9-1-1 DDoS: Threat, Analysis and Mitigation

Mordechai Guri, Yisroel Mirsky, Yuval EloviciarXiv 2016 · OpenReview

TL;DR

This paper reveals a novel and potent threat to the 911 emergency system: anonymous, unblockable DDoS attacks launched from compromised mobile phones. By exploiting cellular network protocols and embedding malware in the baseband firmware, attackers can effectively mask their identity, rendering traditional blocking mechanisms useless and potentially crippling emergency services for entire states with a surprisingly small number of compromised devices.

What Problem Does This Paper Address?

The 911 emergency service is a critical piece of US infrastructure. While Distributed Denial of Service (DDoS) attacks are a known threat, this paper focuses on a specific, highly concerning vulnerability within the 911 system: the potential for mobile phone botnets to launch attacks that are both anonymous and unblockable. Current FCC regulations mandate that all 911 calls be routed regardless of caller identification, a rule designed for accessibility but which also creates an exploitable loophole. The challenge is that traditional DDoS mitigation relies on identifying and blocking malicious sources. This paper addresses how attackers can circumvent this by making their sources untraceable and unidentifiable within the cellular network, thereby posing a significant threat to the availability of this vital service.

Methods

The core of this paper's contribution lies in its novel approach to creating and analyzing anonymous 911 DDoS attacks. The authors developed three types of "bots" that can be deployed on mobile phones:

  • Non-Anonymized (NA) Bot: This is a more straightforward bot that operates from the main OS. While it can launch DDoS attacks, its identifiers (IMSI/IMEI) are still exposed to the network, making it susceptible to blocking.
  • Anonymized (A) Bot: This bot operates at a deeper level, within the baseband firmware. It achieves anonymity by forcing the device into a "no-SIM" state, effectively masking its IMSI and only exposing the IMEI. This makes it harder to trace and block.
  • Persistent Anonymized (A*) Bot: This is the most sophisticated and concerning bot. It also operates within the baseband firmware and goes a step further by randomizing its identifiers (IMSI and/or IMEI) before each 911 call. This means the device presents a different, untraceable identity with every attempt, making it virtually impossible to blacklist.

To demonstrate feasibility, the researchers implemented these bots and tested them on a small, controlled cellular network. They then used a discreet event simulator (DES) to model the E911 infrastructure and quantify the impact of these attacks under realistic conditions.

Datasets

The paper does not explicitly mention training datasets. For evaluation and simulation, they utilized:

  • Published surveys, reports, and statistics: These were used to model the E911 network topology, call volume statistics, and configurations.
  • North Carolina's E911 infrastructure data: Specific data from a 2008 survey of North Carolina's E911 network was used to create a detailed state-level model (GNC).
  • FCC master PSAP registry: Used for country-level analysis.
  • NENA (National Emergency Number Association) data: Used for staffing requirements and call volume statistics.

Key Insights and Ideas

  • Baseband firmware is the new frontier for mobile attacks: The paper highlights that by compromising the baseband firmware (the low-level software managing cellular communication), attackers can achieve a level of anonymity and control that is impossible with OS-level malware.
  • Exploiting "no-SIM" state for anonymity: The "no-SIM" state, designed for emergency calls when a SIM is absent, is cleverly exploited by the A bot to mask its identity.
  • Randomized identifiers are the ultimate evasion: The A* bot's ability to constantly change its identifiers makes it a truly unblockable threat, as any blacklist would be rendered useless in real-time.
  • Surprisingly few bots can cause significant disruption: The paper demonstrates that a relatively small number of compromised devices can overwhelm the 911 system, leading to widespread service denial.
  • The 911 infrastructure is vulnerable due to its design: The reliance on routing all emergency calls regardless of identification, while crucial for accessibility, is a fundamental weakness exploited by these attacks.

Key Results

  • State-level impact: With as few as 6,000 bots (approximately 0.0006% of North Carolina's population), attackers can deny 20-50% of wireline and wireless callers access to 911 services.
  • Country-level impact: Deploying 200,000 bots (approximately 0.0006% of the US population) can cause 33% of legitimate callers nationwide to give up on reaching 911.
  • Call processing time increase: A 6K bot attack can increase legitimate call service time by 40%, meaning callers might wait an additional 45 seconds to 3 minutes and make multiple redial attempts.
  • SR overload: In older networks using MF signaling, an attack with 6K bots could overload 87% of SRs, impacting 98% of 911 callers.
  • Cost-effectiveness: The paper estimates that blocking emergency services in an entire state could be achieved with less than $100K in hardware (for the bots).

Limitations and Future Work

  • Ethical considerations: The paper acknowledges that some proposed countermeasures, like disallowing NSI calls, raise ethical concerns and could prevent legitimate users from accessing emergency services.
  • Implementation complexity: While the A* bot offers the highest anonymity, its implementation requires deep access to baseband firmware, which can be challenging.
  • Future work: The authors suggest further research into device-level and network-level countermeasures, particularly those that can detect and mitigate anonymous attacks. They also emphasize the need for lawmakers and security professionals to understand this threat to develop effective prevention strategies. The paper also points to the need for upgrading older signaling methods (like MF) to SS7 to improve network resilience.

Summary updated

Abstract

The 911 emergency service belongs to one of the 16 critical infrastructure sectors in the United States. Distributed denial of service (DDoS) attacks launched from a mobile phone botnet pose a significant threat to the availability of this vital service. In this paper we show how attackers can exploit the cellular network protocols in order to launch an anonymized DDoS attack on 911. The current FCC regulations require that all emergency calls be immediately routed regardless of the caller's identifiers (e.g., IMSI and IMEI). A rootkit placed within the baseband firmware of a mobile phone can mask and randomize all cellular identifiers, causing the device to have no genuine identification within the cellular network. Such anonymized phones can issue repeated emergency calls that cannot be blocked by the network or the emergency call centers, technically or legally. We explore the 911 infrastructure and discuss why it is susceptible to this kind of attack. We then implement different forms of the attack and test our implementation on a small cellular network. Finally, we simulate and analyze anonymous attacks on a model of current 911 infrastructure in order to measure the severity of their impact. We found that with less than 6K bots (or $100K hardware), attackers can block emergency services in an entire state (e.g., North Carolina) for days. We believe that this paper will assist the respective organizations, lawmakers, and security professionals in understanding the scope of this issue in order to prevent possible 911-DDoS attacks in the future.
BibTeX
@misc{guri2016ddos,
  author = {Mordechai Guri and Yisroel Mirsky and Yuval Elovici},
  title = {{9-1-1 DDoS: Threat, Analysis and Mitigation}},
  year = {2016},
  eprint = {1609.02353},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1609.02353},
}
Figure