How Memory-Safe is IoT? Assessing the Impact of Memory-Protection Solutions for Securing Wireless Gateways Conference: ; ; CCS: Security and privacy Systems security CCS: Networks Wireless access points, base stations and infrastructure

Vadim Safronov, Ionut Bostan, Nicholas Allott, Andrew Martin · 2024

Read How Memory-Safe is IoT? Assessing the Impact of Memory-Protection Solutions for Securing Wireless Gateways Conference: ; ; CCS: Security and privacy Systems security CCS: Networks Wireless access points, base stations and infrastructure by Vadim Safronov, Ionut Bostan, Nicholas Allott, Andrew Martin (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2411-01377.txt.

Description

A 2024 arXiv research paper by Vadim Safronov, Ionut Bostan, Nicholas Allott, Andrew Martin. The rapid development of the Internet of Things (IoT) has enabled novel user-centred applications, including many in safety-critical areas such as healthcare, smart environment security, and emergency response systems. The diversity in IoT manufacturers, standards, and devices creates a combinatorial explosion of such deployment scenarios, leading to increased security and safety threats due to the difficulty of managing such heterogeneity.

Catalog record

CREATORVadim Safronov, Ionut Bostan, Nicholas Allott, Andrew Martin
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2411-01377

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

How Memory-Safe is IoT? Assessing the Impact of Memory-Protection Solutions for Securing Wireless Gateways Conference: ; ; CCS: Security and privacy Systems security CCS: Networks Wireless access points, base stations and infrastructure Vadim Safronov, Ionut Bostan, Nicholas Allott, Andrew Martin Abstract . The rapid development of the Internet of Things (IoT) has enabled novel user-centred applications, including many in safety-critical areas such as healthcare, smart environment security, and emergency response systems. The diversity in IoT manufacturers, standards, and devices creates a combinatorial explosion of such deployment scenarios, leading to increased security and safety threats due to the difficulty of managing such heterogeneity. In almost every IoT deployment, wireless gateways are crucial for interconnecting IoT devices and providing services, yet they are vulnerable to external threats and serve as key entry points for large-scale IoT attacks. Memory-based vulnerabilities are among the most serious threats in software, with no universal solution yet available. Legacy memory protection mechanisms, such as canaries, RELRO, NX, and Fortify, have enhanced memory safety but remain insufficient for comprehensive protection. Emerging technologies like ARM-MTE, CHERI, and Rust are based on more universal and robust Secure-by-Design (SbD) memory safety principles, yet each entails different trade-offs in hardware or code modifications. Given the challenges of balancing security levels with associated overheads in IoT systems, this paper explores the impact of memory safety on the IoT domain through an empirical large-scale analysis of memory-related vulnerabilities in modern wireless gateways. Our results show that memory vulnerabilities constitute the majority of IoT gateway threats, underscoring the necessity for SbD solutions, with the choice of memory-protection technology depending on specific use cases and associated overheads. 1. Introduction The rapid rise of IoT devices and applications is revolutionising various industries. Currently, there are billions of deployed IoT devices, with billions more expected in the coming years ( Sinha 2024 ) . To provide users and businesses with a wide array of smart features, such as adaptive climate control, smart parking, traffic safety, healthcare, and smart environmental safety, IoT devices are manufactured globally, transmit data worldwide, receive over-the-air updates from various locations, and utilise a myriad of third-party software libraries. This uncontrolled heterogeneity in the production of IoT devices, their communication patterns, and applications may result in significant safety, security and privacy threats on a global scale. Even a small vulnerability in one component of this complex IoT ecosystem ( Conti et al. 2016 ; Copos et al. 2016 ) could significantly disrupt the entire security and safety infrastructure of large organisations, including cyber-critical systems ( Karale 2021 ; Swessi and Idoudi 2022 ) , such as healthcare or emergency response systems. A large number of cyberthreats can be triggered by memory-based vulnerabilities ( Obaidat et al. 2023 ) , resulting in broad and intricate attack chains. For example, a single compromised router can become the entry point for attacking IoT devices, which subsequently compromises entire networks, escalating to widespread cyber threats. Notable incidents highlight the urgency of this issue: the Dyn DNS attack, where a multitude of IoT devices were used to launch a massive Distributed Denial of Service (DDoS) attacks, the SolarWinds breach, and Log4j incidents all exemplify the wide systemic consequences of vulnerabilities (however they arise). Although risk management has helped to stabilise many contexts, incident responses, and disaster containment plans, it is increasingly clear that step-changes are needed in order to eliminate whole classes of vulnerabilities. Doing so offers immense potential for improved security resilience: failure to address these fundamentals will leave cyber space vulnerable to unpredictable systemic failures indefinitely, posing risks not just to individual systems but to large ecosystems, the wider economy, and public health. Millions of new IoT devices emerge annually, potentially introducing novel, unknown security threats, partly due to the lack of centralised control and standardisation. Moreover, this trend highlights the continuous introduction of fresh vulnerabilities, both in new systems and in the patches intended to fix old ones. This procession of endlessly-broken interconnected IoT devices, networks and systems pushes us to the emerging paradigm of Security by Design (SbD). Whilst system architects have long understood — or been enjoined to take account of — the security of their systems and the privacy of the data they process, it has not always been a priority. Improved designs — and better implementation primitives — help to eliminate large classes of vulnerabilities. The UK Government’s ‘Digital Security by Design’ initiative is an example of such an approach, promoting the CHERI architecture with a promise of low-cost memory protection, claimed to eliminate at a stroke around 70% of the vulnerabilities present in a wide range of systems, according to analysis from Microsoft Security Response Centre ( Miller 2019 ) . Along with the CHERI architecture, other promising SbD approaches for memory safety include memory-safe languages like Rust ( Matsakis and Klock II 2014 ) and memory-tagging techniques such as ARM-MTE ( Arm 2019 ) . Each approach involves trade-offs between security guarantees and associated overheads. Rust and CHERI offer deterministic memory protection, capable of defending against known and future memory-safety threats, but require hardware modifications for CHERI and significant code rewrites for Rust during the porting of memory-unsafe software. In contrast, memory-tagging techniques like ARM-MTE have less security robustness due to the probabilistic nature of the tagging approach, yet they require fewer hardware and code changes, making them easier to implement in modern processors and to adapt to existing software. 1.1. Research questions Motivated by the urgent need for a thorough investigation into the cyber threats prevalent in IoT and the role of memory protection in this domain, our study maps out and classifies cyber threats within wireless gateway firmware, exploring the impact of memory-protection technologies on their mitigation. We conducted an empirical large-scale vulnerability analysis based on the examination of 6,335 firmware images of modern wireless gateways. The focus on wireless gateways stems from the fact that routers serve as sweet spots for attacks originating both from inside and outside IoT networks, with the potential to propagate across entire network infrastructures, particularly as they integrate IoT networks with safety-critical systems like building security, traffic management, and healthcare. Specifically, our research addresses the following questions: • RQ1: What is the ratio and classification of memory safety threats compared to all vulnerabilities in modern wireless gateway firmware? (§ 4 ) • RQ2: What is the quantitative impact of memory protection on wireless gateways, and how effectively can SbD solutions guarantee it in terms of vulnerability coverage, transfer overheads, and future-proof potential? (§ 5 ) 1.2. Contributions The key contributions of our research are as follows. • We empirically analysed existing vulnerabilities in wireless gateway firmware (§ 4 ). From a total of 6,335 router firmware images, we sampled and analysed 502 firmware binary samples, identifying 17,341 occurrences of common vulnerabilities and exposures (CVEs). We quantitatively categorised the identified vulnerabilities by memory relevance and severity level, providing insights into the ratio and types of memory safety threats in relation to all identified threats. • Based on the vulnerability analysis, we quantitatively measured the impact of SbD memory protection (§ 5 ). Our findings show that deploying SbD solutions in wireless gateways can eliminate 74% of known CVEs, increasing the security of average router

Read the complete text