We are currently experiencing the most dynamic period the telecommunications industry has ever faced.
The Internet of Things (IoT) is transforming the world into an extension of ourselves. Beyond our smartphones and tablets, the number of Australian adults using wearable devices has increased from 7% in 2017 to 12% in 2018. By 2023, there could be more than 50 billion connected devices worldwide. In dollar value, the IoT market is expected to grow from US$157 billion in 2016 to US$457 billion by 2020.
Read more: Explainer: what is 5G?
Mobile providers are competing to increase their share of this ever growing market. But the current wireless networks are unequipped to service the extraordinary amount of information the IoT market will require. While 5G – the next generation mobile standard – could solve this problem, the question of how effective 5G will be in the future remains unanswered.
Networks must find new ways to respond this tremendous demand for data, using technologies that allow them to allocate bandwidth where it’s needed.
Recent network outages
Currently almost 96% of Australian adults use mobile phones, and 87% of those are using smartphones.
Recent incidents involving Australia’s largest mobile networks, Telstra and Optus, are prompting questions about how Australian networks will cope with increasing data demand, exacerbated by the emergence of IoT.
Telstra went through several major service outages in April and May that affected many Australians. The outage could have been catastrophic as the emergency number, 000, was also affected. The reason wasn’t thoroughly explained to the public.
In July, Optus was unable to provide the service that was promised to both new and existing costumers who signed up to the network to watch the 2018 FIFA World Cup.
While the former incident shows core network problems, the latter is due to network planning issues and insufficient bandwidth being allocated to service customers.
Network capacity: the limiting factor
All generations of mobile communication standards, including the current 4G and the future 5G, operate over a licensed spectrum – a frequency band which is predominantly used for this purpose.
But the spectrum is limited. Only a limited number of concurrent users can receive data at the rate promised by the operator. The more people using the network, the slower it gets. That’s why most of our mobile plans come with limited data allowance.
Offering unlimited data plans without proper network planning means slow speeds during peak congestion times, service interruptions and service failure.
Optus’ recent trial of unlimited data plans could only provide a very limited speed at peak hours, about 30 times less than the speed of 4G.
Considering demand for data is likely to increase, what can service providers do?
Dynamic data plans
Mobile broadband networks sell fixed amounts of data to users on a monthly basis, and allocate more bandwidth to users who request it when their original allocation is used up.
But most people do not use their entire data allowance. This should be taken into account when allocating bandwidth and billing customers.
Networks could offer dynamic data plans that are priced according to network demand.
A dynamic service would use advanced machine learning techniques to predict data demand. This could be used to allocate network bandwidth where it’s needed, leading to a more reliable service overall. People should pay only for what they are using according to the time of their use, overall network traffic and user demand. This is how Uber’s surge pricing works, where users pay more to use the service during peak periods.
Mobile providers could also consider offering plans that can be shared across devices. Relying on the conventional plans, such as pre-paid plans with a fixed allowance, is not efficient for many IoT devices. People may carry up to six smart devices in the future, and having a separate plan for each device will be frustrating.
On top of that, some devices use only a small fraction of the spectrum, so why pay for large bills? A smart meter at the basement of an apartment, for example, needs to send only few kilobytes of data per day to fulfil its function reporting daily energy usage.
New business models developed around dynamic network planning will ensure people receive reliable speeds at reasonable prices.
Won’t 5G fix this?
5G is expected to launch in Australia by 2020 and increase data speeds tenfold. It will operate over a wider spectrum, meaning more users can be serviced and possibly more bandwidth can be allocated to each user.
The service is designed to cater to mobile broadband users, as well as the potentially massive number of IoT applications. 5G has promised to provide connectivity for a large number of ultra-low power devices, such as smart home monitoring, and for applications that demand almost real-time, error-free communication, such as remote surgery.
The main question is how to fully unleash its potential. It’s not clear if 5G can service the data demand across such a diverse range of applications with the promised level of quality. Network providers have never experienced such diversity.
Without proper network planning, most of the 5G spectrum will be wasted. We will get higher data rate on average, but not all the time.
What’s at stake here? Without urgent change, users will be left with frustratingly slow data speeds and possibly more frequent network outages. As IoT technologies become more integrated into our lives, outages and failures are no longer tolerable and even a few seconds of service outage could be devastating.
Authors: Mahyar Shirvanimoghaddam, Academic Fellow, University of Sydney