Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Federated Hyperdimensional Computing for Resource-Constrained Industrial IoT

Nikita Zeulin, Olga Galinina, Nageen Himayat, Sergey Andreev

Abstract

In the Industrial Internet of Things (IIoT) systems, edge devices often operate under strict constraints in memory, compute capability, and wireless bandwidth. These limitations challenge the deployment of advanced data analytics tasks, such as predictive and prescriptive maintenance. In this work, we explore hyperdimensional computing (HDC) as a lightweight learning paradigm for resource-constrained IIoT. Conventional centralized HDC leverages the properties of high-dimensional vector spaces to enable energy-efficient training and inference. We integrate this paradigm into a federated learning (FL) framework where devices exchange only prototype representations, which significantly reduces communication overhead. Our numerical results highlight the potential of federated HDC to support collaborative learning in IIoT with fast convergence speed and communication efficiency. These results indicate that HDC represents a lightweight and resilient framework for distributed intelligence in large-scale and resource-constrained IIoT environments.

Federated Hyperdimensional Computing for Resource-Constrained Industrial IoT

Abstract

In the Industrial Internet of Things (IIoT) systems, edge devices often operate under strict constraints in memory, compute capability, and wireless bandwidth. These limitations challenge the deployment of advanced data analytics tasks, such as predictive and prescriptive maintenance. In this work, we explore hyperdimensional computing (HDC) as a lightweight learning paradigm for resource-constrained IIoT. Conventional centralized HDC leverages the properties of high-dimensional vector spaces to enable energy-efficient training and inference. We integrate this paradigm into a federated learning (FL) framework where devices exchange only prototype representations, which significantly reduces communication overhead. Our numerical results highlight the potential of federated HDC to support collaborative learning in IIoT with fast convergence speed and communication efficiency. These results indicate that HDC represents a lightweight and resilient framework for distributed intelligence in large-scale and resource-constrained IIoT environments.
Paper Structure (12 sections, 5 figures, 2 algorithms)

This paper contains 12 sections, 5 figures, 2 algorithms.

Table of Contents

  1. Introduction
  2. Principles of Hyperdimensional Computing
  3. Foundational Concepts
  4. Fundamental Procedures in HDC Systems
  5. Challenges in Federated Learning for IIoT
  6. Resource Constraints
  7. Scalability in Massive Deployments
  8. Federated HDC Framework
  9. Numerical Example
  10. Datasets and Data Preprocessing
  11. Discussion of Numerical Results
  12. Discussion and Conclusions

Figures (5)

  • Figure 1: Illustration of envisioned industrial IoT system.
  • Figure 2: Architecture of proposed resource-efficient federated HDC framework for resource-constrained devices. Edge devices encode local data into hypervectors and update class prototypes. Instead of transmitting full model parameters, devices send compact prototype vectors to a central server for aggregation.
  • Figure 3: Illustrative comparison of maximum accuracy achieved by baseline and proposed federated HDC methods.
  • Figure 4: Total uplink traffic consumption to achieve baseline federated HDC accuracy.
  • Figure 5: Comparison of maximum accuracy and volume of uplink traffic between proposed method and baseline for non-i.i.d. scenario.