arXiv subjects

Sheetal Kalyani

Publications and source records attributed to Sheetal Kalyani.

At least 37 records · Page 2Linked to original sources

arXiv2024

Textless NLP -- Zero Resource Challenge with Low Resource Compute

This work addresses the persistent challenges of substantial training time and GPU resource requirements even when training lightweight encoder-vocoder models for Textless NLP. We reduce training steps significantly while improving performance by a) leveraging learning rate schedulers for efficient and faster convergence b) optimizing hop length and c) tuning the interpolation scale factors for better audio quality. Additionally, we explore the latent space representation for Indian languages such as Tamil and Bengali for the acoustic unit discovery and voice conversion task. Our approach leverages a quantized encoder architecture, in conjunction with a vocoder which utilizes the proposed mixture of optimized hop length, tuned interpolation scale factors and a cyclic learning rate scheduler. We obtain consistently good results across English, Tamil and Bengali datasets. The proposed method excels in capturing complex linguistic patterns, resulting in clear reconstructed audio during voice conversion with significantly reduced training time.

cs.CL↗

arXiv2024

Practical Radar Sensing Using Two Stage Neural Network for Denoising OTFS Signals

Our objective is to derive the range and velocity of multiple targets from the delay-Doppler domain for radar sensing using orthogonal time frequency space (OTFS) signaling. Noise contamination affects the performance of OTFS signals in real-world environments, making radar sensing challenging. This work introduces a two-stage approach to tackle this issue. In the first stage, we use a generative adversarial network to denoise the corrupted OTFS samples, significantly improving the data quality. Following this, the denoised signals are passed to a convolutional neural network model to predict the values of the velocities and ranges of multiple targets. The proposed two-stage approach can predict the range and velocity of multiple targets, even in very low signal-to-noise ratio scenarios, with high accuracy and outperforms existing methods.

cs.IT↗

arXiv2024

Golden Ratio Search: A Low-Power Adversarial Attack for Deep Learning based Modulation Classification

We propose a minimal power white box adversarial attack for Deep Learning based Automatic Modulation Classification (AMC). The proposed attack uses the Golden Ratio Search (GRS) method to find powerful attacks with minimal power. We evaluate the efficacy of the proposed method by comparing it with existing adversarial attack approaches. Additionally, we test the robustness of the proposed attack against various state-of-the-art architectures, including defense mechanisms such as adversarial training, binarization, and ensemble methods. Experimental results demonstrate that the proposed attack is powerful, requires minimal power, and can be generated in less time, significantly challenging the resilience of current AMC methods.

cs.CR↗

arXiv2024

Learning Rate Optimization for Deep Neural Networks Using Lipschitz Bandits

Learning rate is a crucial parameter in training of neural networks. A properly tuned learning rate leads to faster training and higher test accuracy. In this paper, we propose a Lipschitz bandit-driven approach for tuning the learning rate of neural networks. The proposed approach is compared with the popular HyperOpt technique used extensively for hyperparameter optimization and the recently developed bandit-based algorithm BLiE. The results for multiple neural network architectures indicate that our method finds a better learning rate using a) fewer evaluations and b) lesser number of epochs per evaluation, when compared to both HyperOpt and BLiE. Thus, the proposed approach enables more efficient training of neural networks, leading to lower training time and lesser computational cost.

cs.LG↗

arXiv2024

First line of defense: A robust first layer mitigates adversarial attacks

Adversarial training (AT) incurs significant computational overhead, leading to growing interest in designing inherently robust architectures. We demonstrate that a carefully designed first layer of the neural network can serve as an implicit adversarial noise filter (ANF). This filter is created using a combination of large kernel size, increased convolution filters, and a maxpool operation. We show that integrating this filter as the first layer in architectures such as ResNet, VGG, and EfficientNet results in adversarially robust networks. Our approach achieves higher adversarial accuracies than existing natively robust architectures without AT and is competitive with adversarial-trained architectures across a wide range of datasets. Supporting our findings, we show that (a) the decision regions for our method have better margins, (b) the visualized loss surfaces are smoother, (c) the modified peak signal-to-noise ratio (mPSNR) values at the output of the ANF are higher, (d) high-frequency components are more attenuated, and (e) architectures incorporating ANF exhibit better denoising in Gaussian noise compared to baseline architectures. Code for all our experiments are available at \url{https://github.com/janani-suresh-97/first-line-defence.git}.

cs.LG↗

arXiv2024

Product and Ratio of Two $α-κ-μ$ Shadowed Random Variables and its Application to Wireless Communication

This work studies the product and ratio statistics of independent and non-identically distributed (i.n.i.d) $ α-κ- μ$ shadowed random variables. We derive the series expression for the probability density function (PDF), cumulative distribution function (CDF), and moment generating function (MGF) of the product and ratio of i.n.i.d $ α- κ- μ$ shadowed random variables. We then give the single integral representation for the derived PDF expressions. Further, as application examples, 1) outage probability has been derived for cascaded wireless systems, and 2) physical-layer security metrics like secrecy outage probability and strictly positive secrecy capacity are derived for the classic three-node model with $α-κ-μ$ shadowed fading. Next, we discuss an intelligent reflecting surface-assisted communication system over $α-κ-μ$ shadowed fading.

cs.IT↗

arXiv2024

A DRL Approach for RIS-Assisted Full-Duplex UL and DL Transmission: Beamforming, Phase Shift and Power Optimization

We propose a deep reinforcement learning (DRL) approach for a full-duplex (FD) transmission that predicts the phase shifts of the reconfigurable intelligent surface (RIS), base station (BS) active beamformers, and the transmit powers to maximize the weighted sum rate of uplink and downlink users. Existing methods require channel state information (CSI) and residual self-interference (SI) knowledge to calculate exact active beamformers or the DRL rewards, which typically fail without CSI or residual SI. Especially for time-varying channels, estimating and signaling CSI to the DRL agent is required at each time step and is costly. We propose a two-stage DRL framework with minimal signaling overhead to address this. The first stage uses the least squares method to initiate learning by partially canceling the residual SI. The second stage uses DRL to achieve performance comparable to existing CSI-based methods without requiring the CSI or the exact residual SI. Further, the proposed DRL framework for quantized RIS phase shifts reduces the signaling from BS to the RISs using $32$ times fewer bits than the continuous version. The quantized methods reduce action space, resulting in faster convergence and $7.1\%$ and $22.28\%$ better UL and DL rates, respectively than the continuous method.

cs.IT↗

arXiv2023

Multi-RIS Communication Systems: Asymptotic analysis of best RIS selection for i.n.i.d. Random Variables using Extreme Value Theory

This paper investigates the performance of multiple reconfigurable intelligent surfaces (multi-RIS) communication systems where the RIS link with the highest signal-to-noise-ratio (SNR) is selected at the destination. In practice, all the RISs will not have the same number of reflecting elements. Hence, selecting the RIS link with the highest SNR will involve characterizing the distribution of the maximum of independent, non-identically distributed (i.n.i.d.) SNR random variables (RVs). Using extreme value theory (EVT), we derive the asymptotic distribution of the normalized maximum of i.n.i.d. non-central chi-square (NCCS) distributed SNR RVs with one degree of freedom (d.o.f) and then extend the results for k-th order statistics. Using these asymptotic results, the outage capacity and average throughput expressions are derived for the multi-RIS system. The results for independent and identically distributed (i.i.d.) SNR RVs are then derived as a special case of i.n.i.d. RVs. All the derivations are validated through extensive Monte Carlo simulations, and their utility is discussed.

cs.IT↗

arXiv2023

Optimal phase shift design for fair allocation in RIS aided uplink network using statistical CSI

Reconfigurable intelligent surfaces (RIS) can be crucial in next-generation communication systems. However, designing the {RIS} phases according to the instantaneous channel state information (CSI) can be challenging in practice due to the short coherent time of the channel. In this regard, we propose a novel algorithm based on the channel statistics of massive multiple input multiple output systems rather than the instantaneous {CSI}. The beamforming at the base station (BS), power allocation of the users, and phase shifts at the RIS elements are optimized to maximize the minimum signal-to-interference and noise ratio (SINR), guaranteeing fair operation among various users. In particular, we design the RIS phases by leveraging the asymptotic deterministic equivalent of the minimum {SINR} that depends only on the channel statistics. This significantly reduces the computational complexity and the amount of controlling data between the {BS} and {RIS} for updating the phases. This setup is also useful for electromagnetic fields (EMF)-aware systems with constraints on the maximum user's exposure to EMF. The numerical results show that the proposed algorithms achieve more than $100 \%$ gain in terms of minimum SINR, compared to a system with random RIS phase shifts, when $40$ RIS elements, $20$ antennas at the BS and $10$ users, are considered.

cs.IT↗

arXiv2023

Grafting Laplace and Gaussian distributions: A new noise mechanism for differential privacy

The framework of differential privacy protects an individual's privacy while publishing query responses on congregated data. In this work, a new noise addition mechanism for differential privacy is introduced where the noise added is sampled from a hybrid density that resembles Laplace in the centre and Gaussian in the tail. With a sharper centre and light, sub-Gaussian tail, this density has the best characteristics of both distributions. We theoretically analyze the proposed mechanism, and we derive the necessary and sufficient condition in one dimension and a sufficient condition in high dimensions for the mechanism to guarantee ($ε$,$δ$)-differential privacy. Numerical simulations corroborate the efficacy of the proposed mechanism compared to other existing mechanisms in achieving a better trade-off between privacy and accuracy.

cs.CR↗

arXiv2023

Outage Probability Analysis of Uplink Cell-Free Massive MIMO Network with and without Pilot Contamination

This paper derives approximate outage probability (OP) expressions for uplink cell-free massive multiple-input-multiple-output (CF-mMIMO) systems with and without pilot contamination. The system's access points (APs) are considered to have imperfect channel state information (CSI). The signal-to-interference-plus-noise ratio (SINR) of the CF-mMIMO system is approximated via a Log-normal distribution using a two-step moment matching method. OP and ergodic rate expressions are derived with the help of the approximated Log-normal distribution. For the no-pilot contamination scenario, an exact expression is first derived using conditional expectations in terms of a multi-fold integral. Then, a novel dimension reduction method is used to approximate it by the sum of single-variable integrations. Both the approximations derived for the CF-mMIMO systems are also useful for single-cell collocated massive MIMO (mMIMO) systems and lead to closed-form expression. The derived expressions closely match the simulated numerical values for OP and ergodic rate.

cs.IT↗

arXiv2023

Binarized ResNet: Enabling Robust Automatic Modulation Classification at the resource-constrained Edge

Recently, deep neural networks (DNNs) have been used extensively for automatic modulation classification (AMC), and the results have been quite promising. However, DNNs have high memory and computation requirements making them impractical for edge networks where the devices are resource-constrained. They are also vulnerable to adversarial attacks, which is a significant security concern. This work proposes a rotated binary large ResNet (RBLResNet) for AMC that can be deployed at the edge network because of low memory and computational complexity. The performance gap between the RBLResNet and existing architectures with floating-point weights and activations can be closed by two proposed ensemble methods: (i) multilevel classification (MC), and (ii) bagging multiple RBLResNets while retaining low memory and computational power. The MC method achieves an accuracy of $93.39\%$ at $10$dB over all the $24$ modulation classes of the Deepsig dataset. This performance is comparable to state-of-the-art (SOTA) performances, with $4.75$ times lower memory and $1214$ times lower computation. Furthermore, RBLResNet also has high adversarial robustness compared to existing DNN models. The proposed MC method with RBLResNets has an adversarial accuracy of $87.25\%$ over a wide range of SNRs, surpassing the robustness of all existing SOTA methods to the best of our knowledge. Properties such as low memory, low computation, and the highest adversarial robustness make it a better choice for robust AMC in low-power edge devices.

cs.IT↗

arXiv2022

Unrolling SVT to obtain computationally efficient SVT for n-qubit quantum state tomography

Quantum state tomography aims to estimate the state of a quantum mechanical system which is described by a trace one, Hermitian positive semidefinite complex matrix, given a set of measurements of the state. Existing works focus on estimating the density matrix that represents the state, using a compressive sensing approach, with only fewer measurements than that required for a tomographically complete set, with the assumption that the true state has a low rank. One very popular method to estimate the state is the use of the Singular Value Thresholding (SVT) algorithm. In this work, we present a machine learning approach to estimate the quantum state of n-qubit systems by unrolling the iterations of SVT which we call Learned Quantum State Tomography (LQST). As merely unrolling SVT may not ensure that the output of the network meets the constraints required for a quantum state, we design and train a custom neural network whose architecture is inspired from the iterations of SVT with additional layers to meet the required constraints. We show that our proposed LQST with very few layers reconstructs the density matrix with much better fidelity than the SVT algorithm which takes many hundreds of iterations to converge. We also demonstrate the reconstruction of the quantum Bell state from an informationally incomplete set of noisy measurements.

quant-ph↗

arXiv2022

Instantaneous Channel Oblivious Phase Shift Design for an IRS-Assisted SIMO System with Quantized Phase Shift

We design the phase shifts of an intelligent reflecting surface (IRS)-assisted single-input-multiple-output communication system to minimize the outage probability (OP) and to maximize the ergodic rate. Our phase shifts design uses only statistical channel state information since these depend only on the large-scale fading coefficients; the obtained phase shift design remains valid for a longer time frame. We further assume that one has access to only quantized phase values. The closed-form expressions for OP and ergodic rate are derived for the considered system. Next, two optimization problems are formulated to choose the phase shifts of IRS such that (i) OP is minimized and (ii) the ergodic rate is maximized. We used the multi-valued particle swarm optimization (MPSO) and particle swarm optimization (PSO) algorithms to solve the optimization problems. Numerical simulations are performed to study the impact of various parameters on the OP and ergodic rate. We also discuss signaling overhead between BS and IRS controller. It is shown that the overhead can be reduced up to $99.69 \%$ by using statistical CSI for phase shift design and $5$ bits to represent the phase shifts without significantly compromising on the performance.

cs.IT↗

arXiv2022

SINR Analysis of an IRS Assisted MU-MISO System

In this work, we characterize the outage probability (OP) of an intelligent reflecting surface (IRS) assisted multi-user multiple-input-single-output (MU-MISO) communication system. Using a two-step approximation method, we approximate the signal-to-interference-plus-noise ratio (SINR) for any downlink user by a Log-Normal random variable. The impact of various system parameters is studied using the closed-form expression of OP. It is concluded that the position of IRS has a critical role, but an appropriate increase in the number of IRS elements would help to compensate for the loss in performance if the position of IRS is suboptimal.

cs.IT↗

arXiv2022

Max-Min Data Rate Optimization for RIS-aided Uplink Communications with Green Constraints

Smart radio environments aided by reconfigurable intelligent reflecting surfaces (RIS) have attracted much research attention recently. We propose a joint optimization strategy for beamforming, RIS phases, and power allocation to maximize the minimum SINR of an uplink RIS-aided communication system. The users are subject to constraints on their transmit power. We derive a closed-form expression for the beam forming vectors and a geometric programming-based solution for power allocation. We also propose two solutions for optimizing the phase shifts at the RIS, one based on the matrix lifting method and one using an approximation for the minimum function. We also propose a heuristic algorithm for optimizing quantized phase shift values. The proposed algorithms are of practical interest for systems with constraints on the maximum allowable electromagnetic field exposure. For instance, considering $24$-element RIS, $12$-antenna BS, and $6$ users, numerical results show that the proposed algorithm achieves close to $300 \%$ gain in terms of minimum SINR compared to a scheme with random RIS phases.

cs.IT↗

arXiv2022

Joint Power-control and Antenna Selection in User-Centric Cell-Free Systems with Mixed Resolution ADC

In this paper, we propose a scheme for the joint optimization of the user transmit power and the antenna selection at the access points (AP)s of a user-centric cell-free massive multiple-input-multiple-output (UC CF-mMIMO) system. We derive an approximate expression for the achievable uplink rate of the users in a UC CF-mMIMO system in the presence of a mixed analog-to-digital converter (ADC) resolution profile at the APs. Using the derived approximation, we propose to maximize the uplink sum rate of UC CF-mMIMO systems subject to energy constraints at the APs. An alternating-optimization solution is proposed using binary particle swarm optimization (BPSO) and successive convex approximation (SCA). We also study the impact of various system parameters on the performance of the system.

cs.IT↗

arXiv2022

Introducing the Huber mechanism for differentially private low-rank matrix completion

Performing low-rank matrix completion with sensitive user data calls for privacy-preserving approaches. In this work, we propose a novel noise addition mechanism for preserving differential privacy where the noise distribution is inspired by Huber loss, a well-known loss function in robust statistics. The proposed Huber mechanism is evaluated against existing differential privacy mechanisms while solving the matrix completion problem using the Alternating Least Squares approach. We also propose using the Iteratively Re-Weighted Least Squares algorithm to complete low-rank matrices and study the performance of different noise mechanisms in both synthetic and real datasets. We prove that the proposed mechanism achieves ε-differential privacy similar to the Laplace mechanism. Furthermore, empirical results indicate that the Huber mechanism outperforms Laplacian and Gaussian in some cases and is comparable, otherwise.

cs.CR↗