Six months ago, I was skeptical about millimeter-wave (mmWave) adoption in smartphones, and I debated with my friends at Qualcomm about whether it would ramp up. During the discussion, I told Qualcomm that I would “eat humble pie” if major consumer adoption happened in 2020.
Now, I’m sitting down to the table and getting ready to take a bite. I’ve seen some encouraging changes in the ecosystem that make me much more optimistic now. Here’s a rundown of the important factors:
On the negative side:
· The technical challenges are still there. Performance of mmWave smartphones is still fairly poor (low power in the uplink), causing challenges in actual usage. The problem here is that 5G NR in the mmWave bands needs channel-estimation data to be transmitted in the mmWave uplink, not in another band. In other words, this system needs to be a closed loop and the uplink RF front end is too weak today.
· Propagation is still poor, and penetration for indoor users will be a challenge. In some cases, a window or a single wall can be penetrated with only about 20-30 dB attenuation. However, attenuation of 30 to 90 dB will be commonplace, even for light construction. Forget about penetration of concrete buildings with metalized glass.
· Millimeter-wave components are expensive, adding about 65% to the cost of the RF components in a handset. A typical mmWave implementation involves three sub-arrays, each of which includes four dual-polarized antennas, up/downconverters, and wideband processing.
· We still have no ‘killer apps’ that drive consumers to buy new phones for gigabit performance.
On the positive side:
· Repeaters, CPEs, and indoor gNodeB infrastructure are coming to market to improve performance for indoor users. Options from companies like Pivotalcomm and Movandi are FCC-certified and have been thoroughly tested by Verizon, and are starting to be deployed in the market.
· Verizon is aggressively pushing their users toward mmWave phones. In particular, Verizon has refused to offer any 5G phones, tablets, PCs, or similar devices that are not mmWave-capable. They have announced plans for at least 20 devices by year-end.
· Verizon appears to be subsidizing the extra cost of the mmWave radio. They’re offering 5G mmWave handsets for prices similar to other non-mmWave handsets. They’re doing this because fat pipes can deliver data at lower cost than skinny pipes.
· It turns out that the “killer app” for 5G mmWave is not related to virtual reality or any other new use case. Rather, it’s simply the need for mobile operators to increase their capacity by a factor of five. So, it’s good old-fashioned video streaming that is driving the need for 5G mmWave.
So far, all of the above is based on the announcements of a single company: Verizon. But we believe that Verizon is spearheading a new trend in which the operators will demand mmWave in the network, and they will bear the cost.
After 30 years of toiling as a ‘niche’ market, mmWave is about to hit the mainstream.