Our mobile networks still depend on fibre

Mobile is not truly wireless; Every call depends on fiber and without fiber the whole system would break down, writes Paul Budde.

PEOPLE OFTEN ask a seemingly logical question: if most Australians use mobile phones for calls and internet access, why do we still need to invest billions of dollars in fixed networks such as fibre? At first glance, it may seem that mobile networks can do this job on their own.

The reality is very different. Nearly all mobile traffic ultimately depends on fixed, fiber optic infrastructure. Without this, mobile networks cannot operate at the scale, reliability or quality people expect, especially for voice services and emergency calls.

When you make a mobile phone call, your phone connects wirelessly to the nearest cell tower. This is the only part of the trip that is truly “wireless”. From the tower, your call reaches the operator’s core network via landlines and from there to the person you are calling. In metropolitan and regional areas of Australia these towers are almost always connected via fibre-optic cables.

This means that the mobile signal usually only covers the last few kilometers, sometimes much less. The heavy lifting is done by fiber, which provides the capacity, stability and low latency that modern communications require.

In regions where fiber is not available, operators have traditionally relied on microwave connections. These are point-to-point radio links between towers or relay zones, using line-of-sight links that often exceed tens of kilometers. Microwaves can work well, but require towers, power, maintenance access, and open transmission paths. In very remote areas of Australia this infrastructure may be difficult or uneconomic to install.

This is where satellite backhaul comes into play. In recent years, low Earth orbit (LEO) satellite systems have been introduced as a way to connect remote mobile towers that cannot be reached by fiber or microwave. in Australia, Telstra LEO became the first major carrier to deploy satellite backhaul at scale. OneWebowned by Eutelsat.

The way this works is relatively simple in principle. A mobile tower in a remote location sends calls and data to a passing LEO satellite. The satellite then transmits these signals to a ground station that connects via fiber to Telstra’s core network. From there, traffic behaves like any other call or data session on the national network.

This approach has allowed Telstra to extend mobile coverage to places that would otherwise be impossible or prohibitively expensive. But as recent reports and community feedback show, it’s also important limitations.

LEO satellite backhaul is still a relatively new technology in this context. Telstra’s rollout only started in early 2024, and the OneWeb constellation is still being expanded. Capacity will remain limited until more satellites come into orbit. When too many users share a limited satellite connection, audio quality is often the first thing to suffer. Calls may be interrupted, voices may sound garbled, or conversations may stop altogether.

Unlike data apps that can pause, resend information, and recover from brief interruptions, voice calls require a constant, real-time connection. Even brief outages that are not recorded as “downtime” in network monitoring systems can make it extremely difficult or impossible to resume phone calls.

Another challenge is that LEO satellites are constantly moving. Mobile towers must transfer traffic from one satellite to another as it passes overhead. Although this process is designed to be uninterrupted, in practice it can cause momentary interruptions. To users, this may seem like random interruptions rather than clearly defined interruptions.

These technical facts help explain why some remote communities are reporting ongoing call quality problems even though operators indicate services are largely available. From a network perspective, the system may be “working”. From a user perspective, it can be unreliable in daily use, especially for critical services like Triple Zero calling.

More satellites would help. Eutelsat has announced large additional orders for the new LEO spacecraft, which will increase capacity, flexibility and turnover performance over time. But even with these upgrades, satellite backhaul will never match the reliability of fiber or well-designed microwave links. It remains, by definition, a compromise solution.

There is also growing interest in a different satellite approach: direct-to-device services, where ordinary cell phones connect directly to satellites without being tied to a local tower. This model is being aggressively pursued by Starlink (among others) and we have discussed its potential in previous articles. Direct-to-satellite services could ultimately provide a valuable safety net for essential connectivity and emergency messaging in remote areas.

However, they do not eliminate the need for a fixed network. High capacity fiber remains the backbone of Australia’s communications system. Mobile networks, satellites and fixed infrastructure are not alternatives to each other; these are complementary layers of the same ecosystem.

The current debate about satellite backhaul failures is therefore not an argument against innovation. We would like to remind you that there are no shortcuts in telecommunication. Extending coverage to the most remote parts of Australia is crucial – but this honesty realistic expectations about the limitations and what satellite-based solutions can and cannot provide.

Paul Budde IA is a columnist and managing director of independent telecommunications research and consultancy. Paul Budde Consulting. You can follow Paul on Twitter @PaulBudde.

Support independent journalism Subscribe to IA.