Poster Abstract: Towards Scalable and Trustworthy Decentralized Collaborative Intrusion Detection System for IoT

Guntur Dharma Putra, Volkan Dedeoglu, Salil S. Kanhere, Raja Jurdak

Introduction

The rapid expansion of the Internet of Things (IoT) presents significant security challenges due to the massive scale and resource-constrained nature of connected devices. While individual Intrusion Detection Systems (IDS) can monitor single devices, a Collaborative IDS (CIDS) offers a more comprehensive defense by allowing multiple monitors to share threat intelligence. However, traditional CIDS architectures often struggle with two primary issues: trust in shared data and the scalability of storing large detection models.

Proposed Architecture

To address these challenges, this work proposes a modular, four-layer decentralized architecture that leverages blockchain and the InterPlanetary File System (IPFS).

The proposed four-layer decentralized CIDS architecture, illustrating the interaction between the application, trust management, storage, and blockchain layers.

The architecture is divided into the following layers:

  • Application Layer: Executes the IDS using a hybrid approach. It combines signature-based detection for known threats with anomaly-based detection using lightweight Support Vector Machines (SVM) for zero-day attacks.
  • Trust Management Layer: Utilizes Smart Contracts to enforce accountability. It manages an incentive and penalty system to ensure nodes contribute high-quality data and are penalized for malicious behavior.
  • Storage Layer: Offloads heavy data, such as SVM models and signature databases, to IPFS. This peer-to-peer storage protocol prevents the blockchain from becoming bloated.
  • Blockchain Layer: Serves as the immutable source of truth, storing only the cryptographic hashes of the models stored in IPFS and managing the consensus mechanism to filter out deceptive data.

Implementation and Experimental Setup

The system is designed with a gateway-centric approach, where resource-intensive tasks like blockchain consensus and machine learning are handled by gateways rather than individual sensors.

Experimental test-bed implementation featuring Ethereum nodes, IPFS clusters, and Raspberry Pi-based IoT nodes under simulated attack.

The proof-of-concept implementation utilizes a local test-bed consisting of:

  • IoT Nodes: Raspberry Pis acting as gateways, running Ethereum light nodes.
  • Blockchain Infrastructure: A private Ethereum network with full nodes for consensus.
  • Storage: A local IPFS cluster for model distribution.
  • Attacker Node: A dedicated system used to simulate DoS, MitM, and reconnaissance attacks.

Key Insights

A critical innovation of this design is the decoupling of data from metadata. By storing large models off-chain in IPFS and only keeping their "fingerprints" (hashes) on the blockchain, the system achieves high scalability. Furthermore, the use of Smart Contracts shifts the trust model from subjective reputation to objective, math-based verification.

Conclusion and Future Work

This ongoing research focuses on validating the architectural design and evaluating its performance against common IoT threats. Future work will address the challenges of adversarial machine learning, where attackers might attempt to poison models subtly. Additionally, the authors aim to quantify the computational and latency overhead introduced by the blockchain and IPFS layers to ensure the system remains viable for real-world IoT deployments.

Summary updated

Abstract

An Intrusion Detection System (IDS) aims to alert users of incoming attacks by deploying a detector that monitors network traffic continuously. As an effort to increase detection capabilities, a set of independent IDS detectors typically work collaboratively to build intelligence of holistic network representation, which is referred to as Collaborative Intrusion Detection System (CIDS). However, developing an effective CIDS, particularly for the IoT ecosystem raises several challenges. Recent trends and advances in blockchain technology, which provides assurance in distributed trust and secure immutable storage, may contribute towards the design of effective CIDS. In this poster abstract, we present our ongoing work on a decentralized CIDS for IoT, which is based on blockchain technology. We propose an architecture that provides accountable trust establishment, which promotes incentives and penalties, and scalable intrusion information storage by exchanging bloom filters. We are currently implementing a proof-of-concept of our modular architecture in a local test-bed and evaluate its effectiveness in detecting common attacks in IoT networks and the associated overhead.
BibTeX
@inproceedings{putra2020poster,
  author = {Guntur Dharma Putra and Volkan Dedeoglu and Salil S. Kanhere and Raja Jurdak},
  title = {{Poster Abstract: Towards Scalable and Trustworthy Decentralized Collaborative Intrusion Detection System for IoT}},
  booktitle = {IoTDI},
  year = {2020},
  url = {https://openreview.net/forum?id=HRP8Lg8k22z},
}
Figure