adversarial examples at scale is a core primitive for robustness evaluation, adversarial training, and red-teaming, yet even "fast" attacks such as FGSM remain throughput-limited by the cost of a backward pass. We introduce a family of attacks that eliminates the backward pass by predicting the input gradient from forward-pass hidden states via a lightweight linear regression. The appro… ▽ More Generating adversarial examples at scale is a core primitive for robustness evaluation, adversarial training, and red-teaming, yet even "fast" attacks such as FGSM remain throughput-limited by the cost of a backward pass. We introduce a family of attacks that eliminates the backward pass by predicting the input gradient from forward-pass hidden states via a lightweight linear regression. The approach is motivated by a kernel view of neural networks and is exact in the Neural Tangent Kernel regime, while remaining effective for practical finite-width models. Empirically, our methods recover much of FGSM's attack performance while using only a small fraction of the time, corresponding to a $532\%$ increase in throughput. These results suggest gradient prediction as a simple and general route to significantly faster adversarial generation under realistic wall-clock constraints. △ Less

Submitted 14 May, 2026; originally announced May 2026.

Comments: 17 pages

Toward the Internet of Space Things: Performance Analysis of LEO Satellite Relay Networks using mmWave and sub-THz links

Authors: Sergi Aliaga, Ahmad Masihi, Vitaly Petrov, Marc Sanchez Net, Josep M. Jornet

Abstract: As the commercial space economy expands, existing ground-based infrastructure faces severe bottlenecks in supporting the data-intensive continuous connectivity needs of next-generation "space users," including CubeSats, space data centers, and more. Even when utilizing existing Ku-band ground relay networks, the contact time with a CubeSat at low-Earth orbit (LEO) is often still limited to minutes… ▽ More As the commercial space economy expands, existing ground-based infrastructure faces severe bottlenecks in supporting the data-intensive continuous connectivity needs of next-generation "space users," including CubeSats, space data centers, and more. Even when utilizing existing Ku-band ground relay networks, the contact time with a CubeSat at low-Earth orbit (LEO) is often still limited to minutes per day only. This paper analyzes an alternative system design that leverages emerging high-rate millimeter-wave (mmWave) and sub-terahertz (sub-THz) inter-satellite links to build a high-throughput and high-availability satellite-based relay backbone for space vehicles. To evaluate this concept, we develop a comprehensive mathematical framework that jointly incorporates complex time-variant orbital dynamics and mmWave/sub-THz link characteristics. We then derive the key performance indicators, including contact probability, channel capacity, and energy efficiency. The numerical results, cross-verified by computer simulations, demonstrate that such systems can provide improvements of up to several orders of magnitude compared to existing networks of ground stations. Notably, we identify a fundamental bound on download capacity and show that continuous 24/7 connectivity becomes achievable with only ten LEO relay satellites. These findings establish mmWave and sub-THz satellite relay networks as a promising, scalable, and energy-efficient solution, thus unlocking improved connectivity with various space vehicles of tomorrow. △ Less

Submitted 3 May, 2026; originally announced May 2026.

Comments: 13 Pages, 10 Figures, 2 Tables. The work has been submitted to IEEE Transactions on Wireless Communications

Harnessing the Freedom of Non-Uniformity in Monostatic ISAC with Antenna Flexibility

Authors: Zhe Wang, Mahmoud Zaher, Vitaly Petrov, Emil Björnson

Abstract: This paper studies flexible non-uniform array design for monostatic integrated sensing and communication (ISAC) systems. An antenna pool is considered at the base station, where each candidate antenna can be dynamically assigned to transmit, receive, or inactive modes, such that a non-uniform effective array is jointly constructed with the ISAC precoding design. We formulate a sum communication ra… ▽ More This paper studies flexible non-uniform array design for monostatic integrated sensing and communication (ISAC) systems. An antenna pool is considered at the base station, where each candidate antenna can be dynamically assigned to transmit, receive, or inactive modes, such that a non-uniform effective array is jointly constructed with the ISAC precoding design. We formulate a sum communication rate maximization problem by jointly optimizing the ISAC beamforming schemes and antenna-mode assignment under sensing, power, and antenna mode constraints. We develop an alternating-optimization-based solution framework mainly with the aid of weighted minimum mean square error, continuous relaxation-based penalty, and successive convex approximation. Numerical results show that the proposed non-uniform array achieves higher sum-rates than the uniform-array baselines, with particularly large gains when the number of activated antennas is small. Moreover, the proposed non-uniform array can achieve, and in some cases exceed, the performance of uniform array baselines with substantially fewer activated antennas, highlighting geometry-aware non-uniform array design as a compelling alternative to brute-force antenna scaling-based array design. △ Less

Submitted 30 April, 2026; originally announced April 2026.

Comments: 6 pages, 3 figures, submitted to IEEE for possible publication

On the Failure of Topic-Matched Contrast Baselines in Multi-Directional Refusal Abliteration

Authors: Valentin Petrov

Abstract: Inasmuch as the removal of refusal behavior from instruction-tuned language models by directional abliteration requires the extraction of refusal-mediating directions from the residual stream activation space, and inasmuch as the construction of the contrast baseline against which harmful prompt activations are compared has been treated in the existing literature as an implementation detail rather… ▽ More Inasmuch as the removal of refusal behavior from instruction-tuned language models by directional abliteration requires the extraction of refusal-mediating directions from the residual stream activation space, and inasmuch as the construction of the contrast baseline against which harmful prompt activations are compared has been treated in the existing literature as an implementation detail rather than a methodological concern, the present work investigates whether a topically matched contrast baseline yields superior refusal directions. The investigation is carried out on the Qwen~3.5 2B model using per-category matched prompt pairs, per-class Self-Organizing Map extraction, and Singular Value Decomposition orthogonalization. It was found that topic-matched contrast produces no functional refusal directions at any tested weight level on any tested layer, while unmatched contrast on the same model, same extraction code, and same evaluation protocol achieves complete refusal elimination on six layers. The geometric analysis of the failure establishes that topic-matched subtraction cancels the dominant activation component shared between harmful and harmless prompts of the same subject, reducing the extracted direction magnitude below the threshold at which weight-matrix projection perturbs the residual stream. The implications for the design of contrast baselines in abliteration research are discussed. △ Less

Submitted 23 March, 2026; originally announced March 2026.

Statistics Approximation-Enabled Distributed Beamforming for Cell-Free Massive MIMO

Authors: Zhe Wang, Emil Björnson, Jiayi Zhang, Peng Zhang, Vitaly Petrov, Bo Ai

Abstract: We study a distributed beamforming approach for cell-free massive multiple-input multiple-output networks, referred to as Global Statistics & Local Instantaneous information-based minimum mean-square error (GSLI-MMSE). The scenario with multi-antenna access points (APs) is considered over three different channel models: correlated Rician fading with fixed or random line-of-sight (LoS) phase-shifts… ▽ More We study a distributed beamforming approach for cell-free massive multiple-input multiple-output networks, referred to as Global Statistics & Local Instantaneous information-based minimum mean-square error (GSLI-MMSE). The scenario with multi-antenna access points (APs) is considered over three different channel models: correlated Rician fading with fixed or random line-of-sight (LoS) phase-shifts, and correlated Rayleigh fading. With the aid of matrix inversion derivations, we can construct the conventional MMSE combining from the perspective of each AP, where global instantaneous information is involved. Then, for an arbitrary AP, we apply the statistics approximation methodology to approximate instantaneous terms related to other APs by channel statistics to construct the distributed combining scheme at each AP with local instantaneous information and global statistics. With the aid of uplink-downlink duality, we derive the respective GSLI-MMSE precoding schemes. Numerical results showcase that the proposed GSLI-MMSE scheme demonstrates performance comparable to the optimal centralized MMSE scheme, under the stable LoS conditions, e.g., with static users having Rician fading with a fixed LoS path. △ Less

Submitted 4 February, 2026; v1 submitted 3 February, 2026; originally announced February 2026.

Comments: 6 pages, 3 figures, accepted by IEEE International Conference on Communications (ICC) 2026

From Antenna Abundance to Antenna Intelligence in 6G Gigantic MIMO Systems

Authors: Emil Björnson, Amna Irshad, Özlem Tugfe Demir, Giuseppe Thadeu Freitas de Abreu, Alva Kosasih, Vitaly Petrov

Abstract: Current cellular systems achieve high spectral efficiency through Massive MIMO, which leverages an abundance of antennas to create favorable propagation conditions for multiuser spatial multiplexing. Looking towards future networks, the extrapolation of this paradigm leads to systems with many hundreds of antennas per base station, raising concerns regarding hardware complexity, cost, and power co… ▽ More Current cellular systems achieve high spectral efficiency through Massive MIMO, which leverages an abundance of antennas to create favorable propagation conditions for multiuser spatial multiplexing. Looking towards future networks, the extrapolation of this paradigm leads to systems with many hundreds of antennas per base station, raising concerns regarding hardware complexity, cost, and power consumption. This article suggests more intelligent array designs that reduce the need for excessive antenna numbers. We revisit classical uniform array design principles and explain how their uniform spatial sampling leads to unnecessary redundancies in practical deployment scenarios. By exploiting non-uniform sparse arrays with site-specific antenna placements -- based on either pre-optimized irregular arrays or real-time movable antennas -- we demonstrate how superior multiuser MIMO performance can be achieved with far fewer antennas. These principles are inspired by previous works on wireless localization. We explain and demonstrate numerically how these concepts can be adapted for communications to improve the average sum rate and similar metrics. The results suggest a paradigm shift for future antenna array design, where antenna intelligence replaces sheer antenna count. This opens new opportunities for efficient, adaptable, and sustainable Gigantic MIMO systems. △ Less

Submitted 13 January, 2026; originally announced January 2026.

Comments: Submitted for publication, 8 pages, 6 figures

OpenAI GPT-5 System Card

Authors: Aaditya Singh, Adam Fry, Adam Perelman, Adam Tart, Adi Ganesh, Ahmed El-Kishky, Aidan McLaughlin, Aiden Low, AJ Ostrow, Akhila Ananthram, Akshay Nathan, Alan Luo, Alec Helyar, Aleksander Madry, Aleksandr Efremov, Aleksandra Spyra, Alex Baker-Whitcomb, Alex Beutel, Alex Karpenko, Alex Makelov, Alex Neitz, Alex Wei, Alexandra Barr, Alexandre Kirchmeyer, Alexey Ivanov , et al. (461 additional authors not shown)

Abstract: This is the system card published alongside the OpenAI GPT-5 launch, August 2025. GPT-5 is a unified system with a smart and fast model that answers most questions, a deeper reasoning model for harder problems, and a real-time router that quickly decides which model to use based on conversation type, complexity, tool needs, and explicit intent (for example, if you say 'think hard about this' in… ▽ More This is the system card published alongside the OpenAI GPT-5 launch, August 2025. GPT-5 is a unified system with a smart and fast model that answers most questions, a deeper reasoning model for harder problems, and a real-time router that quickly decides which model to use based on conversation type, complexity, tool needs, and explicit intent (for example, if you say 'think hard about this' in the prompt). The router is continuously trained on real signals, including when users switch models, preference rates for responses, and measured correctness, improving over time. Once usage limits are reached, a mini version of each model handles remaining queries. This system card focuses primarily on gpt-5-thinking and gpt-5-main, while evaluations for other models are available in the appendix. The GPT-5 system not only outperforms previous models on benchmarks and answers questions more quickly, but -- more importantly -- is more useful for real-world queries. We've made significant advances in reducing hallucinations, improving instruction following, and minimizing sycophancy, and have leveled up GPT-5's performance in three of ChatGPT's most common uses: writing, coding, and health. All of the GPT-5 models additionally feature safe-completions, our latest approach to safety training to prevent disallowed content. Similarly to ChatGPT agent, we have decided to treat gpt-5-thinking as High capability in the Biological and Chemical domain under our Preparedness Framework, activating the associated safeguards. While we do not have definitive evidence that this model could meaningfully help a novice to create severe biological harm -- our defined threshold for High capability -- we have chosen to take a precautionary approach. △ Less

Submitted 1 May, 2026; v1 submitted 19 December, 2025; originally announced January 2026.

Comments: May 2026: Added monitorability evals and authors

From IR to RecSys: Evaluating LLM-based Judges in Cranfield-style Recommendation Collections

Authors: Gustavo Penha, Aleksandr V. Petrov, Claudia Hauff, Enrico Palumbo, Ali Vardasbi, Edoardo D'Amico, Francesco Fabbri, Alice Wang, Praveen Chandar, Henrik Lindstrom, Hugues Bouchard, Mounia Lalmas

Abstract: The Cranfield paradigm has long provided reliable, reproducible evaluation in ad hoc retrieval, and recent work has begun extending this framework to recommender systems. A recent development in IR is the use of Large Language Models (LLMs) as automatic relevance judges, showing promising agreement with human assessors. Whether this LLM-judge paradigm---studied predominantly on query--document pai… ▽ More The Cranfield paradigm has long provided reliable, reproducible evaluation in ad hoc retrieval, and recent work has begun extending this framework to recommender systems. A recent development in IR is the use of Large Language Models (LLMs) as automatic relevance judges, showing promising agreement with human assessors. Whether this LLM-judge paradigm---studied predominantly on query--document pairs---transfers to the subjective, profile-driven nature of recommendation remains an open question. This paper bridges the IR and RecSys evaluation traditions by systematically investigating LLM-based judges within a Cranfield-style recommendation collection. Using the ML-32M-ext movie recommendation collection, we first demonstrate that traditional train--test splits yield substantially incomplete relevance labels and unreliable system rankings compared to Cranfield-style pooling. We then assess LLM-judge alignment with human labels, finding that richer item metadata and longer user histories improve agreement, although item-level agreement remains moderate overall. Rankings derived from LLM-judge labels achieve high agreement with human-based rankings (Kendall's tau up to 0.92 for nDCG@100 across 52 system configurations), comparable to values reported for TREC ad hoc retrieval collections. Crucially, LLM-judge recovers system rankings that are distorted under traditional evaluation---correctly identifying systems that are undervalued or overvalued by incomplete labels. An industrial case study in podcast recommendation further demonstrates the practical value of LLM-judge for model selection. Rather than positioning LLM-judges as a replacement for human or interaction-based evaluation, our results support their use as a promising complementary signal: item-level agreement with humans is moderate, yet system-level rankings---which aggregate judgments over many user--item pairs---remain stable. △ Less

Submitted 17 August, 2026; v1 submitted 28 November, 2025; originally announced November 2025.

Comments: v2 paper accepted at the RecSys'26 Unified Search & Recommendation Workshop

Impact of Antenna Arrays Misalignment on the Near Field Distance in Terahertz Communications

Authors: Peng Zhang, Vitaly Petrov, Emil Björnson

Abstract: The extremely short wavelength of terahertz (THz) communications leads to an extended radiative near-field region, in which some canonical far-field assumptions fail. Existing near-field boundary formulations (Fraunhofer distance) for uniform linear/planar array (ULA/UPA) configurations assume ideal alignment between transceivers, overlooking practical misalignments caused by mobility or mechanica… ▽ More The extremely short wavelength of terahertz (THz) communications leads to an extended radiative near-field region, in which some canonical far-field assumptions fail. Existing near-field boundary formulations (Fraunhofer distance) for uniform linear/planar array (ULA/UPA) configurations assume ideal alignment between transceivers, overlooking practical misalignments caused by mobility or mechanical imperfections. This paper addresses this critical gap by analyzing the impact of spatial misalignment on near-field distance calculations in THz systems. We derive exact analytical expressions and simplified approximations for the near-field boundary in both ULA--ULA and UPA--UPA configurations under arbitrary misalignment offsets. Through numerical simulations, we validate our theoretical models and quantify how misalignment reshapes the near-field region. These findings provide essential guidelines for optimizing THz system deployment in realistic scenarios. △ Less

Submitted 1 November, 2025; originally announced November 2025.

Comments: Accepted to IEEE Globecom, 2025. Copyright 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material, creating new works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

LLMs for estimating positional bias in logged interaction data

Authors: Aleksandr V. Petrov, Michael Murtagh, Karthik Nagesh

Abstract: Recommender and search systems commonly rely on Learning To Rank models trained on logged user interactions to order items by predicted relevance. However, such interaction data is often subject to position bias, as users are more likely to click on items that appear higher in the ranking, regardless of their actual relevance. As a result, newly trained models may inherit and reinforce the biases… ▽ More Recommender and search systems commonly rely on Learning To Rank models trained on logged user interactions to order items by predicted relevance. However, such interaction data is often subject to position bias, as users are more likely to click on items that appear higher in the ranking, regardless of their actual relevance. As a result, newly trained models may inherit and reinforce the biases of prior ranking models rather than genuinely improving relevance. A standard approach to mitigate position bias is Inverse Propensity Scoring (IPS), where the model's loss is weighted by the inverse of a propensity function, an estimate of the probability that an item at a given position is examined. However, accurate propensity estimation is challenging, especially in interfaces with complex non-linear layouts. In this paper, we propose a novel method for estimating position bias using Large Language Models (LLMs) applied to logged user interaction data. This approach offers a cost-effective alternative to online experimentation. Our experiments show that propensities estimated with our LLM-as-a-judge approach are stable across score buckets and reveal the row-column effects of Viator's grid layout that simpler heuristics overlook. An IPS-weighted reranker trained with these propensities matches the production model on standard NDCG@10 while improving weighted NDCG@10 by roughly 2%. We will verify these offline gains in forthcoming live-traffic experiments. △ Less

Submitted 3 September, 2025; originally announced September 2025.

Comments: Accepted at the CONSEQUENCES Workshop @ RecSys'25

eSASRec: Enhancing Transformer-based Recommendations in a Modular Fashion

Authors: Daria Tikhonovich, Nikita Zelinskiy, Aleksandr V. Petrov, Mayya Spirina, Andrei Semenov, Andrey V. Savchenko, Sergei Kuliev

Abstract: Since their introduction, Transformer-based models, such as SASRec and BERT4Rec, have become common baselines for sequential recommendations, surpassing earlier neural and non-neural methods. A number of following publications have shown that the effectiveness of these models can be improved by, for example, slightly updating the architecture of the Transformer layers, using better training object… ▽ More Since their introduction, Transformer-based models, such as SASRec and BERT4Rec, have become common baselines for sequential recommendations, surpassing earlier neural and non-neural methods. A number of following publications have shown that the effectiveness of these models can be improved by, for example, slightly updating the architecture of the Transformer layers, using better training objectives, and employing improved loss functions. However, the additivity of these modular improvements has not been systematically benchmarked - this is the gap we aim to close in this paper. Through our experiments, we identify a very strong model that uses SASRec's training objective, LiGR Transformer layers, and Sampled Softmax Loss. We call this combination eSASRec (Enhanced SASRec). While we primarily focus on realistic, production-like evaluation, in our preliminarily study we find that common academic benchmarks show eSASRec to be 23% more effective compared to the most recent state-of-the-art models, such as ActionPiece. In our main production-like benchmark, eSASRec resides on the Pareto frontier in terms of the accuracy-coverage tradeoff (alongside the recent industrial models HSTU and FuXi. As the modifications compared to the original SASRec are relatively straightforward and no extra features are needed (such as timestamps in HSTU), we believe that eSASRec can be easily integrated into existing recommendation pipelines and can can serve as a strong yet very simple baseline for emerging complicated algorithms. To facilitate this, we provide the open-source implementations for our models and benchmarks in repository https://github.com/blondered/transformer_benchmark △ Less

Submitted 8 August, 2025; originally announced August 2025.

Comments: Accepted at ACM RecSys 2025

Balancing Accuracy and Novelty with Sub-Item Popularity

Authors: Chiara Mallamaci, Aleksandr Vladimirovich Petrov, Alberto Carlo Maria Mancino, Vito Walter Anelli, Tommaso Di Noia, Craig Macdonald

Abstract: In the realm of music recommendation, sequential recommenders have shown promise in capturing the dynamic nature of music consumption. A key characteristic of this domain is repetitive listening, where users frequently replay familiar tracks. To capture these repetition patterns, recent research has introduced Personalised Popularity Scores (PPS), which quantify user-specific preferences based on… ▽ More In the realm of music recommendation, sequential recommenders have shown promise in capturing the dynamic nature of music consumption. A key characteristic of this domain is repetitive listening, where users frequently replay familiar tracks. To capture these repetition patterns, recent research has introduced Personalised Popularity Scores (PPS), which quantify user-specific preferences based on historical frequency. While PPS enhances relevance in recommendation, it often reinforces already-known content, limiting the system's ability to surface novel or serendipitous items - key elements for fostering long-term user engagement and satisfaction. To address this limitation, we build upon RecJPQ, a Transformer-based framework initially developed to improve scalability in large-item catalogues through sub-item decomposition. We repurpose RecJPQ's sub-item architecture to model personalised popularity at a finer granularity. This allows us to capture shared repetition patterns across sub-embeddings - latent structures not accessible through item-level popularity alone. We propose a novel integration of sub-ID-level personalised popularity within the RecJPQ framework, enabling explicit control over the trade-off between accuracy and personalised novelty. Our sub-ID-level PPS method (sPPS) consistently outperforms item-level PPS by achieving significantly higher personalised novelty without compromising recommendation accuracy. Code and experiments are publicly available at https://github.com/sisinflab/Sub-id-Popularity. △ Less

Submitted 7 August, 2025; originally announced August 2025.

GLoSS: Generative Language Models with Semantic Search for Sequential Recommendation

Authors: Krishna Acharya, Aleksandr V. Petrov, Juba Ziani

Abstract: We propose Generative Low-rank language model with Semantic Search (GLoSS), a generative recommendation framework that combines large language models with dense retrieval for sequential recommendation. Unlike prior methods such as GPT4Rec, which rely on lexical matching via BM25, GLoSS uses semantic search to retrieve relevant items beyond lexical matching. For query generation, we employ 4-bit qu… ▽ More We propose Generative Low-rank language model with Semantic Search (GLoSS), a generative recommendation framework that combines large language models with dense retrieval for sequential recommendation. Unlike prior methods such as GPT4Rec, which rely on lexical matching via BM25, GLoSS uses semantic search to retrieve relevant items beyond lexical matching. For query generation, we employ 4-bit quantized LlaMA-3 models fine-tuned with low-rank adaptation (LoRA), enabling efficient training and inference on modest hardware. We evaluate GLoSS on three real-world Amazon review datasets: Beauty, Toys, and Sports, and find that it achieves state-of-the-art performance. Compared to traditional ID-based baselines, GLoSS improves Recall@5 by 33.3%, 52.8%, and 15.2%, and NDCG@5 by 30.0%, 42.6%, and 16.1%, respectively. It also outperforms LLM-based recommenders such as P5, GPT4Rec, LlamaRec and E4SRec with Recall@5 gains of 4.3%, 22.8%, and 29.5%. Additionally, user segment evaluations show that GLoSS performs particularly well for cold-start users in the Amazon Toys and Sports datasets, and benefits from longer user histories in Amazon Beauty dataset, demonstrating robustness across different levels of interaction lengths. △ Less

Submitted 9 June, 2025; v1 submitted 2 June, 2025; originally announced June 2025.

Comments: Our code and model checkpoints are publicly available at:https://github.com/krishnacharya/GLoSS

Efficient Recommendation with Millions of Items by Dynamic Pruning of Sub-Item Embeddings

Authors: Aleksandr V. Petrov, Craig Macdonald, Nicola Tonellotto

Abstract: A large item catalogue is a major challenge for deploying modern sequential recommender models, since it makes the memory footprint of the model large and increases inference latency. One promising approach to address this is RecJPQ, which replaces item embeddings with sub-item embeddings. However, slow inference remains problematic because finding the top highest-scored items usually requires sco… ▽ More A large item catalogue is a major challenge for deploying modern sequential recommender models, since it makes the memory footprint of the model large and increases inference latency. One promising approach to address this is RecJPQ, which replaces item embeddings with sub-item embeddings. However, slow inference remains problematic because finding the top highest-scored items usually requires scoring all items in the catalogue, which may not be feasible for large catalogues. By adapting dynamic pruning concepts from document retrieval, we propose the RecJPQPrune dynamic pruning algorithm to efficiently find the top highest-scored items without computing the scores of all items in the catalogue. Our RecJPQPrune algorithm is safe-up-to-rank K since it theoretically guarantees that no potentially high-scored item is excluded from the final top K recommendation list, thereby ensuring no impact on effectiveness. Our experiments on two large datasets and three recommendation models demonstrate the efficiency achievable using RecJPQPrune: for instance, on the Tmall dataset with 2.2M items, we can reduce the median model scoring time by 64 times compared to the Transformer Default baseline, and 5.3 times compared to a recent scoring approach called PQTopK. Overall, this paper demonstrates the effective and efficient inference of Transformer-based recommendation models at catalogue scales not previously reported in the literature. Indeed, our RecJPQPrune algorithm can score 2 million items in under 10 milliseconds without GPUs, and without relying on Approximate Nearest Neighbour (ANN) techniques. △ Less

Submitted 1 May, 2025; originally announced May 2025.

Comments: Accepted as a full research paper at SIGIR 2025

Optimizing Movable Antennas in Wideband Multi-User MIMO With Hardware Impairments

Authors: Amna Irshad, Emil Björnson, Alva Kosasih, Vitaly Petrov

Abstract: Movable antennas represent an emerging field in telecommunication research and a potential approach to achieving higher data rates in multiple-input multiple-output (MIMO) communications when the total number of antennas is limited. Most solutions and analyses to date have been limited to \emph{narrowband} setups. This work complements the prior studies by quantifying the benefit of using movable… ▽ More Movable antennas represent an emerging field in telecommunication research and a potential approach to achieving higher data rates in multiple-input multiple-output (MIMO) communications when the total number of antennas is limited. Most solutions and analyses to date have been limited to \emph{narrowband} setups. This work complements the prior studies by quantifying the benefit of using movable antennas in \emph{wideband} MIMO communication systems. First, we derive a novel uplink wideband system model that also accounts for distortion from transceiver hardware impairments. We then formulate and solve an optimization task to maximize the average sum rate by adjusting the antenna positions using particle swarm optimization. Finally, the performance with movable antennas is compared with fixed uniform arrays and the derived theoretical upper bound. The numerical study concludes that the data rate improvement from movable antennas over other arrays heavily depends on the level of hardware impairments, the richness of the multi-path environments, and the number of subcarriers. The present study provides vital insights into the most suitable use cases for movable antennas in future wideband systems. △ Less

Submitted 2 June, 2025; v1 submitted 17 April, 2025; originally announced April 2025.

Comments: To appear at 2025 IEEE International Workshop on Signal Processing and Artificial Intelligence for Wireless Communications (SPAWC), 5 pages, 6 figures

Enhancing Sequential Music Recommendation with Personalized Popularity Awareness

Authors: Davide Abbattista, Vito Walter Anelli, Tommaso Di Noia, Craig Macdonald, Aleksandr Vladimirovich Petrov

Abstract: In the realm of music recommendation, sequential recommender systems have shown promise in capturing the dynamic nature of music consumption. Nevertheless, traditional Transformer-based models, such as SASRec and BERT4Rec, while effective, encounter challenges due to the unique characteristics of music listening habits. In fact, existing models struggle to create a coherent listening experience du… ▽ More In the realm of music recommendation, sequential recommender systems have shown promise in capturing the dynamic nature of music consumption. Nevertheless, traditional Transformer-based models, such as SASRec and BERT4Rec, while effective, encounter challenges due to the unique characteristics of music listening habits. In fact, existing models struggle to create a coherent listening experience due to rapidly evolving preferences. Moreover, music consumption is characterized by a prevalence of repeated listening, i.e., users frequently return to their favourite tracks, an important signal that could be framed as individual or personalized popularity. This paper addresses these challenges by introducing a novel approach that incorporates personalized popularity information into sequential recommendation. By combining user-item popularity scores with model-generated scores, our method effectively balances the exploration of new music with the satisfaction of user preferences. Experimental results demonstrate that a Personalized Most Popular recommender, a method solely based on user-specific popularity, outperforms existing state-of-the-art models. Furthermore, augmenting Transformer-based models with personalized popularity awareness yields superior performance, showing improvements ranging from 25.2% to 69.8%. The code for this paper is available at https://github.com/sisinflab/personalized-popularity-awareness. △ Less

Submitted 6 September, 2024; originally announced September 2024.

Comments: Accepted by RecSys'24 as an LBR paper

Efficient Inference of Sub-Item Id-based Sequential Recommendation Models with Millions of Items

Authors: Aleksandr V. Petrov, Craig Macdonald, Nicola Tonellotto

Abstract: Transformer-based recommender systems, such as BERT4Rec or SASRec, achieve state-of-the-art results in sequential recommendation. However, it is challenging to use these models in production environments with catalogues of millions of items: scaling Transformers beyond a few thousand items is problematic for several reasons, including high model memory consumption and slow inference. In this respe… ▽ More Transformer-based recommender systems, such as BERT4Rec or SASRec, achieve state-of-the-art results in sequential recommendation. However, it is challenging to use these models in production environments with catalogues of millions of items: scaling Transformers beyond a few thousand items is problematic for several reasons, including high model memory consumption and slow inference. In this respect, RecJPQ is a state-of-the-art method of reducing the models' memory consumption; RecJPQ compresses item catalogues by decomposing item IDs into a small number of shared sub-item IDs. Despite reporting the reduction of memory consumption by a factor of up to 50x, the original RecJPQ paper did not report inference efficiency improvements over the baseline Transformer-based models. Upon analysing RecJPQ's scoring algorithm, we find that its efficiency is limited by its use of score accumulators for each item, which prevents parallelisation. In contrast, LightRec (a non-sequential method that uses a similar idea of sub-ids) reported large inference efficiency improvements using an algorithm we call PQTopK. We show that it is also possible to improve RecJPQ-based models' inference efficiency using the PQTopK algorithm. In particular, we speed up RecJPQ-enhanced SASRec by a factor of 4.5 x compared to the original SASRec's inference method and by a factor of 1.56 x compared to the method implemented in RecJPQ code on a large-scale Gowalla dataset with more than a million items. Further, using simulated data, we show that PQTopK remains efficient with catalogues of up to tens of millions of items, removing one of the last obstacles to using Transformer-based models in production environments with large catalogues. △ Less

Submitted 19 August, 2024; originally announced August 2024.

Comments: Accepted by RecSys 2024

The Evolution of Applications, Hardware Design, and Channel Modeling for Terahertz (THz) Band Communications and Sensing: Ready for 6G?

Authors: Josep M. Jornet, Vitaly Petrov, Hua Wang, Zoya Popovic, Dipankar Shakya, Jose V. Siles, Theodore S. Rappaport

Abstract: For decades, the terahertz (THz) frequency band had been primarily explored in the context of radar, imaging, and spectroscopy, where multi-gigahertz (GHz) and even THz-wide channels and the properties of terahertz photons offered attractive target accuracy, resolution, and classification capabilities. Meanwhile, the exploitation of the terahertz band for wireless communication had originally been… ▽ More For decades, the terahertz (THz) frequency band had been primarily explored in the context of radar, imaging, and spectroscopy, where multi-gigahertz (GHz) and even THz-wide channels and the properties of terahertz photons offered attractive target accuracy, resolution, and classification capabilities. Meanwhile, the exploitation of the terahertz band for wireless communication had originally been limited due to several reasons, including (i) no immediate need for such high data rates available via terahertz bands and (ii) challenges in designing sufficiently high power terahertz systems at reasonable cost and efficiency, leading to what was often referred to as "the terahertz gap". This roadmap paper first reviews the evolution of the hardware design approaches for terahertz systems, including electronic, photonic, and plasmonic approaches, and the understanding of the terahertz channel itself, in diverse scenarios, ranging from common indoors and outdoors scenarios to intra-body and outer-space environments. The article then summarizes the lessons learned during this multi-decade process and the cutting-edge state-of-the-art findings, including novel methods to quantify power efficiency, which will become more important in making design choices. Finally, the manuscript presents the authors' perspective and insights on how the evolution of terahertz systems design will continue toward enabling efficient terahertz communications and sensing solutions as an integral part of next-generation wireless systems. △ Less

Submitted 10 June, 2024; originally announced June 2024.

Comments: 31 pages. Accepted to the Proceedings of the IEEE, 2024. \c{opyright}2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material, creating new works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

Electromagnetic Nanonetworks Beyond 6G: From Wearable and Implantable Networks to On-chip and Quantum Communication

Authors: Sergi Abadal, Chong Han, Vitaly Petrov, Laura Galluccio, Ian F. Akyildiz, Josep M. Jornet

Abstract: Emerging from the symbiotic combination of nanotechnology and communications, the field of nanonetworking has come a long way since its inception more than fifteen years ago. Significant progress has been achieved in several key communication technologies as enablers of the paradigm, as well as in the multiple application areas that it opens. In this paper, the focus is placed on the electromagnet… ▽ More Emerging from the symbiotic combination of nanotechnology and communications, the field of nanonetworking has come a long way since its inception more than fifteen years ago. Significant progress has been achieved in several key communication technologies as enablers of the paradigm, as well as in the multiple application areas that it opens. In this paper, the focus is placed on the electromagnetic nanonetworking paradigm, providing an overview of the advances made in wireless nanocommunication technology from microwave through terahertz to optical bands. The characteristics and potential of the compared technologies are then confronted with the requirements and challenges of the broad set of nanonetworking applications in the Internet of NanoThings (IoNT) and on-chip networks paradigms, including quantum computing applications for the first time. Finally, a selection of cross-cutting issues and possible directions for future work are given, aiming to guide researchers and practitioners towards the next generation of electromagnetic nanonetworks. △ Less

Submitted 13 May, 2024; originally announced May 2024.

Comments: 22 pages, 8 figures, 3 tables; Accepted for publication at IEEE Journal on Selected Areas in Communications, 2024

Shallow Cross-Encoders for Low-Latency Retrieval

Authors: Aleksandr V. Petrov, Sean MacAvaney, Craig Macdonald

Abstract: Transformer-based Cross-Encoders achieve state-of-the-art effectiveness in text retrieval. However, Cross-Encoders based on large transformer models (such as BERT or T5) are computationally expensive and allow for scoring only a small number of documents within a reasonably small latency window. However, keeping search latencies low is important for user satisfaction and energy usage. In this pape… ▽ More Transformer-based Cross-Encoders achieve state-of-the-art effectiveness in text retrieval. However, Cross-Encoders based on large transformer models (such as BERT or T5) are computationally expensive and allow for scoring only a small number of documents within a reasonably small latency window. However, keeping search latencies low is important for user satisfaction and energy usage. In this paper, we show that weaker shallow transformer models (i.e., transformers with a limited number of layers) actually perform better than full-scale models when constrained to these practical low-latency settings since they can estimate the relevance of more documents in the same time budget. We further show that shallow transformers may benefit from the generalized Binary Cross-Entropy (gBCE) training scheme, which has recently demonstrated success for recommendation tasks. Our experiments with TREC Deep Learning passage ranking query sets demonstrate significant improvements in shallow and full-scale models in low-latency scenarios. For example, when the latency limit is 25ms per query, MonoBERT-Large (a cross-encoder based on a full-scale BERT model) is only able to achieve NDCG@10 of 0.431 on TREC DL 2019, while TinyBERT-gBCE (a cross-encoder based on TinyBERT trained with gBCE) reaches NDCG@10 of 0.652, a +51% gain over MonoBERT-Large. We also show that shallow Cross-Encoders are effective even when used without a GPU (e.g., with CPU inference, NDCG@10 decreases only by 3% compared to GPU inference with 50ms latency), which makes Cross-Encoders practical to run even without specialized hardware acceleration. △ Less

Submitted 29 March, 2024; originally announced March 2024.

Comments: Accepted by ECIR2024

Modeling Interference from Millimeter Wave and Terahertz Bands Cross-links in Low Earth Orbit Satellite Networks for 6G and Beyond

Authors: Sergi Aliaga, Vitaly Petrov, Josep M. Jornet

Abstract: High-rate satellite communications among hundreds and even thousands of satellites deployed at low-Earth orbits (LEO) will be an important element of the forthcoming sixth-generation (6G) of wireless systems beyond 2030. With millimeter wave communications (mmWave, ~30GHz-100GHz) completely integrated into 5G terrestrial networks, exploration of its potential, along with sub-terahertz (sub-THz, 10… ▽ More High-rate satellite communications among hundreds and even thousands of satellites deployed at low-Earth orbits (LEO) will be an important element of the forthcoming sixth-generation (6G) of wireless systems beyond 2030. With millimeter wave communications (mmWave, ~30GHz-100GHz) completely integrated into 5G terrestrial networks, exploration of its potential, along with sub-terahertz (sub-THz, 100GHz-300GHz), and even THz (300GHz-3THz) frequencies, is underway for space-based networks. However, the interference problem between LEO mmWave/THz satellite cross-links in the same or different constellations is undeservedly forgotten. This article presents a comprehensive mathematical framework for modeling directional interference in all key possible scenario geometries. The framework description is followed by an in-depth numerical study on the impact of cross-link interference on various performance indicators, where the delivered analytical results are cross-verified via computer simulations. The study reveals that, while highly directional mmWave and, especially, THz beams minimize interference in many cases, there are numerous practical configurations where the impact of cross-link interference cannot be neglected and must be accounted for. △ Less

Submitted 7 February, 2024; v1 submitted 21 December, 2023; originally announced December 2023.

Comments: 16 pages, 10 Figures, 2 Tables. The work has been accepted for publication in IEEE Journal on Selected Areas in Communications (JSAC), 2024

RecJPQ: Training Large-Catalogue Sequential Recommenders

Authors: Aleksandr V. Petrov, Craig Macdonald

Abstract: Sequential Recommendation is a popular recommendation task that uses the order of user-item interaction to model evolving users' interests and sequential patterns in their behaviour. Current state-of-the-art Transformer-based models for sequential recommendation, such as BERT4Rec and SASRec, generate sequence embeddings and compute scores for catalogue items, but the increasing catalogue size make… ▽ More Sequential Recommendation is a popular recommendation task that uses the order of user-item interaction to model evolving users' interests and sequential patterns in their behaviour. Current state-of-the-art Transformer-based models for sequential recommendation, such as BERT4Rec and SASRec, generate sequence embeddings and compute scores for catalogue items, but the increasing catalogue size makes training these models costly. The Joint Product Quantisation (JPQ) method, originally proposed for passage retrieval, markedly reduces the size of the retrieval index with minimal effect on model effectiveness, by replacing passage embeddings with a limited number of shared sub-embeddings. This paper introduces RecJPQ, a novel adaptation of JPQ for sequential recommendations, which takes the place of item embeddings tensor and replaces item embeddings with a concatenation of a limited number of shared sub-embeddings and, therefore, limits the number of learnable model parameters. The main idea of RecJPQ is to split items into sub-item entities before training the main recommendation model, which is inspired by splitting words into tokens and training tokenisers in language models. We apply RecJPQ to SASRec, BERT4Rec, and GRU4rec models on three large-scale sequential datasets. Our results showed that RecJPQ could notably reduce the model size (e.g., 48% reduction for the Gowalla dataset with no effectiveness degradation). RecJPQ can also improve model performance through a regularisation effect (e.g. +0.96% NDCG@10 improvement on the Booking.com dataset). Overall, RecJPQ allows the training of state-of-the-art transformer recommenders in industrial applications, where datasets with millions of items are common. △ Less

Submitted 18 December, 2023; v1 submitted 11 December, 2023; originally announced December 2023.

Comments: Accepted by ACM WSDM 2024

Near-field 6G Networks: Why Mobile Terahertz Communications MUST Operate in the Near Field

Authors: Vitaly Petrov, Josep Miquel Jornet, Arjun Singh

Abstract: Near-field mobile terahertz (THz) communications is one of the candidate enablers for high-rate wireless data exchange in sixth-generation (6G) networks. However, operating in the THz near field brings both attractive opportunities and severe challenges. Hence, it becomes of interest to explore if it is possible to design a realistic mobile THz communication system without working in the THz near… ▽ More Near-field mobile terahertz (THz) communications is one of the candidate enablers for high-rate wireless data exchange in sixth-generation (6G) networks. However, operating in the THz near field brings both attractive opportunities and severe challenges. Hence, it becomes of interest to explore if it is possible to design a realistic mobile THz communication system without working in the THz near field. To answer this question, a mathematical framework is presented modeling a mobile THz link that works exclusively in the far field. The study leads to an interesting theoretical conclusion: while the actual frequency is of (almost) no interest, such a system must operate over a limited bandwidth not exceeding a certain threshold. It is then numerically shown that operating only in the far field imposes stringent limitations on mobile THz communications, thus making them less attractive to prospective high-rate services. In contrast, it is shown that a stationary THz link can still be broadband even when staying exclusively in the THz far field. Hence, broadband mobile THz communications MUST be near-field, while broadband stationary THz links do not have to. △ Less

Submitted 8 August, 2023; originally announced August 2023.

Comments: Accepted to IEEE Globecom, 2023. Copyright 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material, creating new works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

Generative Sequential Recommendation with GPTRec

Authors: Aleksandr V. Petrov, Craig Macdonald

Abstract: Sequential recommendation is an important recommendation task that aims to predict the next item in a sequence. Recently, adaptations of language models, particularly Transformer-based models such as SASRec and BERT4Rec, have achieved state-of-the-art results in sequential recommendation. In these models, item ids replace tokens in the original language models. However, this approach has limitatio… ▽ More Sequential recommendation is an important recommendation task that aims to predict the next item in a sequence. Recently, adaptations of language models, particularly Transformer-based models such as SASRec and BERT4Rec, have achieved state-of-the-art results in sequential recommendation. In these models, item ids replace tokens in the original language models. However, this approach has limitations. First, the vocabulary of item ids may be many times larger than in language models. Second, the classical Top-K recommendation approach used by these models may not be optimal for complex recommendation objectives, including auxiliary objectives such as diversity, coverage or coherence. Recent progress in generative language models inspires us to revisit generative approaches to address these challenges. This paper presents the GPTRec sequential recommendation model, which is based on the GPT-2 architecture. GPTRec can address large vocabulary issues by splitting item ids into sub-id tokens using a novel SVD Tokenisation algorithm based on quantised item embeddings from an SVD decomposition of the user-item interaction matrix. The paper also presents a novel Next-K recommendation strategy, which generates recommendations item-by-item, considering already recommended items. The Next-K strategy can be used for producing complex interdependent recommendation lists. We experiment with GPTRec on the MovieLens-1M dataset and show that using sub-item tokenisation GPTRec can match the quality of SASRec while reducing the embedding table by 40%. We also show that the recommendations generated by GPTRec on MovieLens-1M using the Next-K recommendation strategy match the quality of SASRec in terms of NDCG@10, meaning that the model can serve as a strong starting point for future research. △ Less

Submitted 19 June, 2023; originally announced June 2023.

Comments: Accepted at Gen-IR@SIGIR2023 workshop

Cross-link Interference Modeling in 6G Millimeter Wave and Terahertz LEO Satellite Communications

Authors: Sergi Aliaga, Vitaly Petrov, Josep M. Jornet

Abstract: One of the important questions when discussing next-generation near-Earth mmWave and terahertz (THz) band satellite communications as an integral part of the 5G-Advanced and 6G landscape is the potential interference-related issues when deploying such systems. While the space-to-ground and ground-to-space interference has been explored in multiple works already, the interference at mmWave and THz… ▽ More One of the important questions when discussing next-generation near-Earth mmWave and terahertz (THz) band satellite communications as an integral part of the 5G-Advanced and 6G landscape is the potential interference-related issues when deploying such systems. While the space-to-ground and ground-to-space interference has been explored in multiple works already, the interference at mmWave and THz cross-links, the links between the satellites themselves, have not been extensively studied yet. However, severe cross-link interference may both challenge the reliability of the data exchange within the constellation, as well as compromise the efficient co-existence of multiple satellite constellations (i.e., by different providers) covering the same or neighboring areas. In this paper, both relevant mathematical models and extensive simulation studies are presented for cross-link mmWave and THz satellite communications. Our results indicate that the cross-link interference in the considered setups is a non-negligible factor that must be further explored and accounted for in the design and deployment of next-general mmWave and THz satellite communication systems. △ Less

Submitted 21 April, 2023; v1 submitted 8 February, 2023; originally announced February 2023.

Comments: 7 pages, 7 figures, accepted to IEEE International Conference on Communications, 2023. The copyright may be transferred without further notice after which this version may not be longer available

Terahertz Communications Can Work in Rain and Snow: Impact of Adverse Weather Conditions on Channels at 140 GHz

Authors: Priyangshu Sen, Jacob Hall, Michele Polese, Vitaly Petrov, Duschia Bodet, Francesco Restuccia, Tommaso Melodia, Josep M. Jornet

Abstract: Next-generation wireless networks will leverage the spectrum above 100 GHz to enable ultra-high data rate communications over multi-GHz-wide bandwidths. The propagation environment at such high frequencies, however, introduces challenges throughout the whole protocol stack design, from physical layer signal processing to application design. Therefore, it is fundamental to develop a holistic unders… ▽ More Next-generation wireless networks will leverage the spectrum above 100 GHz to enable ultra-high data rate communications over multi-GHz-wide bandwidths. The propagation environment at such high frequencies, however, introduces challenges throughout the whole protocol stack design, from physical layer signal processing to application design. Therefore, it is fundamental to develop a holistic understanding of the channel propagation and fading characteristics over realistic deployment scenarios and ultra-wide bands. In this paper, we conduct an extensive measurement campaign to evaluate the impact of weather conditions on a wireless link in the 130-150 GHz band through a channel sounding campaign with clear weather, rain, and snow in a typical urban backhaul scenario. We present a novel channel sounder design that captures signals with -82 dBm sensitivity and 20 GHz of bandwidth. We analyze link budget, capacity, as well as channel parameters such as the delay spread and the K-factor. Our experimental results indicate that in the considered context the adverse weather does not interrupt the link, but introduces some additional constraints (e.g., high delay spread and increase in path loss in snow conditions) that need to be accounted for in the design of reliable Sixth Generation (6G) communication links above 100 GHz. △ Less

Submitted 29 August, 2022; originally announced August 2022.

Comments: P. Sen, J. Hall, M. Polese, V. Petrov, D. Bodet, F. Restuccia, T. Melodia, J. M. Jornet. 2022. Terahertz Communications Can Work in Rain and Snow: Impact of Adverse Weather Conditions on Channels at 140 GHz. In 6th ACM Workshop on Millimeter-Wave and Terahertz Networks and Sensing Systems (mmNets'22), October 17, 2022, Sydney, NSW, Australia. ACM, New York, NY, USA, 6 pages

Standardization of Extended Reality (XR) over 5G and 5G-Advanced 3GPP New Radio

Authors: Margarita Gapeyenko, Vitaly Petrov, Stefano Paris, Andrea Marcano, Klaus I. Pedersen

Abstract: Extended Reality (XR) is one of the major innovations to be introduced in 5G/5G-Advanced communication systems. A combination of augmented reality, virtual reality, and mixed reality, supplemented by cloud gaming, revisits the way how humans interact with computers, networks, and each other. However, efficient support of XR services imposes new challenges for existing and future wireless networks.… ▽ More Extended Reality (XR) is one of the major innovations to be introduced in 5G/5G-Advanced communication systems. A combination of augmented reality, virtual reality, and mixed reality, supplemented by cloud gaming, revisits the way how humans interact with computers, networks, and each other. However, efficient support of XR services imposes new challenges for existing and future wireless networks. This article presents a tutorial on integrating support for the XR into the 3GPP New Radio (NR), summarizing a range of activities handled within various 3GPP Service and Systems Aspects (SA) and Radio Access Networks (RAN) groups. The article also delivers a case study evaluating the performance of different XR services in state-of-the-art NR Release 17. The paper concludes with a vision of further enhancements to better support XR in future NR releases and outlines open problems in this area. △ Less

Submitted 28 May, 2023; v1 submitted 4 March, 2022; originally announced March 2022.

Comments: Accepted to IEEE Network, 2023. Copyright 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material, creating new works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

IEEE 802.15.3d: First Standardization Efforts for Sub-Terahertz Band Communications towards 6G

Authors: Vitaly Petrov, Thomas Kürner, and Iwao Hosako

Abstract: With the ratification of the IEEE 802.15.3d amendment to the 802.15.3, a first step has been made to standardize consumer wireless communications in the sub-THz frequency band. The IEEE 802.15.3d offers switched point-to-point connectivity with the data rates of 100\,Gbit/s and higher at distances ranging from tens of centimeters up to a few hundred meters. In this article, we provide a detailed i… ▽ More With the ratification of the IEEE 802.15.3d amendment to the 802.15.3, a first step has been made to standardize consumer wireless communications in the sub-THz frequency band. The IEEE 802.15.3d offers switched point-to-point connectivity with the data rates of 100\,Gbit/s and higher at distances ranging from tens of centimeters up to a few hundred meters. In this article, we provide a detailed introduction to the IEEE 802.15.3d and the key design principles beyond the developed standard. We particularly describe the target applications and usage scenarios, as well as the specifics of the IEEE 802.15.3d physical and medium access layers. Later, we present the results of the initial performance evaluation of IEEE 802.15.3d wireless communications. The obtained first-order performance predictions show non-incremental benefits compared to the characteristics of the fifth-generation wireless systems, thus paving the way towards the six-generation (6G) THz networks. We conclude the article by outlining the further standardization and regulatory activities on wireless networking in the THz frequency band. △ Less

Submitted 13 December, 2020; v1 submitted 3 November, 2020; originally announced November 2020.

Comments: Accepted to IEEE Communications Magazine, November 2020. ©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material, creating new works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

Hover or Perch: Comparing Capacity of Airborne and Landed Millimeter-Wave UAV Cells

Authors: Vitaly Petrov, Margarita Gapeyenko, Dmitri Moltchanov, Sergey Andreev, Robert W. Heath Jr

Abstract: On-demand deployments of millimeter-wave (mmWave) access points (APs) carried by unmanned aerial vehicles (UAVs) are considered today as a potential solution to enhance the performance of 5G+ networks. The battery lifetime of modern UAVs, though, limits the flight times in such systems. In this letter, we evaluate a feasible deployment alternative for temporary capacity boost in the areas with hig… ▽ More On-demand deployments of millimeter-wave (mmWave) access points (APs) carried by unmanned aerial vehicles (UAVs) are considered today as a potential solution to enhance the performance of 5G+ networks. The battery lifetime of modern UAVs, though, limits the flight times in such systems. In this letter, we evaluate a feasible deployment alternative for temporary capacity boost in the areas with highly fluctuating user demands. The approach is to land UAV-based mmWave APs on the nearby buildings instead of hovering over the area. Within the developed mathematical framework, we compare the system-level performance of airborne and landed deployments by taking into account the full operation cycle of the employed drones. Our numerical results demonstrate that the choice of the UAV deployment option is determined by an interplay of the separation distance between the service area and the UAV charging station, drone battery lifetime, and the number of aerial APs in use. The presented methodology and results can support efficient on-demand deployments of UAV-based mmWave APs in prospective 5G+ networks. △ Less

Submitted 26 July, 2020; originally announced July 2020.

Comments: Accepted to IEEE Wireless Communications Letters on July 20, 2020. Copyright may be transferred without further notice after which this version may become non-available

Capacity and Outage of Terahertz Communications with User Micro-mobility and Beam Misalignment

Authors: Vitaly Petrov, Dmitri Moltchanov, Yevgeni Koucheryavy, Josep Miquel Jornet

Abstract: User equipment mobility is one of the primary challenges for the design of reliable and efficient wireless links over millimeter-wave and terahertz bands. These high-rate communication systems use directional antennas and therefore have to constantly maintain alignment between transmitter and receiver beams. For terahertz links, envisioned to employ radiation patterns of no more than few degrees w… ▽ More User equipment mobility is one of the primary challenges for the design of reliable and efficient wireless links over millimeter-wave and terahertz bands. These high-rate communication systems use directional antennas and therefore have to constantly maintain alignment between transmitter and receiver beams. For terahertz links, envisioned to employ radiation patterns of no more than few degrees wide, not only the macro-scale user mobility (human walking, car driving, etc.) but also the micro-scale mobility - spontaneous shakes and rotations of the device - becomes a severe issue. In this paper, we propose a mathematical framework for the first-order analysis of the effects caused by micro-mobility on the capacity and outage in terahertz communications. The performance of terahertz communications is compared with and without micro-mobility illustrating the difference of up to 1 Tbit/s or 75%. In response to this gap, it is finally shown how the negative effects of the micro-mobility can be partially addressed by a proper adjustment of the terahertz antenna arrays and the period of beam realignment procedure. △ Less

Submitted 10 April, 2020; originally announced April 2020.

Comments: Accepted to IEEE Transactions on Vehicular Technology on April 9, 2020. Copyright may be transferred without further notice after which this version may become non-available

Analysis of Intelligent Vehicular Relaying in Urban 5G+ Millimeter-Wave Cellular Deployments

Authors: Vitaly Petrov, Dmitri Moltchanov, Sergey Andreev, Robert W. Heath Jr

Abstract: The capability of smarter networked devices to dynamically select appropriate radio connectivity options is especially important in the emerging millimeter-wave (mmWave) systems to mitigate abrupt link blockage in complex environments. To enrich the levels of diversity, mobile mmWave relays can be employed for improved connection reliability. These are considered by 3GPP for on-demand densificatio… ▽ More The capability of smarter networked devices to dynamically select appropriate radio connectivity options is especially important in the emerging millimeter-wave (mmWave) systems to mitigate abrupt link blockage in complex environments. To enrich the levels of diversity, mobile mmWave relays can be employed for improved connection reliability. These are considered by 3GPP for on-demand densification on top of the static mmWave infrastructure. However, performance dynamics of mobile mmWave relaying is not nearly well explored, especially in realistic conditions, such as urban vehicular scenarios. In this paper, we develop a mathematical framework for the performance evaluation of mmWave vehicular relaying in a typical street deployment. We analyze and compare alternative connectivity strategies by quantifying the performance gains made available to smart devices in the presence of mmWave relays. We identify situations where the use of mmWave vehicular relaying is particularly beneficial. Our methodology and results can support further standardization and deployment of mmWave relaying in more intelligent 5G+ "all-mmWave" cellular networks. △ Less

Submitted 16 August, 2019; originally announced August 2019.

Comments: 6 pages, 8 figures. The paper has been accepted for IEEE GLOBECOM 2019

On Unified Vehicular Communications and Radar Sensing in Millimeter-Wave and Low Terahertz Bands

Authors: Vitaly Petrov, Gabor Fodor, Joonas Kokkoniemi, Dmitri Moltchanov, Janne Lehtomaki, Sergey Andreev, Yevgeni Koucheryavy, Markku Juntti, Mikko Valkama

Abstract: Future smart vehicles will incorporate high-data-rate communications and high-resolution radar sensing capabilities operating in the millimeter-wave and higher frequencies. These two systems are preparing to share and reuse a lot of common functionalities, such as steerable millimeter-wave antenna arrays. Motivated by this growing overlap, and advanced further by the space and cost constraints, th… ▽ More Future smart vehicles will incorporate high-data-rate communications and high-resolution radar sensing capabilities operating in the millimeter-wave and higher frequencies. These two systems are preparing to share and reuse a lot of common functionalities, such as steerable millimeter-wave antenna arrays. Motivated by this growing overlap, and advanced further by the space and cost constraints, the vehicular community is pursuing a vision of unified vehicular communications and radar sensing, which represents a major paradigm shift for next-generation connected and self-driving cars. This article outlines a path to materialize this decisive transformation. We begin by reviewing the latest developments in hybrid vehicular communications and radar systems, and then propose a concept of unified channel access over millimeter-wave and higher frequencies. Our supporting system-level performance characterization relies upon real-life measurements and massive ray-based modeling to confirm the significant improvements brought by our proposal to mitigating the interference and deafness effects. Since our results aim to open the door to unified vehicular communications and radar sensing, we conclude by outlining the potential research directions in this rapidly developing field. △ Less

Submitted 21 January, 2019; originally announced January 2019.

Comments: 8 pages, 5 figures, 1 table. The work has been accepted for publication in IEEE Wireless Communications, 2019

Challenges of Multi-Factor Authentication for Securing Advanced IoT (A-IoT) Applications

Authors: Aleksandr Ometov, Vitaly Petrov, Sergey Bezzateev, Sergey Andreev, Yevgeni Koucheryavy, Mario Gerla

Abstract: The unprecedented proliferation of smart devices together with novel communication, computing, and control technologies have paved the way for the Advanced Internet of Things~(A-IoT). This development involves new categories of capable devices, such as high-end wearables, smart vehicles, and consumer drones aiming to enable efficient and collaborative utilization within the Smart City paradigm. Wh… ▽ More The unprecedented proliferation of smart devices together with novel communication, computing, and control technologies have paved the way for the Advanced Internet of Things~(A-IoT). This development involves new categories of capable devices, such as high-end wearables, smart vehicles, and consumer drones aiming to enable efficient and collaborative utilization within the Smart City paradigm. While massive deployments of these objects may enrich people's lives, unauthorized access to the said equipment is potentially dangerous. Hence, highly-secure human authentication mechanisms have to be designed. At the same time, human beings desire comfortable interaction with their owned devices on a daily basis, thus demanding the authentication procedures to be seamless and user-friendly, mindful of the contemporary urban dynamics. In response to these unique challenges, this work advocates for the adoption of multi-factor authentication for A-IoT, such that multiple heterogeneous methods - both well-established and emerging - are combined intelligently to grant or deny access reliably. We thus discuss the pros and cons of various solutions as well as introduce tools to combine the authentication factors, with an emphasis on challenging Smart City environments. We finally outline the open questions to shape future research efforts in this emerging field. △ Less

Submitted 21 January, 2019; originally announced January 2019.

Comments: 7 pages, 4 figures, 2 tables. The work has been accepted for publication in IEEE Network, 2019

Dense Moving Fog for Intelligent IoT: Key Challenges and Opportunities

Authors: Sergey Andreev, Vitaly Petrov, Kaibin Huang, Maria A. Lema, Mischa Dohler

Abstract: As the ratification of 5G New Radio technology is being completed, enabling network architectures are expected to undertake a matching effort. Conventional cloud and edge computing paradigms may thus become insufficient in supporting the increasingly stringent operating requirements of \emph{intelligent~Internet-of-Things (IoT) devices} that can move unpredictably and at high speeds. Complementing… ▽ More As the ratification of 5G New Radio technology is being completed, enabling network architectures are expected to undertake a matching effort. Conventional cloud and edge computing paradigms may thus become insufficient in supporting the increasingly stringent operating requirements of \emph{intelligent~Internet-of-Things (IoT) devices} that can move unpredictably and at high speeds. Complementing these, the concept of fog emerges to deploy cooperative cloud-like functions in the immediate vicinity of various moving devices, such as connected and autonomous vehicles, on the road and in the air. Envisioning gradual evolution of these infrastructures toward the increasingly denser geographical distribution of fog functionality, we in this work put forward the vision of dense moving fog for intelligent IoT applications. To this aim, we review the recent powerful enablers, outline the main challenges and opportunities, and corroborate the performance benefits of collaborative dense fog operation in a characteristic use case featuring a connected fleet of autonomous vehicles. △ Less

Submitted 20 December, 2018; originally announced December 2018.

Comments: 7 pages, 5 figures, 1 table. The work has been accepted for publication in IEEE Communications Magazine, 2019

Breaking the Limits in Urban Video Monitoring: Massive Crowd Sourced Surveillance over Vehicles

Authors: Vitaly Petrov, Sergey Andreev, Mario Gerla, Yevgeni Koucheryavy

Abstract: Contemporary urban environments are in prompt need of means for intelligent decision-making, where a crucial role belongs to smart video surveillance systems. While existing deployments of stationary monitoring cameras already deliver notable societal benefits, the proposed concept of massive video surveillance over connected vehicles that we contribute in this paper may further augment these impo… ▽ More Contemporary urban environments are in prompt need of means for intelligent decision-making, where a crucial role belongs to smart video surveillance systems. While existing deployments of stationary monitoring cameras already deliver notable societal benefits, the proposed concept of massive video surveillance over connected vehicles that we contribute in this paper may further augment these important capabilities. We therefore introduce the envisioned system concept, discuss its implementation, outline the high-level architecture, and identify major data flows, while also offering insights into the corresponding design and deployment aspects. Our conducted case study confirms the potential of the described crowd sourced vehicular system to effectively complement and eventually surpass even the best of today's static video surveillance setups. We expect that our proposal will become of value and integrate seamlessly into the future Internet-of-Things landscape, thus enabling a plethora of advanced urban applications. △ Less

Submitted 24 June, 2018; originally announced June 2018.

Comments: 8 pages, 5 figures, accepted to IEEE Wireless Communications, 2019

Last Meter Indoor Terahertz Wireless Access: Performance Insights and Implementation Roadmap

Authors: Vitaly Petrov, Joonas Kokkoniemi, Dmitri Moltchanov, Janne Lehtomaki, Yevgeni Koucheryavy, Markku Juntti

Abstract: The terahertz (THz) band, 0.1-10 THz, has sufficient resources not only to satisfy the 5G requirements of 10 Gbit/s peak data rate but to enable a number of tempting rate-greedy applications. However, the THz band brings novel challenges, never addressed at lower frequencies. Among others, the scattering of THz waves from any object, including walls and furniture, and ultra-wideband highly-directi… ▽ More The terahertz (THz) band, 0.1-10 THz, has sufficient resources not only to satisfy the 5G requirements of 10 Gbit/s peak data rate but to enable a number of tempting rate-greedy applications. However, the THz band brings novel challenges, never addressed at lower frequencies. Among others, the scattering of THz waves from any object, including walls and furniture, and ultra-wideband highly-directional links lead to fundamentally new propagation and interference structures. In this article, we review the recent progress in THz propagation modeling, antenna and testbed designs, and propose a step-by-step roadmap for wireless THz Ethernet extension for indoor environments. As a side effect, the described concept provides a second life to the currently underutilized Ethernet infrastructure by using it as a universally available backbone. By applying real THz band propagation, reflection, and scattering measurements as well as ray-tracing simulations of a typical office, we analyze two representative scenarios at 300 GHz and 1.25 THz frequencies illustrating that extremely high rates can be achieved with realistic system parameters at room scales. △ Less

Submitted 26 March, 2018; v1 submitted 9 August, 2017; originally announced August 2017.

Comments: Accepted to IEEE Communications Magazine, 2018. 8 pages, 4 figures, 1 table

Future of Ultra-Dense Networks Beyond 5G: Harnessing Heterogeneous Moving Cells

Authors: Sergey Andreev, Vitaly Petrov, Mischa Dohler, Halim Yanikomeroglu

Abstract: For the past 40 years, cellular industry has been relying on static radio access deployments with gross over-provisioning. However, to meet the exponentially growing volumes of irregular data, the very notion of a cell will have to be rethought to allow them be (re-)configured on-demand and in automated manner. This work puts forward a vision of moving networks to match dynamic user demand with ne… ▽ More For the past 40 years, cellular industry has been relying on static radio access deployments with gross over-provisioning. However, to meet the exponentially growing volumes of irregular data, the very notion of a cell will have to be rethought to allow them be (re-)configured on-demand and in automated manner. This work puts forward a vision of moving networks to match dynamic user demand with network access supply in the beyond-5G cellular systems. The resulting adaptive and flexible network infrastructures will leverage intelligent capable devices (e.g., cars and drones) by employing appropriate user involvement schemes. This work is a recollection of our efforts in this space with the goal to contribute a comprehensive research agenda. Particular attention is paid to quantifying the network performance scaling and session continuity gains with ultra-dense moving cells. Our findings argue for non-incremental benefits of integrating moving access points on a par with conventional (static) cellular access infrastructure. △ Less

Submitted 26 April, 2019; v1 submitted 16 June, 2017; originally announced June 2017.

Comments: 7 pages, 5 figures, accepted to IEEE Communications Magazine, 2019. The copyright may be transferred without further notice after which this version may not be longer available

When IoT Keeps People in the Loop: A Path Towards a New Global Utility

Authors: Vitaly Petrov, Konstantin Mikhaylov, Dmitri Moltchanov, Sergey Andreev, Gabor Fodor, Johan Torsner, Halim Yanikomeroglu, Markku Juntti, Yevgeni Koucheryavy

Abstract: While the Internet of Things (IoT) has made significant progress along the lines of supporting individual machine-type applications, it is only recently that the importance of people as an integral component of the overall IoT infrastructure has started to be fully recognized. Several powerful concepts have emerged to facilitate this vision, whether involving the human context whenever required or… ▽ More While the Internet of Things (IoT) has made significant progress along the lines of supporting individual machine-type applications, it is only recently that the importance of people as an integral component of the overall IoT infrastructure has started to be fully recognized. Several powerful concepts have emerged to facilitate this vision, whether involving the human context whenever required or directly impacting user behavior and decisions. As these become the stepping stones to develop the IoT into a novel people-centric utility, this paper outlines a path to materialize this decisive transformation. We begin by reviewing the latest progress in human-aware wireless networking, then classify the attractive human-machine applications and summarize the enabling IoT radio technologies. We continue with a unique system-level performance characterization of a representative urban IoT scenario and quantify the benefits of keeping people in the loop on various levels. Our comprehensive numerical results confirm the significant gains that have been made available with tighter user involvement, and also corroborate the development of efficient incentivization mechanisms, thereby opening the door to future commoditization of the global people-centric IoT utility. △ Less

Submitted 18 September, 2018; v1 submitted 1 March, 2017; originally announced March 2017.

Comments: 8 pages, 5 figures. This work has been accepted to IEEE Communications Magazine, 2019

Journal ref: IEEE Communications Magazine ( Volume: 57 , Issue: 1 , January 2019 )

Readout Optical System of Sapphire Disks intended for Long-Term Data Storage

Authors: V. V. Petrov, V. P. Semynozhenko, V. M. Puzikov, A. A. Kryuchyn, A. S. Lapchuk, S. M. Shanoilo, I. V. Kosyak, Yu. O. Borodin, I. V. Gorbov, Ye. M. Morozov

Abstract: The development of long-term data storage technology is one of the urging problems of our time. This paper presents the results of implementation of technical solution for long-term data storage technology proposed a few years ago on the basis of single crystal sapphire. It is shown that the problem of reading data through a substrate of negative single crystal sapphire can be solved by using for… ▽ More The development of long-term data storage technology is one of the urging problems of our time. This paper presents the results of implementation of technical solution for long-term data storage technology proposed a few years ago on the basis of single crystal sapphire. It is shown that the problem of reading data through a substrate of negative single crystal sapphire can be solved by using for reading a special optical system with a plate of positive single crystal quartz. The experimental results confirm the efficiency of the proposed method of compensation. △ Less

Submitted 3 December, 2014; v1 submitted 12 March, 2014; originally announced March 2014.

Comments: 10 pages, 3 figures, 2 tables, ArXiv.org:1310.2961v1

A Probabilistic Model of Machine Translation

Authors: G. E. Miram, V. K. Petrov

Abstract: A probabilistic model for computer-based generation of a machine translation system on the basis of English-Russian parallel text corpora is suggested. The model is trained using parallel text corpora with pre-aligned source and target sentences. The training of the model results in a bilingual dictionary of words and "word blocks" with relevant translation probability. A probabilistic model for computer-based generation of a machine translation system on the basis of English-Russian parallel text corpora is suggested. The model is trained using parallel text corpora with pre-aligned source and target sentences. The training of the model results in a bilingual dictionary of words and "word blocks" with relevant translation probability. △ Less

Submitted 10 May, 2004; originally announced May 2004.

ACM Class: I.2.7