Components > Advanced Semiconductors Blog



Will 5G Benefit the Compound Semiconductor Industry?...Yes!

by Eric Higham | Feb 26, 2016

As I sit down to write this blog, it occurs to me that I haven’t talked about 5G in some time. I try to be consistent, like the post office, delivering bits of commentary on a regular basis, but I’ve missed 5G. This analogy actually fits pretty well. Who hasn’t had that important piece of mail slip the schedule for whatever reason and the issue becomes when will it arrive, what will it look like and who will be most concerned? This describes the 5G situation to a tee! I’ll give everyone a moment to catch your breath and think about how I got from the post office to 5G, but bear with me and let me explain.

The 5G “vision” (I won’t use the term standard, because one doesn’t currently exist) has the potential to be a transformative event for society and the compound semiconductor industry. The promise of 5G is “unlimited experience” with everyone getting at least 100 Mbps service and peak data rates exceeding 10 Gbps. This provides the network glue that binds everyone and everything together. The everything aspect is important, because the Internet of Things (IoT) and 5G will be symbiotic, with each enabling and enhancing the other. The final banner headline for 5G is “instant action”. This will enable the “tactile internet”, where actions undertaken remotely will instantaneously influence machine outputs.

The implications are profound. The 5G visionaries picture a world with augmented and virtual reality, intelligent cars and highways, 50+ billion sensors monitoring everything, full feature robots for surgery, assembly and the handy personal assistant! Other than looking like a sci-fi movie, this future world will eliminate traffic fatalities, reduce the global carbon footprint, normalize and improve the availability and quality of healthcare, eliminate waste of water and other precious resources and enable just about any other noble endeavor imaginable.

The stakes are high for the compound semiconductor industry, as well. I said earlier that the 5G wireless network is the glue that will hold this entire picture together by accommodating increasing data traffic that will enable all the these benefits. The 5G network is still developing, but it seems likely to include more intelligence, functionality and capability at the edge, higher frequencies with more bandwidth, MIMO and beam steering to increase spectral efficiency and densification of cells to increase capacity. All of these developments mean higher performance (and more) devices and this has historically been a good situation for compound semiconductors.

So, how is this like the post office? Well, 3GPP has meticulously conceived the schedule, but major deployment milestones link to the Winter Olympics and the World Cup in 2018 and the Summer Olympics in 2020. These global events provide a great forum to showcase new technology and the host countries want to present their best side as the world descends upon them. The challenge for the engineering community becomes how to develop new networks requiring several orders of magnitude improvement in a timely fashion. Even the schedule acknowledges this, because there is an intermediary milestone that won’t provide full 5G functionality, but will ensure something is ready for the 2018 events.

The concern is that if the schedule slips, if the mail gets lost in my analogy, the motivation to finish the task lessens. Operators, many of whom are waiting for returns on their 4G networks, are already adopting a wait-and-see attitude with the business model of 5G. A slip in the schedule or a decrease in the features will make the business model even harder to justify. At this point, no one knows what the network will look like when it arrives in 2018/2020, but it’s safe to say that if it’s too late, operators, equipment manufacturers and consumers will be disappointed.

However, I don’t want to end on a sour note so I will point out that I am unabashedly optimistic about the prospects of 5G. There are extraordinary challenges to address, but the scientific, engineering and supply chain communities have never backed down from a challenge. The vision of 5G is inspiring, but if we don’t quite get all the way there, there will still be many new opportunities and this effort will be the engine for continued growth in the compound semiconductor industry.

I’m giving a presentation at the upcoming CS International Conference entitled “Will chipmakers and foundries benefit from 5G?” I’ll get into more details about this topic at the event, but I hope that this blog inspires some discussion. If you plan to attend the conference, introduce yourself and we can chat about 5G and how Strategy Analytics can work with your company to help you address 5G or any of the other challenges that you are facing.

In the meantime, this Nokia graphic does a great job of detailing the concepts, implementations and requirements that support the 5G vision.

-Eric

Previous Post: Seeking Clarity in the Compound Semiconductor Industry | Next Post: Tales from the Compound Semiconductor Road
Leave a comment

1 comment

Leave a comment
  1. Richard Nichols | Mar 18, 2016
    5G is nothing less than a paradigm shift for the internet. Mobile internet has grown in the past as a result of demand, yes, but enabled by various patches to the basically "fixed location" internet. 5G, on the other hand, is a MOBILE internet and will be revolutionary in its projected capabilities. Higher cell network densities to be sure, but only because of the 1 msec round trip latency and high bandwidth requirements for the huge data rates targeted. If you look carefully, you'll see that 5G divides the RF spectrum into two pieces: 1) below 6 GHz; and 2) above 6 GHz. If one looks a little closer, you'll find that "above 6 GHz" is really in the 18 GHz to 30 GHz region. And above. This places a tremendously high bar for the signal processing world. Using traditional DSP just won't work at these high frequencies for a mobile network dominated by hand-held battery powered devices. Even if cost weren't an issue - which it is. You can ignore the following for blog purposes, but at Analogue Technologies LTD (ATL) we are developing an analog tunable bandpass filter to provide the frequency range coverages required at low power. This filter is frequency agile and the user can adjust the Q - all by software driven filter parameter adjustments.