Papers

This paper investigates energy efficiency (EE) maximization for the downlink of cell-free massive multiple-input multiple-output systems under quality-of-service and per-access point power constraints. We first derive closed-form gradient expressions of the objective function with respect to the power allocation coefficients, and then propose an accelerated projected gradient (APG) approach to solve this problem. To reduce the computational complexity and runtime of APG, we propose a deep-unfolded APG framework that maps iterative APG updates onto a finite number of neural network layers, where parameters such as step sizes and penalty coefficients are learned from data. The proposed approach produces power allocation solutions through a fixed number of gradient-based updates without the need for line search or manual parameter tuning. Numerical results show that the method achieves EE performance comparable to the iterative APG approach while requiring significantly lower computational cost, with up to a 30-fold reduction in floating-point operations under the considered system settings.

This paper proposes a geometry-based joint cross-modal token selection framework for latency-constrained multimodal token communications. To capture cross-modal token dependencies, we leverage the cross-attention mechanism to project modality-specific tokens into a shared query-key space, where the modality with the fewest tokens serves as the anchor modality and the others as non-anchor modalities. Inspired by germ-grain models, we define an angular-distance metric and construct semantic grain regions around anchor queries. Based on this geometric representation, we identify cross-modal evidence shared across multiple anchor queries in this space and develop an intersection-based token selection (IBS) strategy that prioritizes non-anchor tokens whose key are covered by multiple grain regions. We further develop an erasure-aware extension, termed robust-IBS (R-IBS), for token-wise erasure channels using an expected angular-distance formulation. In both IBS and R-IBS, the grain regions are optimized for individual queries under a latency constraint, using block coordinate descent and a low-complexity greedy algorithm. Simulations corroborate the effectiveness of IBS and R-IBS under latency-constrained and token-wise erasure channels, achieving up to 31.6% and 29.2% task accuracy gains on visual question answering (VQA) and audio-visual question answering (AVQA) tasks, respectively, over existing token selection baselines.

Non-linear functions such as exponential and sigmoid are essential in AI and LLM acceleration, although implementing them efficiently on FPGAs is still costly. This paper proposes a PWL interpolation framework based on partial reconfiguration to reduce hardware cost while preserving flexibility. The architecture separates the design into a static region for communication and control, and a reconfigurable region where different interpolation modules can be dynamically loaded. Uniform and non-uniform segmentation strategies are evaluated for exponential and sigmoid functions using FP16 and FP32 arithmetic. Results show that non-uniform segmentation can improve accuracy in high-curvature regions, while uniform segmentation offers lower hardware overhead. At the system level, the reconfigurable implementation achieved significant area savings, reaching up to 43% less LUTs, 50% less flip-flops, BRAMs and DSPs cells, compared against a static design containing both operators; all this with predictable reconfiguration latency. These results show that partial reconfiguration is a practical approach for exploring area-latency trade-offs in FPGA-based acceleration of non-linear functions for LLM workloads.

Pretrained large language models (LLMs) have demonstrated in-context learning (ICL) capabilities for numerical inference over sequences serialized as text. Prior work has identified and characterized this form of numerical inference primarily through output-level evaluations such as prediction error. However, how numerical information is organized within LLM representations remains much less understood. To study this internal organization, we adopt a graph signal processing perspective in which attention induces a weighted graph over tokens, while token hidden states define signals on its nodes. Quantitative graph-spectral diagnostics and qualitative token-graph visualizations reveal that representations become more clearly differentiated by input dynamical complexity as context length increases. Simpler inputs produce attention-induced token graphs with stronger global connectivity and smoother, spectrally concentrated hidden-state signals, whereas more complex inputs produce more localized graphs and hidden-state signals with broader spectral support and greater high-frequency energy. Together, these findings point to systematic, context-dependent internal signatures associated with numerical ICL that are conserved across model families.

Orthogonal frequency-division multiplexing (OFDM) is widely adopted in frequency-selective channels for its ability to simplify equalization, yet it also causes large signal peaks, out-of-band (OOB) emissions, and spectral leakage. We study downlink precoding/combining for multi-user multiple-input multiple-output (MU-MIMO)–OFDM systems by minimizing the sum of the users' mean-squared errors (MSEs) under per-subcarrier transmit-power limits, per-antenna OOB spectral-mask constraints, and per-antenna peak-amplitude (clipping) constraints, which confine the emitted spectrum and limit waveform peaks to reduce power-amplifier saturation, nonlinear distortion, and spectral regrowth. The resulting nonconvex problem is handled by a minimum mean-squared error (MMSE) based block coordinate descent (BCD), with a closed-form combiner update and an alternating direction method of multipliers (ADMM) algorithm with closed-form updates for the precoder subproblem. Simulations show clear gains over well-known benchmark schemes.

Recent zero-shot text-to-speech (TTS) systems achieve remarkable naturalness and speaker similarity but typically require high-quality speaker prompts and either strip away or entangle the acoustic environment with speaker characteristics, limiting their real-world applicability. We present an extended DAIEN-TTS, an environment-aware zero-shot TTS framework that disentangles and jointly models speech, background noise, and reverberation, enabling independent control over timbre and acoustic environment through separate speaker and environment prompts. Built upon the flow-matching-based F5-TTS, it uses a speech-environment separation module to decompose environmental speech into speech, noise, and reverberation components, which are injected into the Diffusion Transformer for environment-aware generation. Training uses simulated data constructed by mixing clean speech with noise and room impulse responses, together with a cross-speaker conditioning strategy that suppresses speaker information leakage from the environment branch. When real-world data are available, the system can be further fine-tuned to bridge the simulated-to-real domain gap.At inference, a triple classifier-free guidance mechanism enables fine-grained control over speech, noise, and reverberation, and a signal-to-noise-ratio adaptation strategy aligns the synthesized speech with the environment prompt. Experiments on simulated and real-world test sets show that DAIEN-TTS generates environmental personalized speech with high naturalness, strong speaker similarity, and faithful noise and reverberation reproduction, while offering controllability beyond prior environment-aware TTS systems.

Nonlinear activation functions are widely used in machine learning workloads, but their direct hardware implementation is often costly, function-specific, or difficult to reuse across different models. This work introduces CAMTA, a 16-bit reconfigurable multi-region activation unit for nonlinear function approximation in FPGA and ASIC accelerators. CAMTA combines independent region thresholds, per-region polynomial degrees, coefficient sets, and execution modes over a shared Horner-based datapath. Unlike conventional polynomial or piecewise approximation units that mainly reconfigure coefficients or segment selection, CAMTA also reconfigures the computational behavior of each region through HORNER, CONST, ZERO, and IDENTITY modes, enabling the same hardware to support functions with different symmetry and tail behavior without resynthesis. FPGA validation on an AMD Alveo platform shows RMSE as low as \(3.60\times10^{-6}\) for CAMTA-assisted Softmax, outperforming the CORDIC-based Softmax baseline considered in this work by nearly one order of magnitude. FPGA HLS synthesis reports 3 DSPs, 802 FFs, 1756 LUTs, and an 11-cycle datapath latency. ASIC synthesis in TSMC 65~nm at 250~MHz reports \(6632.40~\mu\mathrm{m}^2\) total cell area and \(1.3634~\mathrm{mW}\) total power. Compared with a same-node, function-specific PLAC implementation, CAMTA incurs \(2.20\times\) area and \(1.75\times\) power overhead, in exchange for runtime configurability and reuse across multiple nonlinear functions.

Communication-centric Integrated Sensing and Communications (ISAC) is a promising paradigm for sixth-generation (6G) wireless systems, enabling new sensing services by leveraging the already-deployed communication infrastructure. Communication signals typically comprise both known deterministic pilot sequences and unknown random data payloads. For localization and sensing tasks, the prevailing approach in multistatic and distributed ISAC systems relies exclusively on pilot symbols, entirely overlooking the positioning information carried by data payloads, which constitute the majority of each transmitted frame. Alternatively, Decision-Directed (DD) approaches treat data estimates as additional pilots, inherently limiting localization performance to that of the underlying communication system, while Non-Data-Aided (NDA) methods from the literature require prior knowledge of the data symbol distribution and incur a computational cost that grows with constellation size. In this paper, we derive a Joint Maximum Likelihood (JML) estimator that jointly exploits pilot and data symbols for localization without requiring data decoding, in a passive scenario where a distributed sensing receiver localizes a User Equipment (UE) by exploiting its Orthogonal Frequency-Division Multiplexing (OFDM) communication signal as a signal of opportunity. The optimal solution is derived and shown to be computationally intractable for typical 6G parameters. Two tractable approximations are then proposed, achieving localization performance superior to DD baselines at comparable computational complexity, while remaining constellation-agnostic and yielding substantially lower computational requirements than existing NDA approaches. Furthermore, the proposed estimators are shown to admit a geometric interpretation, providing insight into their intrinsic localization behavior.

High-altitude platform stations (HAPS) are promising components of 6G and beyond networks, where antenna array configuration is critical for achieving wide-area coverage and high capacity with massive MIMO. This paper investigates and compares the uplink signal-to-interference-plus-noise ratio (SINR) distributions of user equipments (UEs) for five antenna array structures, including the cylindrical antenna array, the 3GPP antenna array, the hemispherical antenna array, and two proposed architectures, namely the truncated cone and truncated hemispherical antenna arrays, under uniform, Gaussian, and Poisson cluster process UEs distributions. Simulation results show that both proposed arrays achieve performance comparable to the hemispherical array, with the truncated hemispherical array being particularly effective for densely distributed UEs, while the truncated cone array offers a favorable tradeoff between performance and implementation complexity.

We present a method to accurately localize multiple perturbations from fiber outputs. We cross-correlate received polarizations between two wavelengths, apply peak detection, and interpolate peaks to boost accuracy. We localize five simultaneous perturbations in a 10 GSa/s numerical fiber model with 3.5 m median error.

Recent studies have reported clearly identifiable dynamical changes in the high-frequency range of EEG signals recorded during specific stimuli, such as visual or auditory inputs, or in cases of brain disorders like epileptic seizures. In this study, we utilized Dynamic Mode Decomposition (DMD) to extract consistent and persistent dynamical changes in the high-frequency band from the signals of neurologically relevant EEG channels. High-frequency DMD modes were employed as features, composing a feature table. Through post-processing, a random distribution test was performed, revealing that approximately 70% of the samples exhibited consistent high-frequency dynamics within the signal of a specific channel. Furthermore, classification experiments confirmed that the PCA components of the feature table that passed the test formed a consistent pattern that distinguished the alcohol-dependent group from the control group.

Polar codes achieve channel capacity as block length increases, but this asymptotic advantage comes at the cost of decoding speed: the conventional successive cancellation (SC) algorithm is inherently sequential, which limits its practicality for high-throughput applications. This work addresses that limitation by introducing the Speculative Successive-Cancellation (Spec-SC) framework, which incorporates speculative execution into the SC decoding process. While node-based decoders with special nodes enable partial parallelism for specific, predefined structures, Spec-SC extends this capability by permitting parallel execution in general nodes that do not match any special-node pattern. A built-in verification mechanism ensures that this speculation does not compromise decoding correctness. Simulation results show that Spec-SC achieves identical BER and FER performance to standard SC decoding, while reducing the average number of sequentially traversed nodes by up to $69\%$ for a polar code of length $N=4096$. These results demonstrate that Spec-SC enables additional algorithm-level parallelism beyond what special-node approaches alone can achieve, paving the way for speculative hardware decoder implementations.

Apportionment converts fractional entitlements into integer seat allocations. Quota requires each state to receive the floor or ceiling of its standard quota, whereas population monotonicity prohibits a state whose population weakly increases from losing a seat to a state whose population weakly decreases. Gölz, Peters, and Procaccia recently removed the order-preservation assumptions used in classical incompatibility results by giving a five-state construction; together with the three-state Webster possibility result, this left four states as the unresolved boundary. We prove that no deterministic four-state apportionment solution satisfies both axioms under their definition, which permits the compared house sizes to differ. The proof is a finite logical gadget. Conditional on one quota choice at a central profile, twelve auxiliary profiles encode three bits and force a frustrated cycle. The argument uses only transfers between states whose populations are unchanged and assumes neither anonymity, neutrality, order preservation, homogeneity, nor other regularity conditions. It also applies to relative population monotonicity. Geometrically, the result is a global compatibility obstruction for quota-constrained lattice rounding, rather than an average-distortion bound.

Multi-level apportionment allocates integer seats through a hierarchy of groups. Schmidt-Kraepelin, Suksompong, and Wijaya proved that, at every fixed house size, lower and upper quota can be satisfied simultaneously; they also constructed house-monotone rules satisfying either quota separately. They left open whether one rule can satisfy lower quota, upper quota, and house monotonicity together, even when quota is required only relative to the root. We give a negative answer. For a full binary comb with $D$ equally entitled leaves, every house-monotone allocation sequence induces a sequence of seat recipients. Quota for the nested comb groups would force every grid-aligned prefix discrepancy to be below one. A midpoint embedding then bounds the full interval discrepancy by this quantity plus $1/2$, contradicting Schmidt's logarithmic lower bound. Conversely, a binary van der Corput seat schedule has comb-prefix error at most $\log D/(3\log 2)+1$. Thus the optimal worst-case error on the comb is $\Theta(\log D)$, and for sufficiently large finite $D$ no house-monotone quota rule exists. The proof isolates a static–dynamic gap: each house size admits a quota-feasible allocation, but the feasible allocations cannot be embedded into one monotone path. In quantization language, the result characterizes the order of the embedded-quantization penalty for progressive one-hot rounding on the comb.

Accurate gait analysis in Parkinson's disease (PD) typically relies on laboratory-based systems to capture biomechanical data, such as ground reaction forces (GRFs). Estimating GRFs using inertial measurement units (IMUs) provides a feasible alternative. However, this approach remains challenging in pathological gait like PD due to its high variability and complexity. Moreover, existing monitoring approaches often require multiple body-mounted sensors, which limit practicality and reduce patient compliance. To date, no study has investigated the application of deep learning approaches to address this challenge. This study proposes, for the first time, a deep learning framework to estimate bilateral vertical GRFs (vGRFs) in PD using an optimized set of wearable IMUs. A hybrid CNN-BiLSTM model was trained separately on data from 61 PD patients and 65 healthy controls (HC) using 13 IMUs. The model achieved high intra-subject accuracy ($R^2$ = 0.98) and strong inter-subject generalization ($R^2$ = 0.93 for HC, $R^2$ = 0.91 for PD). Sensor configuration was found to significantly influence estimation accuracy, with optimal sensor placement varying between PD patients and HC. For PD patients, estimation accuracy dropped markedly when reducing to a single IMU. The optimal configuration for PD used four IMUs. We identified a minimal setup with only two IMUs still enabled robust estimation. This compact setup offers a practical and scalable solution. Overall, the proposed approach supports the development of wearable vGRF-based gait analysis systems for Parkinsonian gait and potentially other pathological conditions, enabling accessible clinical assessments, remote monitoring, and personalized rehabilitation.

The inverse problem of precipitation retrieval from Doppler power spectral density measurements of a fast-scanning single-polarization X-band weather radar is addressed. The formulation is developed around quantities that remain interpretable and stable across short coherent processing intervals: radial-wind mean, radial-wind spectral width, reflectivity-weighted terminal fall velocity, and median volume diameter. The gamma drop-size distribution parameters are retained as latent variables within the spectral forward model rather than reported as primary retrieval products. This choice is motivated by likelihood analyses showing that the inverse problem is weakly identifiable with respect to the latent distribution parameters, whereas the derived quantities remain substantially more stable. Simulated-data experiments compare the proposed approach with a moment-based retrieval using externally supplied wind information in the short-coherent-processing-interval regime. Real-data experiments demonstrate same-scan spatial likelihood pooling through a one-scan plan-position-indicator retrieval and 33-scan time-series comparisons against two disdrometers within the radar coverage region. The results show that the latent distribution parameters should be treated as intermediate fitting variables rather than primary end products. The proposed approach shows its clearest practical benefit for median volume diameter and reflectivity-weighted terminal fall velocity under incoherent short-record operation. Spatial pooling allows efficient full-scan mapping of both quantities when each Doppler spectrum contains only a few slow-time samples. Radar-disdrometer discrepancies remain physically meaningful even when the spectral fit and temporal evolution are reasonable.

We numerically investigate gas-line absorption in hollow-core fibre C+L-band transmission. Results show limited C-band but severe L-band performance degradation. Ideal suppression enables a 1.5$\times$ higher L-band throughput, reaching 59.1 Tb/s over 1000 km, while reducing repeater number by up to 3.3$\times$.

In integrated sensing and communication (ISAC) systems, stringent sensing performance constraints can severely limit the power available for communication. Hybrid reconfigurable intelligent surfaces (HRISs) with capabilities of both passive reflection and active signal amplification can significantly improve communication performance in the power-limited regime. This motivates us to analyze and optimize the performance of an HRIS-aided multiple-input-multiple-output (mMIMO) ISAC system. We first estimate the effective uplink/downlink channels using the minimum mean square error method. We then derive closed-form expressions for the communication sum-rate and sensing Cramér-Rao lower bound (CRLB). It is shown that under the equal power allocation strategy, the CRLB remains independent of the HRIS coefficients. Then, we formulate a joint optimization problem of power allocation and HRIS beamforming to maximize the communication sum-rate while ensuring specified sensing CRLB constraints. To solve the formulated non-convex problem, we propose an alternating optimization algorithm based on fractional programming and successive convex approximation. Extensive simulations validate our analysis and proposed algorithm, showing significant improvements in both communication and sensing performances enabled by the HRIS. For example, an HRIS with only $4$ active elements offers $97.30\%$ improvement in the communication sum-rate, while ensuring a sensing CRLB constraint of $-30$ dB.

Closed-loop lower-limb exoskeleton control via Electroencephalography (EEG) remains limited by motion artifacts, low signal-to-noise ratio, and binary gait formulations that fail to capture full cortical gait complexity. We propose a 2-block Brain-Computer Interface (BCI) architecture: a trainable session-specific Feature Extraction Block with real-time artifact suppression and multi-domain feature extraction, coupled with a Decoder Block built on a novel Polynomial Time-Varying Layer (PolyTVL)+LSTM for four-state gait classification (Stand, Initiate, Execute, Terminate). Ablation confirmed v01 (PolyTVL+LSTM) outperformed all variants (validation MCC: 0.435, gap: 0.187), with consistent EEG feature discriminability across ROIs and sub-bands (p<0.05). Closed-loop deployment with v01 achieved 55.3% (Rex-assisted) and 52.7% (volitional) gait initiation success, with a mean end-to-end processing time of 70.5~ms (+/-41.5), validating real-time feasibility in this pilot study.

Faster-than-Nyquist signaling improves spectral ef- ficiency by deliberately introducing inter-symbol interference. Classical sequence detectors such as BCJR can approach optimal performance, but their computational cost grows rapidly with channel memory. This paper investigates data-driven FTN BPSK detection under AWGN through a direct comparison between multilayer perceptrons and Kolmogorov Arnold Networks. A large-scale Monte Carlo dataset containing nearly four million labeled windows is generated for a time-packing factor of zero point eight and signal-to-noise ratio values from seven to ten decibels. The best MLP obtained from width sweeping uses hidden width thirty two, whereas the selected KAN uses hidden width four with spline grid size five. At ten decibels, the MLP produces a bit error rate of one point three times ten to the minus four, while the KAN reaches seven times ten to the minus six. This corresponds to an eighteen point six times lower bit error rate while using only one eighth of the MLP hidden width. The results show that KAN provides a more effective and more parameter-efficient neural decision model than the MLP baseline for FTN BPSK detection.

This letter investigates an integrated sensing and communication (ISAC) system aided by movable subarrays (MSAs) using a hybrid near-far field channel model. The sensing target and communication users are assumed to lie in the near field of the overall MSA aperture but in the far-field region of each subarray. Accordingly, a hybrid near-far field channel model is established, and the equivalent Fisher information matrix and Cramér-Rao bound (CRB) for joint range, elevation, and azimuth estimation are derived. The transmit beamforming matrix and subarray positions are jointly optimized to minimize the trace CRB subject to minimum communication signal-to-interference-plus-noise ratio (SINR), maximum transmit power and subarray movement constraints. An alternating optimization algorithm is developed combining iterative rank-one-penalized semidefinite relaxation with projected finite-difference block descent and backtracking. Numerical results show that the hybrid-field model closely matches the spherical-wave model, while MSAs substantially reduce the CRB.

This brief presents a circuit-level conductive-textile interconnection technique for batteryless distributed wearable modules. Two conductive textile layers separated by an insulating fabric layer are used as a garment-wide shared bus that simultaneously conveys DC power and pulse-based data signals without point-to-point wiring. To minimize the circuit overhead of each module, universal asynchronous receiver/transmitter (UART) pulses are directly coupled onto the textile through AC-coupling capacitors without carrier modulation. The textile is modeled as a lumped resistance–capacitance network. The textile waveform is then modeled as the transient response of a second-order circuit. From this model, design conditions for comparator-based waveform recovery and high-bitrate transmission are analytically derived. The resulting design framework determines whether a given combination of data rate, textile capacitance, and resistance is feasible, and also provides the corresponding design range of the decoupling inductors. The required textile parameters are characterized, and robustness to parameter drift is analyzed. These results establish a basic methodology for textile-based simultaneous power and data transfer.

Due to the high complexity of geometry-deterministic wireless channel modeling and the difficulty in its implementation, geometry-based stochastic channel modeling (GBSM) approaches have been used to evaluate system performance of wireless communications. This paper introduces a new method to model a GBSM by training a generative neural network using images formed by channel parameters. Toward this end, we process the data of channel parameters in the form of images and train the generative neural networks where the convolutional layers are mainly employed to capture correlation among multipath components. Through a case study, we demonstrate that the use of channel images facilitates the training of the generative model and ensures that the model learns the correlations among multipath components. We show that the outputs of the generative model faithfully represent the joint distributions of the original data, and that the trained model reliably interpolates across held-out conditions not used during training, demonstrating its practical value as a data-efficient alternative to directly resampling the ray-tracing database. Furthermore, to corroborate applicability of the trained model, we run simple system-level simulations and show the results obtained from the trained model closely match those from the ray-tracing data. Therefore, the proposed model is expected to ease the burden of GBSM implementations with general wireless conditions and capture the statistical joint distributions of the original channel data.

Reconfigurable intelligent surface (RIS) has demon- strated remarkable potential to enhance the performance of integrated sensing and communication (ISAC), particularly when the line-of-sight (LoS) paths are obstructed. By controlling the reconfigurable elements on the surface, RIS can establish virtual LoS paths and provide considerable passive beamforming gains, thereby significantly improving the received signal quality. In this paper, we design a novel multi-hop RIS ISAC system for target positioning, where multiple RISs are deployed to assist the communication from a transmitter to associated users while simultaneously enhancing receiver sensing performance in target positioning. Specifically, we formulate an optimization problem to minimize the root mean square error (RMSE) of the target detection while guaranteeing the communication requirements of the users. To solve this problem, we first unfold the cascaded sens- ing channel through parallel factor decomposition, and develop a low-rank CANDECOMP/PARAFAC decomposition (CPD)-based scheme to extract the location parameters (i.e., angle of arrival, angle of departure and delay) of the sensing targets. Then, we develop a scheme for jointly selecting the transmit beamforming and RIS phase shift configurations to maximize the sensing energy at the receiver, which in turn leads to improved accuracy in target positioning. We also provide a uniqueness analysis, complexity analysis, and Cramér-Rao lower bound (CRLB) of the parameters estimated by our methodology. Simulation results validate the improvement in target positioning obtained by our design relative to baselines.

The 3.5 GHz Citizens Broadband Radio Service (CBRS) is a shared wireless band that allows both government systems and commercial networks (such as private LTE/5G) to use the same spectrum. To prevent interference with critical government systems, especially naval radars, CBRS uses a monitoring system called the Environmental Sensing Capability (ESC). ESC acts like a network of sensors that continuously listens for radar signals and alerts the system when they are detected, so commercial users can temporarily stop or adjust their transmissions. This paper reviews how radar signals are detected within the CBRS band. We first explain the regulatory framework and describe the types of radar signals that need to be identified. We then examine traditional detection methods, such as energy-based and pattern-matching techniques, and compare them with newer approaches based on machine learning and deep learning, which can automatically learn to recognize radar signals from data. We also review publicly available datasets and testing platforms used to evaluate these detection methods, along with key performance requirements such as high detection accuracy (e.g., 99% detection probability (radar overlap recall)) and low delay (e.g., within 60 seconds). Finally, we highlight current challenges, including false alarms, interference from modern wireless systems, and the need for real-time operation. Overall, this survey shows that while traditional methods are simple and reliable in controlled settings, modern learning-based approaches offer better performance in complex environments. The future of CBRS radar detection will likely combine both approaches to achieve accurate, fast, and robust performance in real-world deployments.

Physical-layer security based on pseudo-noise (PN) superposition is a promising approach for mitigating eavesdropping in future wireless systems. However, under Shannon's capacity formulation with Gaussian signaling, achieving secrecy typically requires allocating substantial transmit power to PN, resulting in a significant reduction in achievable information rate and limiting practical applicability. This limitation is alleviated when finite-alphabet modulation schemes, such as M-ary Quadrature Amplitude Modulation ($M$-QAM), are employed, as expected in practical 6G transceivers. In this work, we analyze the information rate performance of PN-assisted systems under $M$-QAM signaling using mutual information and derive the corresponding achievable secrecy rate. The impact of PN power allocation on both the legitimate user and the eavesdropper is investigated across different modulation orders and channel conditions. Monte Carlo simulations are conducted to evaluate system behavior under varying user and eavesdropper channel conditions and to examine how PN power allocation influences secrecy performance. The results show that, at sufficiently high signal-to-noise ratio (SNR), the information rate becomes largely insensitive to PN power allocation, enabling near-perfect secrecy with $M$-QAM modulation-highlighting a key departure from Shannon-capacity-based secrecy analyses and underscoring the practicality of finite-alphabet security mechanisms for 6G wireless systems.

Joint frequency and chirp-rate estimation for a noisy chirp signal arises in radar, sonar, and burst satellite communications. Conventional estimators combine a coarse grid search with fine interpolation; accuracy degrades at the edges of the residual cell (the edge effect) and below the breakdown SNR (the threshold effect). We present a deterministic two-stage estimator that controls both failure modes uniformly over the residual cell. The estimator combines a time-centered, zero-padded dechirp-FFT acquisition bank with alternating selectable-$p$ amplitude-interpolation refinements on DTFT samples at fractional bins; in the centered frame, the frequency-chirp-rate cross-term of the Fisher information vanishes. The paper derives a mean-squared-error and threshold characterization over the full SNR range, in closed form except for one calibrated scalar (an effective cell count), to our knowledge the first for the joint problem: the breakdown threshold is governed by the cell count, and its cell-position dependence is dominated by the scalloping loss of the coarse FFT, which the padding bounds at 0.4 dB. An asymptotic uniformity analysis over the cell, including its corners, gives fixed-point variance ratios of $1.003$ and $0.998$, analytically free of the residual. A closed-form bias analysis under a cubic phase mismatch shows the centered chirp-rate estimate is insensitive to first order. Monte Carlo experiments at $N=256$ (validated at $N=32$-$512$) measure frequency- and chirp-rate-axis efficiencies with median $1.03$ and worst case $1.07$ over $144$ cell positions at $-5$ dB. Threshold predictions hold within $1.0$ dB on four configurations not used in the calibration. The dechirp-FFT bank is fully parallel, and each of the four refinement iterations evaluates three DTFT samples per axis; under fixed operating conditions, per-estimate latency is constant at $O(N\log N)$ cost.

Over the past decade, deep learning has evolved from a tool for replacing isolated wireless blocks into a broader methodology for developing wireless intelligence. This article traces that trajectory through three shifts: learning wireless functional modules, redesigning and re-normalizing communication goals, and enabling generalization under practical physical constraints. Together, these shifts advance the broader pursuit of communication anytime and anywhere, through any appropriate means. Early studies showed that neural networks could approximate difficult physical-layer inference and network-optimization mappings, while subsequent research embedded domain-specific structure, shifted toward task-oriented semantics, and addressed the need for edge-efficient adaptation. Looking ahead, we argue that the next era of wireless artificial intelligence (AI) depends on more than scaling model capacity. Promising directions include physically grounded wireless world models, agentic reasoning and fulfillment, and standardization mechanisms that allow learned components to operate with clear boundaries, physical consistency, and system-level interoperability.

Efficient Channel State Information (CSI) feedback is indispensable for frequency division duplex (FDD) massive multiple-input multiple-output (MIMO) systems. Existing compressed sensing (CS) algorithms exploit delay-domain sparsity but suffer from prohibitive iterative latency and discrete grid mismatch. Conversely, deep learning (DL) approaches achieve rapid inference but lack spatial scalability and domain adaptability, failing to generalize to unseen propagation environments, and demand computationally heavy encoders and decoder. In this paper, we propose TAP, a Tap-Assisted Parametric CSI Compression. TAP is a one-shot neural framework that unifies the speed of DL with the mathematical interpretability of CS. TAP replaces iterative pursuit with a lightweight 1D neural network that extracts dominant continuous propagation delays from temporal channel sequences via a differentiable sub-grid interpolation operator. TAP achieves true architecture independence, enabling zero-shot generalization across diverse array geometries and unseen propagation environments. Furthermore, TAP yields a completely decoder-free payload, allowing the BS to reconstruct the channel via a simple inverse fast Fourier transform (IFFT). Extensive evaluations across five 3GPP environments demonstrate that TAP achieves a 3.13 to 12.22 dB channel frequency response normalized mean square error (CFR-NMSE) improvement over CsiNet while shrinking the model footprint by 660 times to under 1 MB. Operating with sub-millisecond latencies, TAP accelerates inference by 2700 times over classical iterative OMP, providing a scalable and deployment-ready solution for next-generation networks.

In this article, we consider received-power maximization in RIS-assisted wireless links using only limited received signal strength (RSS) feedback, avoiding explicit estimation of the high-dimensional cascaded channel. We propose stochastic and deterministic algorithms for passive and active beamforming. In the deterministic method, each RIS element requires only three fixed-phase RSS measurements to recover the phase, followed by tree-structured quantization. We prove a non-asymptotic received-power guarantee for the fixed-beamformer single-user RIS sweep under cascaded Rayleigh fading and extend the bound to finite-resolution RIS phase shifters. Simulations over Rayleigh and temporally correlated channels show that the proposed methods approach full-CSI alternating optimization. We extend the channel-estimation-free framework to multi-user RIS-assisted transmission by proposing a unified single-bit consensus method for both RIS phase adaptation and transmit beamforming. In this mechanism, all users vote on each shared perturbation using only local SINR comparisons, allowing the update to account for both desired-signal improvement and interference generated to other users. The weighted consensus allows for fairness among users based on their priority.