HOW 5G MAY FAIL AND HOW TO AVOID IT

We are at the start of a rollout of the 5G wireless communication systems, which are promising to fundamentally change various industries, services and vertical sectors. There is a saying that only the even generations are successful (2G, 4G), such that 5G is doomed to fail and wait for the next generation. But, besides this semi-supersitious argument, there may be other solid arguments that 5G will not deliver the revolutionary promises. Motivated by these questions, I was invited by professor Liljana Gavrilovska, the General Chair of BalkanCom 2019 in Skopje, to organize a panel session aiming to identify the threats to the success of the 5G wireless systems and highlight some of the possible remedies. The panelists were:

– Ian F. Akyildiz (Georgia Tech, USA)(slides)

– Luiz DaSilva (Trinity College Dublin, Ireland) (slides)

– Merouane Debbah (Huawei, Paris) (slides)

– Driton Emini (Magenta Telekom, Austria) (slides)

– Alberto Leon Garcia (University of Toronto, Canada) (slides)

– Ivan Seskar (WINLAB, Rutgers University, USA) (slides)

The discussion was organized around several guiding questions, described below.

1. We have LTE, some of the previous G, WiFi, NB-IoT and some LWPA technologies, such as LoRa and SigFox. What is the pressing use case that requires 5G and cannot be supported by these technologies or their combination?

The main use cases that can be monetized right now are in the area of enhanced Mobile Broadband (eMBB), in order to improve wireless capacity. IoT use cases are uncertain and the telecom operators are still hesitant about the potential monetization. 5G may also have a use case in covering the large rural areas. Based on the spectrum licenses sold 5G will be deployed chiefly in the sub-6 GHz bands, such as 700 MHz and C-band, mmWave will come later.

2. Can the telecom operators sell low latency?

A real low-latency, but not high reliability, application that gets traction in China is gaming. However, it is not certain that this application will have proliferation worldwide. Most of the panelists have difficulties to see other low latency applications at this time. However, there was a broad consensus that low latency and ultra-reliability do not need to be always coupled together, as in URLLC, but be treated as two different features that can be coupled in the use cases where it makes sense, such as industrial automation. The accent should not be put to extremely low latency (< 1ms) but rather low guaranteed latency.

3. Will 5G be challenged by the current spectrum regulation and spectrum usage models? For example, some large industries will require private 5G networks with local spectrum licensing.

Dynamic spectrum access has been a topic during the last two decades, but it may get a renewed relevance with the deployment of 5G and post-5G systems, requiring fresh look on the spectrum sharing and agile spectrum usage. Get ready for new research activities in this area.

4. There are a number of concerns, especially on the social media, about the health impact of 5G. Even if there are indications that it is based on deliberate spreading of fake news, we should not forget that fake news were successful in deciding elections. Is there a real health concern andwhat should be done on a short or a long term?

This has been deemed as a very relevant issue that needs to be tackled by our community, although the current fears are unfounded. The wireless research community needs to work closely with the medical community in order to follow the radiation with the 5G deployments and provide scientific ground in order to counter the spread of misinformation and uncertainty in the public.

An example of how misinformation works is illustrated in the figures below. The news titled “Brussels Suspends 5G plan”, where the minister Celie Fremault is allegedly cited, received 1400 shares on Facebook. The actual tweets of the minister where she supports 5G deployment, but has reserves with respect to massive MIMO, which is a very specific technology that can be used in 5G, has been retweeted zero times.

In general, one of the biggest contributors to the possible failure of 5G can be the overoptimistic expecations and ambitions in defining what 5G is. There is also a lack of use cases in the area of Internet of Things (IoT), technology uncertainty/immaturity on Ultra-Reliable Low Latency Communciation (URLLC). On a positive note, the technological advances in communication technology will continue in the direction of connecting more devices, offering low latency, guaranteeing higher reliability and overall capacity in terms of data rate. As Tony Ephremides, who was in the audience, eloquently put it: 5G will fail if we continue to call it 5G.

Written by: Petar Popovski