 |
|
Joe Madden |
People are talking about the Internet of Things as the next major thrust for mobile infrastructure. Hold on a minute: How big will mobile IoT really get? We will see 50 billion connected devices…but will 48 billion of them be connected by Wi-Fi or some other unlicensed technology?
Issue #1: Most devices don't need to be mobile.
Thermostats, refrigerators and office machines can access the Internet through your Wi-Fi router. There is some talk about using LTE technology or 5G technology to improve quality for home appliances, but it's clear that unlicensed technologies will be much cheaper.
Issue #2: Other wide-area technologies are deploying now.
SIGFOX is one example of a low-power wide area (LPWA) technology which is currently available in 12 countries. It uses unlicensed spectrum at 900 MHz and an ultra-low bandwidth airlink to achieve great coverage and in-building penetration. Other proprietary systems such as LoRa and Huawei/Neul narrowband IoT standards are also ready to ramp up. While 3GPP committees are still working on narrowband LTE standards and potential 5G IoT standards, other options are already in the field.
Issue #3: Mobile network OEMs won't like the pricing.
The LPWA standards have set very low pricing on Massive MTC applications. Want a sensor in your bicycle? The price point is less than $1 per year today. Want to put sensors in the soil on your farmland? Less than $1 per year. These applications could possibly be supported from the existing 2G radio heads in the field with a software upgrade to make your GSM radio work for narrowband LTE. Simple math: If 2 billion devices are serviced at $1 per year, that's $2B in service revenue. It doesn't compare to the $1.2 trillion of service revenue in the current mobile telecom market. Even if operators spend 25 percent of revenue on the software upgrades to the base stations, OEMs will only get $500M per year. It's not big enough to move the needle.
Issue #4: The royalty pool is too expensive.
In IEEE standards and proprietary systems, the cost to implement the silicon and the stack is pretty low. Why is an LTE baseband chip more than double the cost of a Wi-Fi chip? It's not twice as much silicon. It's the royalty pool and the extra algorithms that make LTE better -- all that R&D gets paid for through royalties. If we want cheap sensors in our tennis shoes, we can't give half of the chip revenue to a royalty pool.
Mobile Experts has published some case studies on 5G IoT, and will be publishing an LPWA study soon. As we break the market into segments, we're left with a pretty interesting opportunity for high quality, high reliability communications where the robustness of LTE will have real value. These devices won't be connected for $1 per year. They're more likely to reach premium pricing in the range of $100 per year or more, depending on how much data is actually required.
Maybe security issues, interference, quality issues and reliability problems will drive the mainstream market away from unlicensed methods and toward licensed technology. I'm not going to say that 5G IoT is a dead duck. But it's a duck that is still paddling on the water, while the other birds are flying.
Joe Madden is Principal Analyst at Mobile Experts LLC, a network of market and technology experts that analyze wireless markets. The team provides detailed research on Small Cell, Base Station, Carrier Wi-Fi, and IoT markets. Mr. Madden currently focuses on trends in 5G, IoT, and Enterprise markets for wireless infrastructure. Over 26 years in mobile communications, he accurately predicted the rise of Digital Predistortion, Remote Radio Heads, Small Cells, and the rise of a Mobile IT market. He validates his ideas with mobile and cable operators, as well as semiconductor suppliers to find the match between business models and technology. Mr. Madden holds a Physics degree from UCLA. Despite learning about economics at Stanford, he still obeys the laws of physics.