Secure Key Establishment Mechanism for Smart Sensing System Based Robots Network

Qi Xiao, Yunchuan Qin, Cheng Xu, Kenli Li2020 · DOI

TL;DR

This paper introduces a novel, polynomial-based key establishment mechanism for multi-robot systems. It addresses the critical security challenges in dynamic, resource-constrained robot networks by enabling secure, efficient key generation and distribution, allowing robots to join and leave the network seamlessly.

What Problem Does This Paper Address?

The increasing integration of robots into various aspects of society, particularly in multi-robot systems, brings significant advancements but also introduces complex security vulnerabilities. These systems, often operating in open and dynamic environments, rely on communication between sensors, actuators, and edge computing systems. This communication is susceptible to attacks due to the distributed nature and limited hardware resources of individual robots. Traditional key distribution methods struggle to meet the security demands of these environments, especially when robots frequently join or leave the network. The paper specifically targets the challenge of establishing and maintaining secure communication keys in such dynamic and resource-constrained multi-robot networks, where traditional cryptographic approaches may be too computationally intensive or inflexible.

Methods

The core of this paper is a novel security key establishment mechanism based on high-order polynomials. The proposed system operates with a cloud-based Key Distribution Center (KDC).

Here's a breakdown of the key components:

  • Key Generation and Distribution: The KDC generates session keys for clusters of robots. Instead of directly sending keys, it sends coefficients of a high-order polynomial and a key update instruction.
  • Private Information: Each robot component receives unique private information (SNID) during initialization, which is crucial for calculating the session key.
  • Polynomial-Based Key Calculation: Robots use their private information and the received polynomial coefficients to calculate their session key for a given time slot. This calculation is designed to be computationally light, suitable for resource-constrained robots.
  • Dynamic Network Handling: The mechanism is designed to handle robots joining or leaving the network dynamically. The KDC can update the key distribution polynomial to accommodate these changes.
  • Attack Detection and Response: When a robot is detected as attacked, its ID is broadcast to all other components. The KDC then issues a new key update instruction, effectively excluding the attacked robot from future key calculations and communication. This also ensures forward and backward security.
  • Message Authentication: A Message Authentication Code (MAC) is used to verify the integrity and authenticity of the key update instructions.

The use of high-order polynomials allows for a flexible and secure way to distribute keys, where the coefficients themselves act as the secret information, and the polynomial evaluation provides the key. The interference polynomial $\delta(X)$ is a key innovation for handling attacked components.

Datasets

Not discussed in the paper. The evaluation is based on simulations and comparisons with existing schemes.

Key Insights and Ideas

  • Polynomials as a Flexible Key Distribution Tool: The paper demonstrates that high-order polynomials can be effectively used to distribute secret keys in a dynamic network. The coefficients of the polynomial act as the secret, and the polynomial evaluation provides a computationally efficient way to derive the key.
  • Decoupling Key Generation from Direct Transmission: By sending polynomial coefficients instead of raw keys, the system reduces the direct exposure of sensitive key material during transmission.
  • Robustness to Dynamic Network Changes: The mechanism's ability to seamlessly integrate new robots and exclude compromised ones without significant overhead is a major advantage for real-world multi-robot deployments.
  • Balancing Security and Resource Constraints: The proposed method strikes a good balance between strong security guarantees and the limited computational and storage resources of typical robot components.
  • Forward and Backward Security: The design ensures that even if a current key is compromised, past and future keys remain secure, a critical property for long-term system integrity.

Key Results

The paper presents a theoretical mechanism and analyzes its performance through comparisons with existing schemes.

  • Storage Cost: The proposed mechanism requires a minimal storage cost of 160 bits (16 bytes) per component, which is independent of the node's lifetime and significantly lower than some existing schemes.
  • Communication Cost: The communication cost is analyzed and shown to be efficient, particularly when the number of normal and attacked components is balanced. The cost is expressed as a function of the number of normal ($m$) and attacked ($p$) components, generally $C = 2(m+2)$ for $m \ge p$ and $C = 2(p+2)$ for $m < p$.
  • Security: The mechanism is claimed to provide high security, including resistance to collusion attacks and invulnerability to individual component attacks. It ensures forward and backward security.
  • Flexibility: The mechanism is highly flexible, allowing for dynamic joining and leaving of network components, making it suitable for mobile robots.

Limitations and Future Work

  • Assumptions on Network Connectivity: The transferability of the mechanism relies on continuous network connectivity to the cloud-based KDC. If connectivity is lost, key updates may be hindered.
  • Complexity of Polynomial Coefficients: While the calculation is light for robots, the generation and management of polynomial coefficients by the KDC might become complex with a very large number of robots and clusters.
  • Practical Implementation Details: The paper focuses on the theoretical framework. Further research would be needed to explore the practical implementation challenges, such as the secure establishment of the initial private information and the robustness of the KDC itself.
  • Future Work: The authors suggest that the mechanism's flexibility makes it suitable for various application scenarios. Further performance evaluations and real-world deployments would be valuable.

Summary updated

BibTeX
@misc{xiao2020secure,
  author = {Qi Xiao and Yunchuan Qin and Cheng Xu and Kenli Li},
  title = {{Secure Key Establishment Mechanism for Smart Sensing System Based Robots Network}},
  year = {2020},
  doi = {10.3390/s20071970},
}
Figure