arXiv subjects

Alexandre Graell i Amat

Publications and source records attributed to Alexandre Graell i Amat.

At least 55 records · Page 3Linked to original sources

arXiv2023

Rateless Autoencoder Codes: Trading off Decoding Delay and Reliability

Most of today's communication systems are designed to target reliable message recovery after receiving the entire encoded message (codeword). However, in many practical scenarios, the transmission process may be interrupted before receiving the complete codeword. This paper proposes a novel rateless autoencoder (AE)-based code design suitable for decoding the transmitted message before the noisy codeword is fully received. Using particular dropout strategies applied during the training process, rateless AE codes allow to trade off between decoding delay and reliability, providing a graceful improvement of the latter with each additionally received codeword symbol. The proposed rateless AEs significantly outperform the conventional AE designs for scenarios where it is desirable to trade off reliability for lower decoding delay.

cs.IT↗

arXiv2022

Improving the Thresholds of Generalized LDPC Codes with Convolutional Code Constraints

CC-GLPDC codes are a class of generalized low-density parity-check (GLDPC) codes where the constraint nodes (CNs) represent convolutional codes. This allows for efficient decoding in the trellis with the forward-backward algorithm, and the strength of the component codes easily can be controlled by the encoder memory without changing the graph structure. In this letter, we extend the class of CC-GLDPC codes by introducing different types of irregularity at the CNs and investigating their effect on the BP and MAP decoding thresholds for the binary erasure channel (BEC). For the considered class of codes, an exhaustive grid search is performed to find the BP-optimized and MAP-optimized ensembles and compare their thresholds with the regular ensemble of the same design rate. The results show that irregularity can significantly improve the BP thresholds, whereas the thresholds of the MAP-optimized ensembles are only slightly different from the regular ensembles. Simulation results for the AWGN channel are presented as well and compared to the corresponding thresholds.

cs.IT↗

arXiv2022

Concatenated Codes for Multiple Reads of a DNA Sequence

Decoding sequences that stem from multiple transmissions of a codeword over an insertion, deletion, and substitution channel is a critical component of efficient deoxyribonucleic acid (DNA) data storage systems. In this paper, we consider a concatenated coding scheme with an outer nonbinary low-density parity-check code or a polar code and either an inner convolutional code or a time-varying block code. We propose two novel decoding algorithms for inference from multiple received sequences, both combining the inner code and channel to a joint hidden Markov model to infer symbolwise a posteriori probabilities (APPs). The first decoder computes the exact APPs by jointly decoding the received sequences, whereas the second decoder approximates the APPs by combining the results of separately decoded received sequences and has a complexity that is linear with the number of sequences. Using the proposed algorithms, we evaluate the performance of decoding multiple received sequences by means of achievable information rates and Monte-Carlo simulations. We show significant performance gains compared to a single received sequence. In addition, we succeed in improving the performance of the aforementioned coding scheme by optimizing both the inner and outer codes.

cs.IT↗

arXiv2022

CodedPaddedFL and CodedSecAgg: Straggler Mitigation and Secure Aggregation in Federated Learning

We present two novel federated learning (FL) schemes that mitigate the effect of straggling devices by introducing redundancy on the devices' data across the network. Compared to other schemes in the literature, which deal with stragglers or device dropouts by ignoring their contribution, the proposed schemes do not suffer from the client drift problem. The first scheme, CodedPaddedFL, mitigates the effect of stragglers while retaining the privacy level of conventional FL. It combines one-time padding for user data privacy with gradient codes to yield straggler resiliency. The second scheme, CodedSecAgg, provides straggler resiliency and robustness against model inversion attacks and is based on Shamir's secret sharing. We apply CodedPaddedFL and CodedSecAgg to a classification problem. For a scenario with 120 devices, CodedPaddedFL achieves a speed-up factor of 18 for an accuracy of 95% on the MNIST dataset compared to conventional FL. Furthermore, it yields similar performance in terms of latency compared to a recently proposed scheme by Prakash et al. without the shortcoming of additional leakage of private data. CodedSecAgg outperforms the state-of-the-art secure aggregation scheme LightSecAgg by a speed-up factor of 6.6-18.7 for the MNIST dataset for an accuracy of 95%.

cs.LG↗

arXiv2022

Frequency-domain digital predistortion for Massive MU-MIMO-OFDM Downlink

Digital predistortion (DPD) is a method commonly used to compensate for the nonlinear effects of power amplifiers (PAs). However, the computational complexity of most DPD algorithms becomes an issue in the downlink of massive multi-user (MU) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM), where potentially up to several hundreds of PAs in the base station (BS) require linearization. In this paper, we propose a convolutional neural network (CNN)-based DPD in the frequency domain, taking place before the precoding, where the dimensionality of the signal space depends on the number of users, instead of the number of BS antennas. Simulation results on generalized memory polynomial (GMP)-based PAs show that the proposed CNN-based DPD can lead to very large complexity savings as the number of BS antenna increases at the expense of a small increase in power to achieve the same symbol error rate (SER).

eess.SP↗

arXiv2022

Finite-Length Scaling of SC-LDPC Codes With a Limited Number of Decoding Iterations

We propose four finite-length scaling laws to predict the frame error rate (FER) performance of spatially-coupled low-density parity-check codes under full belief propagation (BP) decoding with a limit on the number of decoding iterations and a scaling law for sliding window decoding, also with limited iterations. The laws for full BP decoding provide a choice between accuracy and computational complexity; a good balance between them is achieved by the law that models the number of decoded bits after a certain number of BP iterations by a time-integrated Ornstein-Uhlenbeck process. This framework is developed further to model sliding window decoding as a race between the integrated Ornstein-Uhlenbeck process and an absorbing barrier that corresponds to the left boundary of the sliding window. The proposed scaling laws yield accurate FER predictions.

cs.IT↗

arXiv2022

Computational Code-Based Privacy in Coded Federated Learning

We propose a privacy-preserving federated learning (FL) scheme that is resilient against straggling devices. An adaptive scenario is suggested where the slower devices share their data with the faster ones and do not participate in the learning process. The proposed scheme employs code-based cryptography to ensure \emph{computational} privacy of the private data, i.e., no device with bounded computational power can obtain information about the other devices' data in feasible time. For a scenario with 25 devices, the proposed scheme achieves a speed-up of 4.7 and 4 for 92 and 128 bits security, respectively, for an accuracy of 95\% on the MNIST dataset compared with conventional mini-batch FL.

cs.IT↗

arXiv2022

Generalized Spatially-Coupled Parallel Concatenated Codes With Partial Repetition

A new class of spatially-coupled turbo-like codes (SC-TCs), dubbed generalized spatially coupled parallel concatenated codes (GSC-PCCs), is introduced. These codes are constructed by applying spatial coupling on parallel concatenated codes (PCCs) with a fraction of information bits repeated $q$ times. GSC-PCCs can be seen as a generalization of the original spatially-coupled parallel concatenated codes proposed by Moloudi et al. [2]. To characterize the asymptotic performance of GSC-PCCs, we derive the corresponding density evolution equations and compute their decoding thresholds. The threshold saturation effect is observed and proven. Most importantly, we rigorously prove that any rate-$R$ GSC-PCC ensemble with 2-state convolutional component codes achieves at least a fraction $1-\frac{R}{R+q}$ of the capacity of the binary erasure channel (BEC) for repetition factor $q\geq2$ and this multiplicative gap vanishes as $q$ tends to infinity. To the best of our knowledge, this is the first class of SC-TCs that are proven to be capacity-achieving. Further, the connection between the strength of the component codes, the decoding thresholds of GSC-PCCs, and the repetition factor are established. The superiority of the proposed codes with finite blocklength is exemplified by comparing their error performance with that of existing SC-TCs via computer simulations.

cs.IT↗

arXiv2022

Error Floor Analysis of Irregular Repetition ALOHA

With the rapid expansion of the Internet of Things, the efficient sharing of the wireless medium by a large amount of simple transmitters is becoming essential. Scheduling-based solutions are inefficient for this setting, where small data units are broadcast sporadically by terminals that most of the time are idle. Modern random access has embraced the challenge and provides suitable slot-synchronous and asynchronous multiple access solutions based on replicating the packets and exploiting successive interference cancellation (SIC) at the receiver. In this work, we focus on asynchronous modern random access. Specifically, we derive an analytical approximation of the performance of irregular repetition ALOHA (IRA) in the so-called error floor region. Numerical results show the tightness of the derived approximation under various scenarios.

cs.IT↗

arXiv2022

Coding for Straggler Mitigation in Federated Learning

We present a novel coded federated learning (FL) scheme for linear regression that mitigates the effect of straggling devices while retaining the privacy level of conventional FL. The proposed scheme combines one-time padding to preserve privacy and gradient codes to yield resiliency against stragglers and consists of two phases. In the first phase, the devices share a one-time padded version of their local data with a subset of other devices. In the second phase, the devices and the central server collaboratively and iteratively train a global linear model using gradient codes on the one-time padded local data. To apply one-time padding to real data, our scheme exploits a fixed-point arithmetic representation of the data. Unlike the coded FL scheme recently introduced by Prakash \emph{et al.}, the proposed scheme maintains the same level of privacy as conventional FL while achieving a similar training time. Compared to conventional FL, we show that the proposed scheme achieves a training speed-up factor of $6.6$ and $9.2$ on the MNIST and Fashion-MNIST datasets for an accuracy of $95\%$ and $85\%$, respectively.

cs.LG↗

arXiv2022

Privacy-Preserving Coded Mobile Edge Computing for Low-Latency Distributed Inference

We consider a mobile edge computing scenario where a number of devices want to perform a linear inference $\boldsymbol{W}\boldsymbol{x}$ on some local data $\boldsymbol{x}$ given a network-side matrix $\boldsymbol{W}$. The computation is performed at the network edge over a number of edge servers. We propose a coding scheme that provides information-theoretic privacy against $z$ colluding (honest-but-curious) edge servers, while minimizing the overall latency\textemdash comprising upload, computation, download, and decoding latency\textemdash in the presence of straggling servers. The proposed scheme exploits Shamir's secret sharing to yield data privacy and straggler mitigation, combined with replication to provide spatial diversity for the download. We also propose two variants of the scheme that further reduce latency. For a considered scenario with $9$ edge servers, the proposed scheme reduces the latency by $8\%$ compared to the nonprivate scheme recently introduced by Zhang and Simeone, while providing privacy against an honest-but-curious edge server.

cs.IT↗

arXiv2022

Massive Uncoordinated Access With Random User Activity

We extend the seminal work by Polyanskiy (2017) on massive uncoordinated access to the case where the number of active users is random and unknown a priori. We define a random-access code accounting for both misdetection (MD) and false alarm (FA), and derive a random-coding achievability bound for the Gaussian multiple-access channel. Our bound captures the fundamental trade-off between MD and FA probabilities. The derived bound suggests that, for the scenario considered in Polyanskiy (2017), lack of knowledge of the number of active users entails a small penalty in terms of power efficiency. For example, our bound shows that 0.5-0.7 dB extra power is required to achieve both MD and FA probabilities below 0.1 compared to the case in which the number of active users is known a priori. Taking both MD and FA into account, we show that some of the recently proposed massive random access schemes are highly suboptimal with respect to our bound.

cs.IT↗

arXiv2021

Age of Information in Prioritized Random Access

Age of information (AoI) is a performance metric that captures the freshness of status updates. While AoI has been studied thoroughly for point-to-point links, the impact of modern random-access protocols on this metric is still unclear. In this paper, we extend the recent results by Munari to prioritized random access where devices are divided into different classes according to different AoI requirements. We consider the irregular repetition slotted ALOHA protocol and analyze the AoI evolution by means of a Markovian analysis following similar lines as in Munari (2021). We aim to design the protocol to satisfy the AoI requirements for each class while minimizing the power consumption. To this end, we optimize the update probability and the degree distributions of each class, such that the probability that their AoI exceeds a given threshold lies below a given target and the average number of transmitted packets is minimized.

cs.IT↗

arXiv2021

Model-Based End-to-End Learning for WDM Systems With Transceiver Hardware Impairments

We propose an AE-based transceiver for a WDM system impaired by hardware imperfections. We design our AE following the architecture of conventional communication systems. This enables to initialize the AE-based transceiver to have similar performance to its conventional counterpart prior to training and improves the training convergence rate. We first train the AE in a single-channel system, and show that it achieves performance improvements by putting energy outside the desired bandwidth, and therefore cannot be used for a WDM system. We then train the AE in a WDM setup. Simulation results show that the proposed AE significantly outperforms the conventional approach. More specifically, it increases the spectral efficiency of the considered system by reducing the guard band by 37\% and 50\% for a root-raised-cosine filter-based matched filter with 10\% and 1\% roll-off, respectively. An ablation study indicates that the performance gain can be ascribed to the optimization of the symbol mapper, the pulse-shaping filter, and the symbol demapper. Finally, we use reinforcement learning to learn the pulse-shaping filter under the assumption that the channel model is unknown. Simulation results show that the reinforcement-learning-based algorithm achieves similar performance to the standard supervised end-to-end learning approach assuming perfect channel knowledge.

eess.SP↗

arXiv2021

DSAG: A mixed synchronous-asynchronous iterative method for straggler-resilient learning

We consider straggler-resilient learning. In many previous works, e.g., in the coded computing literature, straggling is modeled as random delays that are independent and identically distributed between workers. However, in many practical scenarios, a given worker may straggle over an extended period of time. We propose a latency model that captures this behavior and is substantiated by traces collected on Microsoft Azure, Amazon Web Services (AWS), and a small local cluster. Building on this model, we propose DSAG, a mixed synchronous-asynchronous iterative optimization method, based on the stochastic average gradient (SAG) method, that combines timely and stale results. We also propose a dynamic load-balancing strategy to further reduce the impact of straggling workers. We evaluate DSAG for principal component analysis, cast as a finite-sum optimization problem, of a large genomics dataset, and for logistic regression on a cluster composed of 100 workers on AWS, and find that DSAG is up to about 50% faster than SAG, and more than twice as fast as coded computing methods, for the particular scenario that we consider.

cs.DC↗

arXiv2021

Symbol-Based Over-the-Air Digital Predistortion Using Reinforcement Learning

We propose an over-the-air digital predistortion optimization algorithm using reinforcement learning. Based on a symbol-based criterion, the algorithm minimizes the errors between downsampled messages at the receiver side. The algorithm does not require any knowledge about the underlying hardware or channel. For a generalized memory polynomial power amplifier and additive white Gaussian noise channel, we show that the proposed algorithm achieves performance improvements in terms of symbol error rate compared with an indirect learning architecture even when the latter is coupled with a full sampling rate ADC in the feedback path. Furthermore, it maintains a satisfactory adjacent channel power ratio.

eess.SP↗

arXiv2021

Multi-Server Weakly-Private Information Retrieval

Private information retrieval (PIR) protocols ensure that a user can download a file from a database without revealing any information on the identity of the requested file to the servers storing the database. While existing protocols strictly impose that no information is leaked on the file's identity, this work initiates the study of the tradeoffs that can be achieved by relaxing the perfect privacy requirement. We refer to such protocols as weakly-private information retrieval (WPIR) protocols. In particular, for the case of multiple noncolluding replicated servers, we study how the download rate, the upload cost, and the access complexity can be improved when relaxing the full privacy constraint. To quantify the information leakage on the requested file's identity we consider mutual information (MI), worst-case information leakage, and maximal leakage (MaxL). We present two WPIR schemes, denoted by Scheme A and Scheme B, based on two recent PIR protocols and show that the download rate of the former can be optimized by solving a convex optimization problem. We also show that Scheme A achieves an improved download rate compared to the recently proposed scheme by Samy et al. under the so-called $ε$-privacy metric. Additionally, a family of schemes based on partitioning is presented. Moreover, we provide an information-theoretic converse bound for the maximum possible download rate for the MI and MaxL privacy metrics under a practical restriction on the alphabet size of queries and answers. For two servers and two files, the bound is tight under the MaxL metric, which settles the WPIR capacity in this particular case. Finally, we compare the performance of the proposed schemes and their gap to the converse bound.

cs.IT↗

arXiv2021

Privacy-Preserving Wireless Federated Learning Exploiting Inherent Hardware Impairments

We consider a wireless federated learning system where multiple data holder edge devices collaborate to train a global model via sharing their parameter updates with an honest-but-curious parameter server. We demonstrate that the inherent hardware-induced distortion perturbing the model updates of the edge devices can be exploited as a privacy-preserving mechanism. In particular, we model the distortion as power-dependent additive Gaussian noise and present a power allocation strategy that provides privacy guarantees within the framework of differential privacy. We conduct numerical experiments to evaluate the performance of the proposed power allocation scheme under different levels of hardware impairments.

cs.IT↗