Telecoms Under Pressure: When Emergency Access Breaks Down

Building end-to-end resilience for 4G and 5G emergency calling

Emergency calling has become a critical public safety service, with reliability depending on far more than just mobile coverage. As fixed-line services and circuit-switched networks are retired, the public and first responders increasingly rely on packet-switched 4G and 5G networks, creating greater exposure to software, device, infrastructure, process and governance failures.

This paper examines how emergency calls can fail – from handsets to emergency call centres – and proposes an end-to-end resilience programme for mobile network operators (MNOs), regulators, emergency service organisations, manufacturers and suppliers.

Emergency services are obviously of vital concern to society at large. They are also of vital concern for the MNOs licensed to provide an emergency calling service, which can incur heavy penalties in the event of failure and for whom this is not just a reliability issue but a survival issue. Emergency callers and first responders may also be on the same network, which further necessitates the need for the service to work.

This paper’s central shift is from measuring network availability to assuring the emergency service outcome. Its consolidated recommendations are to:

• Test completed emergency calls from representative devices to emergency call centres every six to 12 months, including roaming, limited service, unauthenticated and no-SIM use cases.
• Monitor completed call outcomes, set-up time, routing, voice continuity, location and callback rather than the status of network components alone.
• Maintain standards-compliant devices, coordinated device-allowing and blocking processes and common emergency call behaviour.
• Review packet core, voice, timing, firewall, Internet Protocol (IP) addressing and fallback behaviour after material changes.
• Provide base station emergency call logs, session-based alarms, automated synthetic testing and useful handset feedback.
• Map critical physical and logical assets, reduce common mode dependencies and separate user, control and management paths.
• Maintain independent recovery routes, sufficient failover capacity, site power and emergency physical access.
• Treat public emergency calling and responder communications as separate but interdependent life-safety services.
• Inform operators, suppliers, regulators, emergency services, executives and customer teams together.
• Promote open, anonymised, “no-blame” information sharing so that the industry can learn from failures and solutions.

While these actions cannot guarantee that failures will never occur, they can create a system in which failures are less likely, smaller in scope, detected earlier, recovered from faster and less capable of blocking access to emergency assistance when help is needed.