Evolution from 4G to 5G

On behalf of ZTE, I’d like to share the latest activity we have regarding 5G. We will focus more on the road from 4G to 5G. When 5G is mentioned, most of time, we think that it is something far away. But when we consider 5G as a group of technologies, we may find something different.

Pre5G

The first question is: Of all the 5G candidate technologies, is there anything that will be truly revolutionary? The answer is no. We know that some candidate technologies will be highly disruptive, but none will be truly revolutionary because all these technologies are still based on the frequency, time and space domains. 

The second question is: Is there anything we can use from 4G? Some 5G technologies can be evolved from 4G, and we can identify some of the 5G candidate technologies back in 4G. This has been our approach to pre5G. 

Pre5G has four components. First, it uses 5G technology. Second, it uses 4G handsets. This is the most important thing. We don’t have to modify 4G handsets. Third, we should provide a 5G-like experience. It is not 5G experience, but it is similar to 5G experience. Fourth, timing is important. 5G will be available much earlier than 2020. We know the target is 2020, but we can achieve pre5G this year. 

Massive MIMO

The first technology we have identified for pre5G is massive MIMO. We know that this will probably be the most important technology, even for true 5G. According our preliminary simulations, massive MIMO can have four to six times the capacity of 4G. However, massive MIMO may need more than 100 antennas, which is 10 times more than 4G right now.

There are three main challenges that need to be overcome in massive MIMO.

Overhead

When there are 10 times more antennas, the overhead should be 10 times more. We know that for LTE, the overhead from RS is 4%, and for massive MIMO, it may be more than 18%. This is unacceptable. After innovating with TDD mode, we found that we don’t need to depend on the feedback in TDD mode. This means we can achieve massive MIMO in 4G using TDD mode.

Size

The second issue is size. Because we have 10 times more antennas, the concern is how big the antenna beam is. After analysis, we found that the target size for the massive MIMO system comprising 128 antennas may be similar to that of the current 8-antenna system, which China Mobile has commercially deployed.

Complexity

128 antennas may be 10 times or even 100 times more complex than 4G. To address this issue, we developed a baseband chipset for the processor. We use a vector processor, which is good for all kinds of 4G and advanced 5G technologies. With this chipset, we can support massive MIMO right now.

Massive MIMO 3D Coverage Field Trials

We found the massive MIMO is very good for tall building coverage, because the beam forming is three-dimensional, not two-dimensional. The most interesting thing is the boost in capacity. Our field tests showed that we can triple the capacity of 4G. At MWC, we demonstrated how we could even quadruple capacity.

MUSA

The second technology we identified for pre5G is multi-user shared access (MUSA). There are two big advantages of MUSA: 1) It overcomes the near-far effect, so we can better balance the central user and remote user; and 2) it is very good for massive connection. For 5G, massive connection is very important, and MUSA may be a key technology to address this.

Summary