SearcharxivSearch

arXiv subjects

Sergei K. Turitsyn

Publications and source records attributed to Sergei K. Turitsyn.

At least 19 recordsLinked to original sources

PatchDenoiser: Parameter-efficient multi-scale patch learning and fusion denoiser for Low-dose CT imaging

Low-dose CT images are essential for reducing radiation exposure in cancer screening, pediatric imaging, and longitudinal monitoring protocols, but their quality is often degraded by noise from low-dose acquisition, patient motion, or scanner limitations, affecting both clinical interpretation and downstream analysis. Traditional filtering approaches often over-smooth and lose fine anatomical details, while deep learning methods, including CNNs, GANs, and transformers, may struggle to preserve such details or require large, computationally expensive models, limiting clinical practicality. We propose PatchDenoiser, a lightweight, energy-efficient multi-scale patch-based denoising framework. It decomposes denoising into local texture extraction and global context aggregation, fused via a spatially aware patch fusion strategy. This design enables effective noise suppression while preserving fine structural and anatomical details. PatchDenoiser is ultra-lightweight, with far fewer parameters and lower computational complexity than CNN, GAN, and transformer based denoisers. On the 2016 Mayo Low-Dose CT dataset, PatchDenoiser consistently outperforms state-of-the-art CNN- and GAN-based methods in PSNR and SSIM. It is robust to variations in slice thickness, reconstruction kernels, and HU windows, generalizes across scanners without fine-tuning, and reduces parameters by ~9x and energy consumption per inference by ~27x compared with conventional CNN denoisers. PatchDenoiser thus provides a practical, scalable, and computationally efficient solution for medical image denoising, balancing performance, robustness, and clinical deployability.

cs.CV

Hardware-Oriented Inference Complexity of Kolmogorov-Arnold Networks

Kolmogorov-Arnold Networks (KANs) have recently emerged as a powerful architecture for various machine learning applications. However, their unique structure raises significant concerns regarding their computational overhead. Existing studies primarily evaluate KAN complexity in terms of Floating-Point Operations (FLOPs) required for GPU-based training and inference. However, in many latency-sensitive and power-constrained deployment scenarios, such as neural network-driven non-linearity mitigation in optical communications or channel state estimation in wireless communications, training is performed offline and dedicated hardware accelerators are preferred over GPUs for inference. Recent hardware implementation studies report KAN complexity using platform-specific resource consumption metrics, such as Look-Up Tables, Flip-Flops, and Block RAMs. However, these metrics require a full hardware design and synthesis stage that limits their utility for early-stage architectural decisions and cross-platform comparisons. To address this, we derive generalized, platform-independent formulae for evaluating the hardware inference complexity of KANs in terms of Real Multiplications (RM), Bit Operations (BOP), and Number of Additions and Bit-Shifts (NABS). We extend our analysis across multiple KAN variants, including B-spline, Gaussian Radial Basis Function (GRBF), Chebyshev, and Fourier KANs. The proposed metrics can be computed directly from the network structure and enable a fair and straightforward inference complexity comparison between KAN and other neural network architectures.

cs.LG

Small-scale photonic Kolmogorov-Arnold networks using standard telecom nonlinear modules

Photonic neural networks promise ultrafast inference, yet most architectures rely on linear optical meshes with electronic nonlinearities, reintroducing optical-electrical-optical bottlenecks. Here we introduce small-scale photonic Kolmogorov-Arnold networks (SSP-KANs) implemented entirely with standard telecommunications components. Each network edge employs a trainable nonlinear module composed of a Mach-Zehnder interferometer, semiconductor optical amplifier, and variable optical attenuators, providing a four-parameter transfer function derived from gain saturation and interferometric mixing. Despite the constrained functional form of these optical nonlinearities, SSP-KANs comprising only a few optical modules achieve strong nonlinear inference performance across classification, regression, and image recognition tasks, approaching software baselines with significantly fewer parameters. A four-module network achieves $94.3$\% (IQR: $90.3$--$97.4$\%, 10~seeds) accuracy on nonlinear classification benchmarks; a seven-module network attains $R^2 = 0.986 \pm 0.015$ on six-input regression. Performance remains robust under realistic hardware impairments, maintaining high accuracy down to 6-bit input resolution and 14 dB signal-to-noise ratio. By using a fully differentiable physics model for end-to-end optimisation of optical parameters, this work establishes a practical pathway from simulation to experimental demonstration of photonic KANs using commodity telecom hardware.

physics.optics

423.7 + 426.5 Tb/s GMI Bi-Directional HCF Transmission

We demonstrate OESCL-band same-wavelength bi-directional transmission over 60 km HCF with 42.5 THz bandwidth, achieving GMIs comparable with the highest unidirectional SMF data-rates in both directions, with an aggregate of 423.7 + 426.5 Tb/s.

eess.SP

Unlocking the O-Band: high-power, broadband soliton microcomb

The O-band (1260-1360 nm), located near the minimum of chromatic dispersion of standard single-mode fiber, is the transmission window of major interest and importance for short-reach data-center interconnects. However, full capacity offered by this spectral band is yet to be unlocked, due to limited availability of scalable multi-wavelength, high-power, low noise O-band light engines. While Kerr microcombs in CMOS-compatible silicon nitride resonators provide mutually coherent wavelength channels with precise spacing and chip-scale footprints, their practical deployment in the O-band has been hindered by limited pump laser power, insufficient per-line power and the lack of flat, wideband amplification technologies to uniformly boost multiple coherent carriers. Here we demonstrate a high-power O-band soliton microcomb architecture that overcomes this bottleneck by combining self-injection-locked (SIL) operation in a Silicon Nitride microring with a single-stage bismuth-doped phosphosilicate fiber amplifier designed for wideband, flat-top gain. The SIL microcomb operates with an 834 GHz free spectral range and spans over 1050-1650 nm. The amplifier simultaneously boosts 21 O-band lines across 100 nm to powers exceeding 0 dBm per carrier without gain flattening or external equalization, while preserving low-noise characteristics. We validate each amplified microcomb line as a carrier across the entire O-band using dual-polarization 32 GBaud 64-QAM coherent transmission. This approach establishes a practical route towards high-power, broadband O-band microcomb engines for next-generation data-center interconnects and scalable photonic systems.

physics.optics

From Data to Decision: A Multi-Stage Framework for Class Imbalance Mitigation in Optical Network Failure Analysis

Machine learning-based failure management in optical networks has gained significant attention in recent years. However, severe class imbalance, where normal instances vastly outnumber failure cases, remains a considerable challenge. While pre- and in-processing techniques have been widely studied, post-processing methods are largely unexplored. In this work, we present a direct comparison of pre-, in-, and post-processing approaches for class imbalance mitigation in failure detection and identification using an experimental dataset. For failure detection, post-processing methods-particularly Threshold Adjustment-achieve the highest F1 score improvement (up to 15.3%), while Random Under-Sampling provides the fastest inference. In failure identification, GenAI methods deliver the most substantial performance gains (up to 24.2%), whereas post-processing shows limited impact in multi-class settings. When class overlap is present and latency is critical, over-sampling methods such as the SMOTE are most effective; without latency constraints, Meta-Learning yields the best results. In low-overlap scenarios, Generative AI approaches provide the highest performance with minimal inference time.

cs.LG

Dual Path Learning -- learning from noise and context for medical image denoising

Medical imaging plays a critical role in modern healthcare, enabling clinicians to accurately diagnose diseases and develop effective treatment plans. However, noise, often introduced by imaging devices, can degrade image quality, leading to misinterpretation and compromised clinical outcomes. Existing denoising approaches typically rely either on noise characteristics or on contextual information from the image. Moreover, they are commonly developed and evaluated for a single imaging modality and noise type. Motivated by Geng et.al CNCL, which integrates both noise and context, this study introduces a Dual-Pathway Learning (DPL) model architecture that effectively denoises medical images by leveraging both sources of information and fusing them to generate the final output. DPL is evaluated across multiple imaging modalities and various types of noise, demonstrating its robustness and generalizability. DPL improves PSNR by 3.35% compared to the baseline UNet when evaluated on Gaussian noise and trained across all modalities. The code is available at 10.5281/zenodo.15836053.

eess.IV

Efficient and Robust Semantic Image Communication via Stable Cascade

Diffusion Model (DM) based Semantic Image Communication (SIC) systems face significant challenges, such as slow inference speed and generation randomness, that limit their reliability and practicality. To overcome these issues, we propose a novel SIC framework inspired by Stable Cascade, where extremely compact latent image embeddings are used as conditioning to the diffusion process. Our approach drastically reduces the data transmission overhead, compressing the transmitted embedding to just 0.29% of the original image size. It outperforms three benchmark approaches - the diffusion SIC model conditioned on segmentation maps (GESCO), the recent Stable Diffusion (SD)-based SIC framework (Img2Img-SC), and the conventional JPEG2000 + LDPC coding - by achieving superior reconstruction quality under noisy channel conditions, as validated across multiple metrics. Notably, it also delivers significant computational efficiency, enabling over 3x faster reconstruction for 512 x 512 images and more than 16x faster for 1024 x 1024 images as compared to the approach adopted in Img2Img-SC.

eess.IV

Optical neuromorphic computing via temporal up-sampling and trainable encoding on a telecom device platform

Mapping input signals to a high-dimensional space is a critical concept in various neuromorphic computing paradigms, including models such as Reservoir Computing (RC) and Extreme Learning Machines (ELM). We propose using commercially available telecom devices and technologies developed for high-speed optical data transmission to implement these models through nonlinear mapping of optical signals into a high-dimensional space where linear processing can be applied. We manipulate the output feature dimension by applying temporal up-sampling (at the speed of commercially available telecom devices) of input signals and a well-established wave-division-multiplexing (WDM). Our up-sampling approach utilizes a trainable encoding mask, where each input symbol is replaced with a structured sequence of masked symbols, effectively increasing the representational capacity of the feature space. This gives remarkable flexibility in the dynamical phase masking of the input signal. We demonstrate this approach in the context of RC and ELM, employing readily available photonic devices, including a semiconductor optical amplifier and nonlinear Mach-Zender interferometer (MZI). We investigate how nonlinear mapping provided by these devices can be characterized in terms of the increased controlled separability and predictability of the output state.

physics.optics

FPGA Implementation of Complex Value-based Clustering Filter for Chromatic Dispersion Compensation in Coherent Metro Links with Ultra-low Power Consumption

This paper introduces a new machine learning-assisted chromatic dispersion compensation filter, demonstrating its superior power efficiency compared to conventional FFT-based filters for metro link distances. Validations on FPGA confirmed an energy efficiency gain of up to 63.5\% compared to the standard frequency-domain chromatic dispersion equalizer.

eess.SP

Geometric Clustering for Hardware-Efficient Implementation of Chromatic Dispersion Compensation

Power efficiency remains a significant challenge in modern optical fiber communication systems, driving efforts to reduce the computational complexity of digital signal processing, particularly in chromatic dispersion compensation (CDC) algorithms. While various strategies for complexity reduction have been proposed, many lack the necessary hardware implementation to validate their benefits. This paper provides a theoretical analysis of the tap overlapping effect in CDC filters for coherent receivers, introduces a novel Time-Domain Clustered Equalizer (TDCE) technique based on this concept, and presents a Field-Programmable Gate Array (FPGA) implementation for validation. We developed an innovative parallelization method for TDCE, implementing it in hardware for fiber lengths up to 640 km. A fair comparison with the state-of-the-art frequency domain equalizer (FDE) under identical conditions is also conducted. Our findings highlight that implementation strategies, including parallelization and memory management, are as crucial as computational complexity in determining hardware complexity and energy efficiency. The proposed TDCE hardware implementation achieves up to 70.7\% energy savings and 71.4\% multiplier usage savings compared to FDE, despite its higher computational complexity.

eess.SP