Deutsche Telekom VP: 5G at risk of being ‘4G++’

Features

A senior Deutsche Telekom executive has claimed that the forthcoming 3GPP Release 16 was “make or break for 5G”, as he urged the industry to make 5G standards more innovative in its approach.

Speaking at the 5G World conference in London, Deutsche Telekom’s Core Network VP Franz Seiser said the standard is currently too focused on providing enhanced mobile broadband (eMBB) and would not be able to deliver on key use-cases.

“What [3GPP is] doing is 4G++,” Seiser said. “This is just the core network fully optimised for mobile broadband.

“This is still very much based on evolution of an [evolved packet core]. This is still very close to legacy thinking.”

Seiser’s criticisms were focused on Release 15, scheduled for 2018, which will deliver the first set of 5G standards. The next phase of specifications, Release 16, is expected in 2019.

“What 3GPP is currently doing with Release 15 is not revolution, it’s evolution,” said Seiser.

He said three major concepts needed more focus in the 5G standard: ultra-dense networks, network slicing and a cloud-native network.

To provide the ultra-dense network, Seiser highlighted fixed-mobile convergence as a key area. “Borders between fixed and mobile will disappear,” he said, adding that “3GPP still sees itself as mobile-centric”.

He said while the industry agrees on certain principles around network slicing, there are fundamental differences in viewpoints that need to be resolved.

“If I talk to three people I usually get five answers,” he said, adding that some in the industry view network slicing as “a kind of advanced quality of service management” and others a “completely self-contained network”.

As for the cloud-native network, Seiser highlighted NFV and modularisation as areas requiring more focus in the standard.

However, at the same event, Adrian Scrase, CTO of standards body ETSI, told Mobile Europe that the current prioritisation of eMBB in the standard is “natural”.

“It will please everyone but not in the same instance,” he said. “The standard cannot cater for the extreme use-cases at the same time – we’ve had to prioritise. So the very first standard we produce will be heavily focused on evolved mobile broadband. It’s not going to offer an awful lot for massive IoT, that’s being pushed into the second phase.”

Scrase cited the triangle of requirements from 5G. One of these is massive IoT, or enabling a large number of low power and low cost connections. Another is critical communications, based on low latency and high reliability. The other point on the triangle is the eMBB, which mainly means high throughput capacity.

“It won’t be until after 2020 that it satisfies all of the people in that triangle; maybe after 2025 for the extreme cases of low latency, ultra-reliable networks.”

To define the direction of 5G, ETSI took an opinion poll of its members, which include network operators, manufacturers and users from across Europe, with eMBB emerging as the clear winner for initial focus in 5G standardisation. However, some members successfully pushed for some low latency parameters to be built in to cater to vehicle to vehicle communications.

Despite ETSI members' support for eMBB, Scrase perceives a regional divide in priorities for 5G.

“The Asian operators are very clear: it’s mobile broadband because we’ve got customers who are willing to pay. But in Europe, maybe we’re not quite so sure.

“You go for mobile broadband because [you believe] people will pay more for more mobile broadband, but by doing that you are expecting to take more money out of the pockets of the same people.

“Are you sure that will happen? If you go for other two corners of the triangle and do massive IoT or ultra-reliable, you’re going for a different user base and taking money out of different pockets, but you’re not so sure you can monetise that use-case.”

Deutsche Telekom’s Seiser also suggested a divide between European operators and their counterparts in Asia and the US.

“The pushers for early 5G implementation are not Europe, they are Asia and US. Their use cases they envision are very close to enhanced mobile broadband.

“But that’s nothing that helps in Europe. We have a totally different selling point,” he said.

However, Seiser’s assessment was not completely bleak. He highlighted the work in 3GPP to include service-based architecture in the standard as a “good step” towards the all-important cloud-native architecture.

However, he said further action was needed and added: “This is make or break. That’s a call to the industry. We need to have a joint push to come up with much more disruptive ways forward."