Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SliceVision-F2I: A Synthetic Feature-to-Image Dataset for Visual Pattern Representation on Network Slices

Md. Abid Hasan Rafi, Mst. Fatematuj Johora, Pankaj Bhowmik

TL;DR

SliceVision-F2I introduces a synthetic KPI-to-image dataset for network slicing, pairing 10 KPIs with $16×16$ RGB encodings produced by four visual methods under realistic noise. It demonstrates that pattern-based CNNs using Perlin noise, neural wallpapering, fractal branching, or physically guided mappings can achieve near-perfect slice-type classification, outperforming traditional KPI-based ML even at low image resolutions. The dataset supports three slice types (eMBB, URLLC, mIoT) with deliberate class imbalance and realistic data gaps, serving as a benchmark for vision-based network analytics. The work highlights the potential of visual KPI representations to improve robustness to measurement uncertainty, enable real-time processing, and provide a flexible platform for future hybrid models and pattern-design optimization in network management.

Abstract

The emergence of 5G and 6G networks has established network slicing as a significant part of future service-oriented architectures, demanding refined identification methods supported by robust datasets. The article presents SliceVision-F2I, a dataset of synthetic samples for studying feature visualization in network slicing for next-generation networking systems. The dataset transforms multivariate Key Performance Indicator (KPI) vectors into visual representations through four distinct encoding methods: physically inspired mappings, Perlin noise, neural wallpapering, and fractal branching. For each encoding method, 30,000 samples are generated, each comprising a raw KPI vector and a corresponding RGB image at low-resolution pixels. The dataset simulates realistic and noisy network conditions to reflect operational uncertainties and measurement imperfections. SliceVision-F2I is suitable for tasks involving visual learning, network state classification, anomaly detection, and benchmarking of image-based machine learning techniques applied to network data. The dataset is publicly available and can be reused in various research contexts, including multivariate time series analysis, synthetic data generation, and feature-to-image transformations.

SliceVision-F2I: A Synthetic Feature-to-Image Dataset for Visual Pattern Representation on Network Slices

TL;DR

SliceVision-F2I introduces a synthetic KPI-to-image dataset for network slicing, pairing 10 KPIs with RGB encodings produced by four visual methods under realistic noise. It demonstrates that pattern-based CNNs using Perlin noise, neural wallpapering, fractal branching, or physically guided mappings can achieve near-perfect slice-type classification, outperforming traditional KPI-based ML even at low image resolutions. The dataset supports three slice types (eMBB, URLLC, mIoT) with deliberate class imbalance and realistic data gaps, serving as a benchmark for vision-based network analytics. The work highlights the potential of visual KPI representations to improve robustness to measurement uncertainty, enable real-time processing, and provide a flexible platform for future hybrid models and pattern-design optimization in network management.

Abstract

The emergence of 5G and 6G networks has established network slicing as a significant part of future service-oriented architectures, demanding refined identification methods supported by robust datasets. The article presents SliceVision-F2I, a dataset of synthetic samples for studying feature visualization in network slicing for next-generation networking systems. The dataset transforms multivariate Key Performance Indicator (KPI) vectors into visual representations through four distinct encoding methods: physically inspired mappings, Perlin noise, neural wallpapering, and fractal branching. For each encoding method, 30,000 samples are generated, each comprising a raw KPI vector and a corresponding RGB image at low-resolution pixels. The dataset simulates realistic and noisy network conditions to reflect operational uncertainties and measurement imperfections. SliceVision-F2I is suitable for tasks involving visual learning, network state classification, anomaly detection, and benchmarking of image-based machine learning techniques applied to network data. The dataset is publicly available and can be reused in various research contexts, including multivariate time series analysis, synthetic data generation, and feature-to-image transformations.

Paper Structure

This paper contains 17 sections, 7 equations, 5 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Contextual Study
  3. Dataset Design
  4. KPI Simulation Model
  5. Visual Representation Methods
  6. Physically-Guided Pattern Generation
  7. Perlin Noise-Based Pattern
  8. Neural Wallpapering Pattern
  9. Fractal Branching Pattern
  10. Dataset Composition
  11. Potential Use Cases
  12. Experimental Analysis and Discussion
  13. Experimental Setup
  14. Performance Comparison and Comparative Analysis
  15. Key Findings
  16. ...and 2 more sections

Figures (5)

  • Figure 1: Directory Hierarchy of SliceVision-F2I
  • Figure 2: Physically-Guided Pattern Visualizations
  • Figure 3: Perlin Noise-Based Pattern Visualizations
  • Figure 4: Neural Wallpapering Pattern Visualizations
  • Figure 5: Fractal-Inspired Pattern Visualizations