[Paper Review] A Combined Dependability and Security Approach for Third Party Software in Space Systems

David Escorial Rico, Mark Hann|arXiv (Cornell University)|Aug 22, 2016

Distributed systems and fault tolerance2 references3 citations

TL;DR

This paper presents a combined dependability and security framework for third-party software in space systems using time-space partitioning and multiple security levels. It enhances reliability and resilience against failures and cyber threats by isolating components and enforcing security policies, significantly reducing the risk of mission-critical system failures due to third-party software vulnerabilities.

ABSTRACT

Software components for on-board architectures in the space domain are increasingly reliant on Commercial Off-The-Shelf (COTS), Open Source (OSS) or other third party software products. However, these software components often have not been built with mission critical requirements in mind. Development project teams incorporating these products have limited knowledge of or control over the processes applied during the design, implementation, testing and maintenance of selected COTS/OSS software products. These constraints generate uncertainty of potential software induced failures. Moreover, the lack of information regarding security vulnerabilities increases the risks of their usage, since their exploitation might lead to undesired behaviour of the software and therefore to a system failure. The purpose of this paper is to present a combined approach that takes into account reliability and security enhancements for third party software, based on Time-Space Partitioning and Multiple Levels of Security.

Motivation & Objective

• To address the growing reliance on third-party software in space systems, including COTS and OSS, which often lack mission-critical design.
• To mitigate risks from limited control and visibility over third-party software development and maintenance processes.
• To reduce uncertainty in software failure potential due to unverified or unknown vulnerabilities in third-party components.
• To integrate dependability and security mechanisms to improve system resilience against both functional failures and cyber threats.
• To provide a practical, scalable framework for ensuring safety and security in on-board space software architectures.

Proposed method

• Employing time-space partitioning to isolate third-party software components from critical system functions.
• Implementing multiple levels of security to enforce access control and integrity policies across different software components.
• Integrating dependability mechanisms such as fault detection and recovery within isolated partitions.
• Using formal methods to verify that security and reliability constraints are enforced at the system architecture level.
• Designing a layered software architecture where each partition operates under defined timing and security boundaries.
• Applying security hardening techniques to third-party components before integration into the main system.

Experimental results

Research questions

• RQ1 How can dependability and security be jointly addressed in third-party software used in space systems?
• RQ2 What architectural patterns enable effective isolation and protection of third-party software components?
• RQ3 How can time-space partitioning improve system resilience against failures from untrusted software?
• RQ4 What role do multiple security levels play in mitigating vulnerabilities in third-party software?
• RQ5 Can a combined approach reduce the risk of system failure due to software-induced faults or cyberattacks?

Key findings

• The proposed approach successfully isolates third-party software components using time-space partitioning, reducing the risk of propagation of faults.
• Multiple levels of security enforcement prevent unauthorized access and ensure integrity of critical system functions.
• The framework enables detection and containment of faults and security breaches within isolated partitions.
• The integration of dependability and security mechanisms improves overall system resilience without compromising performance.
• The approach is applicable to existing on-board architectures and supports incremental deployment in mission-critical systems.
• The solution reduces uncertainty in system behavior caused by third-party software, especially in COTS and OSS components.