This paper introduces a novel deep learning-based user-side feedback reduction framework, termed self-nomination. The goal of self-nomination is to reduce the number of users (UEs) feeding back channel state information (CSI) to the base station (BS), by letting each UE decide whether to feed back based on its estimated likelihood of being scheduled and its potential contribution to precoding in a multiuser MIMO (MU-MIMO) downlink. Unlike SNR- or SINR-based thresholding methods, the proposed approach uses rich spatial channel statistics and learns nontrivial correlation effects that affect eventual MU-MIMO scheduling decisions. To train the self-nomination network under an average feedback constraint, we propose two different strategies: one based on direct optimization with gradient approximations, and another using policy gradient-based optimization with a stochastic Bernoulli policy to handle non-differentiable scheduling. The framework also supports proportional-fair scheduling by incorporating dynamic user weights. Numerical results confirm that the proposed self-nomination method significantly reduces CSI feedback overhead. Compared to baseline feedback methods, self-nomination can reduce feedback by as much as 65%, saving not only bandwidth but also allowing many UEs to avoid feedback altogether (and thus, potentially enter a sleep mode). Self-nomination achieves this significant savings with negligible reduction in sum-rate or fairness.

We study spectrum sharing between two dense low-earth orbit (LEO) satellite constellations, an incumbent primary system and a secondary system that must respect interference protection constraints on the primary system. In particular, we propose a secondary satellite selection framework and algorithm that maximizes capacity while guaranteeing that the time-average interference and absolute interference inflicted upon each primary ground user never exceeds specified thresholds. We solve this NP-hard constrained, combinatorial satellite selection problem through Lagrangian relaxation to decompose it into simpler problems which can then be solved through subgradient methods. A high-fidelity simulation is developed based on public FCC filings and technical specifications of the Starlink and Kuiper systems. We use this case study to illustrate the effectiveness of our approach and that explicit protection is indeed necessary for healthy coexistence. We further demonstrate that deep learning models can be used to predict the primary satellite system associations, which helps the secondary system avoid inflicting excessive interference and maximize its own capacity.

Deep neural network (DNN)-based algorithms are emerging as an important tool for many physical and MAC layer functions in future wireless communication systems, including for large multi-antenna channels. However, training such models typically requires a large dataset of high-dimensional channel measurements, which are very difficult and expensive to obtain. This paper introduces a novel method for generating synthetic wireless channel data using diffusion-based models to produce user-specific channels that accurately reflect real-world wireless environments. Our approach employs a conditional denoising diffusion implicit model (cDDIM) framework, effectively capturing the relationship between user location and multi-antenna channel characteristics. We generate synthetic high fidelity channel samples using user positions as conditional inputs, creating larger augmented datasets to overcome measurement scarcity. The utility of this method is demonstrated through its efficacy in training various downstream tasks such as channel compression and beam alignment. Our diffusion-based augmentation approach achieves over a 1-2 dB gain in NMSE for channel compression, and an 11dB SNR boost in beamforming compared to prior methods, such as noise addition or the use of generative adversarial networks (GANs).

In this paper, we consider a matrix function generalization of the Laplace transform of a random variable, termed the matrix Laplace transform. We characterize the conditions under which the matrix Laplace transform exists, establish its relation to the higher order moments and CCDF of a random variable, and derive the matrix Laplace transform of general Poisson shot noise. Techniques leveraging matrix Laplace transforms can provide improved tractability in the analysis of wireless networks using stochastic geometry. In particular, when one considers the underlying point process of transmitters in the network to follow a Poisson Point Process (PPP), techniques exploiting matrix Laplace transforms provide tractable expressions for the coverage probability of the network when the fading power on the desired signal follows a general phase-type distribution, the metadistribution of the SINR when the fading power on the desired signal follows an exponential distribution, and the distribution of the interference power observed by the typical user in the network.

This paper introduces a novel framework for high-accuracy outdoor user equipment (UE) positioning that applies a conditional generative diffusion model directly to high-dimensional massive MIMO channel state information (CSI). Traditional fingerprinting methods struggle to scale to large, dynamic outdoor environments and require dense, impractical data surveys. To overcome these limitations, our approach learns a direct mapping from raw uplink Sounding Reference Signal (SRS) fingerprints to continuous geographic coordinates. We demonstrate that our DiffLoc framework achieves unprecedented sub-centimeter precision, with our best model (DiffLoc-CT) delivering 0.5 cm fusion accuracy and 1-2 cm single base station (BS) accuracy in a realistic, ray-traced Tokyo urban macro-cell environment. This represents an order-of-magnitude improvement over existing methods, including supervised regression approaches (over 10 m error) and grid-based fusion (3 m error). Our consistency training approach reduces inference time from 200 steps to just 2 steps while maintaining exceptional accuracy even for high-speed users (15-25 m/s) and unseen user trajectories, demonstrating the practical feasibility of our framework for real-time 6G applications.

We consider a generalization of the Laplace transform of Poisson shot noise defined as an integral transform with respect to a matrix exponential. We denote this as the matrix Laplace transform and establish that it is in general a matrix function extension of the scalar Laplace transform. We show that the matrix Laplace transform of Poisson shot noise admits an expression analogous to that implied by Campbell’s theorem. We demonstrate the utility of this generalization of Campbell’s theorem in two important applications: the characterization of a Poisson shot noise process and the derivation of the complementary CDF (CCDF) and meta-distribution of signal-to-interference-and-noise (SINR) models in Poisson networks. In the former application, we demonstrate how the higher order moments of Poisson shot noise may be obtained directly from the elements of its matrix Laplace transform. We further show how the CCDF of this object may be bounded using a summation of the first row of its matrix Laplace transform. For the latter application, we show how the CCDF of SINR models with phase-type distributed desired signal power may be obtained via an expectation of the matrix Laplace transform of the interference and noise, analogous to the canonical case of SINR models with Rayleigh fading. Additionally, when the power of the desired signal is exponentially distributed, we establish that the meta-distribution may be obtained in terms of the limit of a sequence expressed in terms of the matrix Laplace transform of a related Poisson shot noise process.

This work provides a rigorous methodology for assessing the feasibility of spectrum sharing between large low-earth orbit (LEO) satellite constellations. For concreteness, we focus on the existing Starlink system and the soon-to-be-launched Kuiper system, which is prohibited from inflicting excessive interference onto the incumbent Starlink ground users. We carefully model and study the potential downlink interference between the two systems and investigate how strategic satellite selection may be used by Kuiper to serve its ground users while also protecting Starlink ground users. We then extend this notion of satellite selection to the case where Kuiper has limited knowledge of Starlink's serving satellite. Our findings reveal that there is always the potential for very high and extremely low interference, depending on which Starlink and Kuiper satellites are being used to serve their users. Consequently, we show that Kuiper can protect Starlink ground users with high probability, by strategically selecting which of its satellites are used to serve its ground users. Simultaneously, Kuiper is capable of delivering near-maximal downlink SINR to its own ground users. This highlights a feasible route to the coexistence of two dense LEO satellite systems, even in scenarios where one system has limited knowledge of the other's serving satellites.

We develop tractable characterizations of the interference resulting from terrestrial cellular networks radiating towards passive satellite sensing receivers. Such a setting has important implications for the future allocation and terrestrial use of spectrum in the 100 to 300 GHz band. Building on a recently developed stochastic geometry approach, we focus on the outage probability experienced by to a constellation of satellite sensors, which depends upon the distribution of the interference experienced by a typical satellite sensor. The distribution is a function of spatial and temporal randomness. We obtain upper bounds on the outage probability using a large deviation technique for Poisson shot noise, which is a novel adaptation of the Chernoff technique. This analytical method allows for the distribution of the interference to be tightly and tractably bounded. Our analysis theoretically confirms that the satellite sensor's outage probability decreases exponentially as the interference constraint is relaxed, and allows bounding of very low outage probability values, which would be very difficult to simulate.

This paper proposes and analyzes novel deep learning methods for downlink (DL) single-user multiple-input multiple-output (SU-MIMO) and multi-user MIMO (MU-MIMO) systems operating in time division duplex (TDD) mode. A motivating application is the 6G upper midbands (7-24 GHz), where the base station (BS) antenna arrays are large, user equipment (UE) array sizes are moderate, and theoretically optimal approaches are practically infeasible for several reasons. To deal with uplink (UL) pilot overhead and low signal power issues, we introduce the channel-adaptive pilot, as part of an analog channel state information feedback mechanism. Deep neural network (DNN)-generated pilots are used to linearly transform the UL channel matrix into lower-dimensional latent vectors. Meanwhile, the BS employs a second DNN that processes the received UL pilots to directly generate near-optimal DL precoders. The training is end-to-end which exploits synergies between the two DNNs. For MU-MIMO precoding, we propose a DNN structure inspired by theoretically optimum linear precoding. The proposed methods are evaluated against genie-aided upper bounds and conventional approaches, using realistic upper midband datasets. Numerical results demonstrate the potential of our approach to achieve significantly increased sum-rate, particularly at moderate to high signal-to-noise ratio (SNR) and when UL pilot overhead is constrained.

Cellular networks are becoming increasingly heterogeneous with higher base station (BS) densities and ever more frequency bands, making BS selection and band assignment key decisions in terms of user service rate and coverage. In this paper, we decompose the mobility-aware user association task into (i) forecasting of user data rate and then (ii) convex utility maximization for user association accounting for the effects of BS load and handover overheads. Using a linear combination of normalized mean-squared error (NMSE) and normalized discounted cumulative gain (NDCG) as a novel loss function, a recurrent deep neural network is trained to reliably forecast the mobile users’ future data rates. Based on the forecast, the controller optimizes the association decisions to maximize the service rate-based network utility using our computationally efficient (speed up of 100× versus generic convex solver) algorithm based on the Frank-Wolfe method. Using an industry-grade network simulator developed by Meta, we show that the proposed model predictive control (MPC) approach improves the 5th percentile service rate by 3.5× compared to the traditional signal strength-based association, reduces the median number of handovers by 7× compared to a handover agnostic strategy, and achieves service rates close to a genie-aided scheme. Furthermore, our model-based approach is significantly more sample-efficient (needs 100× less training data) compared to model-free reinforcement learning (RL), and generalizes well across different user drop scenarios.

Coexistence between 5G cellular networks and incumbent radar systems is necessary for an increasing number of spectral bands, including highly valuable spectrum such as the C-band. This paper presents a novel coexistence framework that intelligently adjusts 5G antenna parameters to mitigate interference reaching known radar systems, while simultaneously maximizing cellular network performance. The framework leverages Gaussian process regression and differential evolution to navigate high-dimensional, non-convex spaces while effectively managing uncertainty. We propose a practical approach that utilizes user RSRP measurements to characterize communication interference on radar, addressing the non-cooperative nature of radar systems. Evaluation on AT&T Labs' high-fidelity simulator demonstrates over a 12% increase in sum-log-rate and around a 3.6 dB increase in median SINR compared to the exhaustive search with common parameter configurations across all base stations, while decreasing interference on radar to its lowest achievable level in our simulation setup.

From an information theoretic perspective, joint communication and sensing (JCAS) represents a natural generalization of communication network functionality. However, it requires the re-evaluation of network performance from a multi-objective perspective. We develop a novel mathematical framework for characterizing the sensing and communication coverage probability and ergodic rate in JCAS networks. We employ a formulation of sensing parameter estimation based on mutual information to extend the notions of coverage probability and ergodic rate to the radar setting. We define sensing coverage probability as the probability that the rate of information extracted about the parameters of interest associated with a typical radar target exceeds some threshold, and sensing ergodic rate as the spatial average of the aforementioned rate of information. Using this framework, we analyze the downlink sensing and communication coverage and rate of a mmWave JCAS network employing a shared waveform, directional beamforming, and monostatic sensing. Leveraging tools from stochastic geometry, we derive upper and lower bounds for these quantities. We also develop several general technical results including: i) a generic method for obtaining closed form upper and lower bounds on the Laplace Transform of a shot noise process, ii) a new analog of Hölder’s Inequality to the setting of harmonic means, and iii) a relation between the Laplace and Mellin Transforms of a non-negative random variable. We use the derived bounds to numerically investigate the performance of JCAS networks under varying base station and blockage density. Among several insights, our numerical analysis indicates that network densification improves sensing SINR performance – in contrast to communications.

The contours of 6G – its key technical components and driving requirements – are finally coming into focus. Through twenty questions and answers, this article defines the important aspects of 6G across four categories. First, we identify the key themes and forces driving the development of 6G, and what will make 6G unique. We argue that 6G requirements and system design will be driven by (i) the tenacious pursuit of spectral (bits/Hz/area), energy (bits/Joule), and cost (bits/dollar) efficiencies, and (ii) three new service enhancements: sensing/localization/awareness, compute, and global broadband/emergency connectivity. Second, we overview the important role of spectrum in 6G, what new spectrum to expect in 6G, and outline how the different bands will be used to provide 6G services. Third, we focus our attention on the 6G physical layer, including waveforms, MIMO advancements, and the potential use of deep learning. Finally, we explore how global connectivity will be achieved in 6G, through non-terrestrial networks as well as low-cost network expansion via disaggregation and O-RAN. Although 6G standardization activities will not begin until late 2025, meaning this article is by definition speculative, our predictions are informed by several years of intensive research and discussions. Our goal is to provide a grounded perspective that will be helpful to both researchers and engineers as we move into the 6G era.

This paper presents a sensor-aided pose-aware beamwidth adaptation design for a conceptual extended reality (XR) Head-Mounted Display (HMD) equipped with a 2D planar array. The beam is tracked and adapted on the user side by leveraging HMD orientation estimates. The beamwidth adaptation scheme is effected by selective deactivation of elements in the 2D antenna array, employing the angular estimation covariance matrix to overlap the beam with the estimation confidence interval. The proposed method utilizes the estimation correlations to adapt the beamwidth along the confidence interval of these estimates. Compared to a beamwidth adaptation without leveraging estimation correlations, the proposed method demonstrates the gain of leveraging estimation correlations by improving the coverage area for a given outage probability threshold by approximately 16 %, or equivalently increasing the power efficiency up to 18 %.

Beam alignment (BA) in modern millimeter wave standards, such as 5G NR and WiGig (802.11ay), is based on exhaustive and/or hier-archical beam searches over pre-defined code-books of wide and narrow beams. This approach is slow and bandwidth/power-intensive, and is a considerable hindrance to the wide deployment of millimeter wave bands. A new approach is needed as we move toward 6G. BA is a promising use case for deep learning (DL) in the 6G air interface, offering the possibility of automated custom tuning of the BA procedure for each cell based on its unique propagation environment and user equipment (UE) location patterns. We overview and advocate for such an approach in this article, which we term site-specific beam alignment (SSBA). SSBA largely eliminates wasteful searches and allows UEs to be found much more quickly and reliably, without many of the draw-backs of other machine learning-aided approaches. We first overview and demonstrate new results on SSBA, then identify the key open challenges facing SSBA.

Wireless use cases such as spectrum sharing and Massive Machine Type Communications (mMTC) can benefit from the detection of unknown signals, which includes estimating their received power as well as other key characteristics such as bandwidth, modulation type, and waveform. While conventional signal detection methods are susceptible to noise, deep learning (DL) models offer a more robust alternative. Previously, DL models were used for solving simpler problems, focusing mainly on modulation recognition. We propose an advanced DL neural network structure that extracts the parameters of 5G NR frequency range 2 (FR2) mmWave test model waveforms. We evaluate our framework on a state-of-the-art signal generator and vector signal analyzer (VSA) that mimics real-world detection. Our work shows that incorporating curriculum training (CT) on both additive white Gaussian noise (AWGN) and frequency shift error enhances the model's accuracy across all SNR and frequency shift ranges. We further enhance the accuracy by employing the error vector magnitude (EVM) function to prioritize the top five scored parameters and validate selected parameters. As a result, our method consistently achieves an accuracy rate exceeding 90% when extracting the key parameters from 5G NR FR2 mmWave waveforms at diverse noise levels.

This work investigates the joint optimization of coverage, capacity, and cell load by tuning several cell-specific antenna and cell association parameters via data-driven methods. We are particularly focused on the complexities of macrocell and small cell coexistence, and demonstrate an automated learning method whereby macrocells and small cells can strategically adapt their coverage areas. Coupled with adaptive offloading using a tunable small cell bias, we demonstrate significant throughput and coverage improvement in a realistic 5G network simulator developed by AT&T Labs. Concretely, we formulate an optimization problem to maximize network coverage and the application-layer data rate experienced by users, accounting for delays from congestion, cell loading, and packet retransmissions. We propose an algorithm that approaches the optimum via Gaussian process models and the evolutionary search: efficiently navigating the high-dimensional, nonconvex space while managing uncertainty. Our results show that the joint optimization of antenna tuning and load balancing - exemplified by load-aware cell shaping - more than doubles the cell edge throughput and increases the cell edge SINR by 8 dB, compared to bias-only optimization. Furthermore, our algorithm and overall approach appear viable for implementation.

This work investigates the in-band coexistence between two dense low-earth orbit (LEO) satellite communication systems by analyzing two preeminent large-scale constellations, namely Starlink and Kuiper, both which have been granted non-exclusive rights to operate at 20 GHz. Through extensive simulation of Starlink and Kuiper based on their public filings, we examine downlink performance of both systems when Kuiper is obliged to protect Starlink by not inflicting prohibitive interference onto its ground users. We show that Kuiper is capable of reliably satisfying a strict protection constraint at virtually all times by strategically selecting which overhead satellites are used to serve its ground users. In fact, while protecting Starlink users in this way, our results show that Kuiper can remarkably also deliver near-maximal downlink SINR to its own ground users, revealing a feasible route to fruitful coexistence of both systems. For instance, as the constellations orbit the globe, we show that Kuiper is always capable of keeping its inflicted interference at least 12 dB below noise and in doing so sacrifices only about 1 dB in SINR over 80% of the time.

For millimeter wave (mmWave) communication, fast and accurate beam alignment is essential but challenging. Site-specific beam adaptation using deep learning is a very promising paradigm for beam alignment, but such methods typically require a lot of clean channel measurements for training, which can be difficult or even impossible to achieve in practice. This paper introduces a novel method to learn beam alignment policies using only uplink (UL) pilot measurements. The proposed method integrates a generative adversarial network (GAN)-based channel estimation (CE) model with an unsupervised deep learning model beam alignment engine (BAE). We introduce an efficient form of dataset amplification for improved training that leverages the randomness of the deep generative model (DGM) and an early stopping mechanism. Our experiments show that the GAN-BAE method achieves a better signal-to-noise ratio (SNR) by nearly 3 dB compared to compressed sensing (CS) methods such as orthogonal matching pursuit (OMP) and EM-GM-AMP (an Approximate Message Passing algorithm), especially when there are limited pilot measurements from each mobile user.

Ultradense cell-free massive multiple-input multiple-output (CF-MMIMO) has emerged as a promising technology expected to meet the future ubiquitous connectivity requirements and ever-growing data traffic demands in sixth generation (6G). This article provides a contemporary overview of ultradense CF-MMIMO networks and addresses important unresolved questions on their future deployment. We first present a comprehensive survey of state-of-the-art research on CF-MMIMO and ultradense networks. Then, we discuss the key challenges of CF-MMIMO under ultradense scenarios such as low-complexity architecture and processing, low-complexity/scalable resource allocation, fronthaul limitation, massive access, synchronization, and channel acquisition. Finally, we answer key open questions, considering different design comparisons and discussing suitable methods dealing with the key challenges of ultradense CF-MMIMO. The discussion aims to provide a valuable roadmap for interesting future research directions in this area, facilitating the development of CF-MMIMO for 6G.