The fast adoption of Massive MIMO for high-throughput communications was enabled by many research contributions mostly relying on infinite-blocklength information-theoretic bounds. This makes it hard to assess the suitability ultra-reliable low-latency (URLLC) operating with short-blocklength codes. paper provides a rigorous framework characterization and numerical evaluation (using saddlepoint approximation) error probability achievable in uplink downlink at finite blocklength. encompasses...

To account for the massive uncoordinated random access scenario, which is relevant Internet of Things, Polyanskiy et al. (2017) proposed a novel formulation multiple-access problem, commonly referred to as unsourced multiple access, where all users employ common codebook and receiver decodes up permutation messages. In this paper, we extend seminal work case number active unknown <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a priori</i> ....

We present a general framework for the characterization of packet error probability achievable in cell-free Massive multiple-input multiple output (MIMO) architectures deployed to support ultra-reliable low-latency (URLLC) traffic. The is and encompasses both centralized distributed architectures, arbitrary fading channels channel estimation algorithms at network user-equipment (UE) sides, as well combining precoding schemes. used perform numerical experiments on specific scenarios, which...

Saddlepoint approximations of the meta-converse and random-coding union bounds are derived. These accurately characterize channel coding minimum error probability for symmetric memoryless channels in a wide range system parameters. The proposed simple to compute yield unified analysis both hypothesis-testing lower upper bounds.

We propose a numerically efficient method for evaluating the random-coding union bound with parameter <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$s$ </tex-math></inline-formula> on error probability achievable in finite-blocklength regime by pilot-assisted transmission scheme employing Gaussian codebooks and operating over memoryless block-fading channel. Our relies saddlepoint approximation, which,...

In recent years, the derivation of nonasymptotic converse and achievability bounds on maximum coding rate as a function error probability blocklength has gained attention in information theory literature. While these are accurate for many scenarios interest, they need to be evaluated numerically most wireless channels practical their evaluation is computationally demanding. This paper presents saddlepoint approximations state-of-the-art noncoherent, single-antenna, Rayleigh block-fading...

We propose a saddlepoint approximation of the error probability binary hypothesis test between two i.i.d. distributions. The is accurate, simple to compute, and yields unified analysis in different asymptotic regimes. proposed formulation used efficiently compute meta-converse lower bound for moderate block-lengths several cases interest.

This paper concerns the maximum coding rate at which data can be transmitted over a noncoherent, single-antenna, Rayleigh block-fading channel using an error-correcting code of given blocklength with block-error probability not exceeding value. A high-SNR normal approximation is presented that becomes accurate as signal-to-noise ratio (SNR) and number coherence intervals $L$ we tend to infinity. Numerical analyses suggest SNR values above 15dB when 10 or more.

In this paper, we adapt to cell-free Massive MIMO (multiple-input multiple-output) the finite-blocklength framework introduced by Östman et al. (2020) for characterization of packet error probability achievable with MIMO, in ultra-reliable low-latency communications (URLLC) regime. The considered paper encompasses a architecture imperfect channel-state information, and arbitrary linear signal processing performed at central-processing unit connected access points via fronthaul links. By...

We extend the seminal work by Polyanskiy (2017) on massive uncoordinated access to case where number of active users is random and unknown a priori. define random-access code accounting for both misdetection (MD) false-alarm (FA), derive random-coding achievability bound Gaussian multiple channel. Our captures fundamental trade-off between MD FA probabilities. It suggests that lack knowledge entails small penalty in power efficiency. For typical scenario, achieve probabilities below 0.1,...

We present finite-blocklength achievability bounds for the unsourced A-channel. In this multiple-access channel, users noiselessly transmit codewords picked from a common codebook with entries generated <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$q$</tex> -ary alphabet. At each channel use, receiver observes set of different transmitted symbols but not their multiplicity. show that A-channel finds applications in random-access (URA) and...

We study the single-channel source separation problem involving orthogonal frequency-division multiplexing (OFDM) signals, which are ubiquitous in many modern-day digital communication systems. Related efforts have been pursued monaural separation, where state-of-the-art neural architectures adopted to train an end-to-end separator for audio signals (as 1-dimensional time series). In this work, through a prototype based on OFDM model, we assess -- and question efficacy of using...

We study the potential of data-driven deep learning methods for separation two communication signals from an observation their mixture. In particular, we assume knowledge on generation process one signals, dubbed signal interest (SOI), and no second signal, referred to as interference. This form single-channel source problem is also interference rejection. show that capturing high-resolution temporal structures (nonstationarities), which enables accurate synchronization both SOI...

This paper presents saddlepoint approximations of state-of-the-art converse and achievability bounds for noncoherent, single-antenna, Rayleigh block-fading channels. These can be calculated efficiently are shown to accurate SNR values as small 0 dB, blocklengths 168 channel uses or more, when the channel's coherence interval is not smaller than two. It demonstrated that derived recover both normal approximation reliability function channel.

This paper presents a rigorous finite-blocklength framework for the characterization and numerical evaluation of packet error probability achievable in uplink downlink Massive MIMO ultra-reliable low-latency communications (URLLC). The encompasses imperfect channel-state information, pilot contamination, spatially correlated channels, arbitrary linear signal processing. For practical URLLC network setup involving base stations with M = 100 antennas, we show by means results that target 10...

This paper concerns the maximal achievable rate at which data can be transmitted over a non-coherent, single-antenna, Rayleigh block-fading channel using an error-correcting code of given blocklength with block-error probability not exceeding value. In particular, high-SNR normal approximation is presented that becomes accurate as signal-to-noise ratio (SNR) and number coherence intervals L we tend to infinity. Numerical analyses suggest already SNR values 15 dB.

A bilateral (i.e., upper and lower) bound on the mean-square error under a general model mismatch is developed. The bound, which derived from variational representation of chi-square divergence, applicable in Bayesian nonBayesian frameworks to biased unbiased estimators. Unlike other classical MSE bounds that depend only model, our also estimator-dependent. Thus, it as tool for characterizing specific estimator. proposed bounding technique has variety applications, one proving consistency...

In this work we demonstrate how a lack of synchronization can in fact be advantageous the problem random access. Specifically, consider multiple-access over frame-asynchronous 2-user binary-input adder channel unsourced setup (2-UBAC). Previous has shown that under perfect per-user rates achievable with linear codes 2-UBAC are limited by 0.5 bit per use (compared to capacity 0.75). paper, first arbitrary small (even single-bit) shift between user's frames enables (random) attain full 0.75...

We consider a communication problem in which the receiver must first detect presence of an information packet and, if detected, decode message carried within it. present general nonasymptotic upper and lower bounds on maximum coding rate that depend blocklength, probability false alarm, misdetection, er-ror probability. The bounds, are expressed terms binary-hypothesis-testing performance metrics, generalize finite-blocklength derived previously for scenario when genie informs whether is...

We consider the transmission of short packets over a bidirectional communication link where multiple devices, e.g., sensors and actuators, exchange small-data payloads with base station equipped large antenna array. Using results from finite-blocklength information theory, we characterize minimum SNR required to achieve target error probability for fixed packet length payload size. Our non-asymptotic analysis, which applies scenario in is device-initiated, also more challenging case when it...

We study the problem of single-channel source separation (SCSS), and focus on cyclostationary signals, which are particularly suitable in a variety application domains. Unlike classical SCSS approaches, we consider setting where only examples sources available rather than their models, inspiring data-driven approach. For models with underlying Gaussian constituents, establish lower bound attainable mean squared error (MSE) for any method, model-based or data-driven. Our analysis further...

This paper addresses the critical problem of interference rejection in radio-frequency (RF) signals using a novel, data-driven approach that leverages state-of-the-art AI models. Traditionally, algorithms are manually tailored to specific types interference. work introduces more scalable solution and contains following contributions. First, we present an insightful signal model serves as foundation for developing analyzing algorithms. Second, introduce RF Challenge, publicly available...