arXiv subjects

Priyalakshmi Sheela

Publications and source records attributed to Priyalakshmi Sheela.

2 recordsLinked to original sources

arXiv2026

Advances in Wavelet Denoising for Communication Signals: From Parameter Selection Toward Data-Driven Optimization

Wavelet denoising suppresses nonstationary, impulsive, and interference-like disturbances in communication signals, but its effectiveness depends on jointly selecting the transform family, mother wavelet, decomposition level, thresholding rule, and shrinkage function. This review synthesises studies published during 2020--2025 across ten sources using a PRISMA-aligned protocol and classifies them by parameter-selection focus and application domain. The evidence shows a shift from fixed empirical choices toward similarity-, sparsity-, entropy-, energy-, sub-band-SNR-, and task-loss-driven selection, while revealing limited communication-specific validation. To address this gap, DWT, SWT, and WPT are benchmarked for OFDM denoising under impulsive noise using SNR gain, MSE, BER, EVM, real-time feasibility, Friedman and Wilcoxon tests, efficiency-index ranking, and embedded DSP/FPGA constraints. Results show that improved waveform fidelity does not necessarily translate into better hard-decision performance, motivating receiver-level validation. A Unified Decision Framework is therefore developed and validated on synthetic pilot-aided OFDM channel estimation and measured IEEE 802.11n channels using BER, EVM, NMSE, and SNR gain. The selected configuration significantly outperforms fixed-parameter wavelet and classical baselines $\left(p < 10^{-11}\right)$, achieves the lowest estimation error, generalises to held-out data, adapts to channel conditions, and supports extension to deep-unfolding architectures.

eess.SP↗

arXiv2026

Drift-Aware RL-based Wavelet Denoising for Network-Traffic Anomaly Detection

Traffic-utilisation measurements for network monitoring are corrupted by additive noise and statistical drift: time-dependent change in the signal's mean, variance, distributional shape, or tail behaviour. Static wavelet denoising, calibrated under stationary independent and identically distributed (i.i.d.) Gaussian assumptions, becomes mismatched under drift and, at moderate-to-high signal-to-noise ratio (SNR), over-suppresses useful structure and degrades monitoring decisions. We propose a drift-aware framework treating adaptive wavelet denoising as a preprocessing layer optimised for two tasks: anomaly detection, recovering the multi-scale transient load bursts that noise and drift obscure, and capacity estimation, recovering the operational required capacity $C_{95}$ (95th percentile of utilisation). Because localised bursts are multi-scale structure a wavelet preserves but a low-pass filter removes, detection discriminates denoiser families. A four-detector gate (Page-Hinkley, variance-ratio, Jensen-Shannon, Anderson-Darling) determines when to invoke a learned policy, and a Proximal Policy Optimization agent selects a per-window wavelet configuration over a mixed discrete-continuous action space. Unlike prior work, the reward is downstream task utility, not reconstruction fidelity. The denoiser is benchmarked, per drift type and input SNR, against a low-pass moving-average filter, VisuShrink, SureShrink, BayesShrink, and a Wiener filter. Defining the anomaly target on the clean signal and the drift gate on the corruption keeps both stages non-circular.

eess.SP↗