Five reasons why the Internet of Things needs its own network

【51CTO.com Quick Translation】Last week, AT&T said it would launch a narrowband Internet of Things (NB-IoT) network in the United States and Mexico. This is not the first network AT&T has focused on deploying specifically for the Internet of Things. The operator previously announced the launch of an IoT network using the LTE-M standard, aiming to cover about 400 million people in the United States and Mexico by the end of last year.

Just as important, a number of other US carriers are also working on various versions of low-power IoT networks, including Verizon, T-Mobile, Sprint and even Dish Network.

[[234492]]

So what’s attracting these operators? Why do they feel it’s necessary (or at least beneficial) to build entirely new wireless networks to support low-power machine-to-machine (M2M) communications? After all, don’t existing LTE, 4G, and planned 5G mobile networks provide the connectivity needed?

Yes, but there are other reasons.

Why do low-power IoT networks make sense?

It turns out that there are five reasons why these low-power IoT networks make sense in addition to existing and planned mobile networks.

Reason 1: Cost

As always, cost is the key concern. Despite being built from scratch, these new IoT networks can offer much cheaper service. For example, T-Mobile offers a $6 per year plan for machines on its new NB-IoT network. The company claims that’s only one-tenth the cost of a similar plan from Verizon, but even $60 per year is much cheaper than a standard cellular connection.

Just as important, the low-power devices that use these networks are much cheaper than standard LTE devices like cell phones. As AT&T said in a press release last year: "Now we can reach new areas and connect new devices more affordably than before."

The second reason: power consumption

Power consumption is another factor. Standard LTE connections are relatively power-hungry, which can be a problem for devices that are expected to last for years on a small battery. If a device is connected to a network with extremely limited bandwidth (NB-IoT drops data rates to 120 Kbps or less), it can last much longer than if it were connected to a full-power network.

Reason 3: Spectrum

Efficient use of scarce, expensive wireless spectrum is the third reason why dedicated IoT networks make sense. Both NB-IoT and LTE-M can be deployed in very small sliced ​​spectrum channels compared to 4G deployments. NB-IoT can even be deployed in so-called “guard bands” of LTE spectrum between LTE channels to prevent interference. This means that NB-IoT communications do not share spectrum resources with standard smartphone traffic.

Reason 4: Coverage

Obviously, these low-power networks could provide better signal coverage for IoT devices deployed deep underground or inside buildings. Low-power devices running on these networks could also be much smaller than standard LTE devices, so they could be used in more places. Some estimates suggest that NB-IoT device modules could be only one-sixth the size of current modules.

Reason No. 5: Aging 2G Technology

Building new NB-IoT and LTE-M networks can also help operators migrate IoT customers off aging 2G cellular networks, which many operators are eager to decommission.

2. One major argument against dedicated IoT networks

So far, so good. But there is a degree of risk in deploying these low-power narrowband IoT networks. Simply put, the demand for bandwidth has only ever been going in one direction: higher and higher. Regardless of the use case originally envisioned, people want to do more and more with the available networks. It seems likely that these new networks will soon be limited by the very factors that made them so attractive in the first place.

For example, NB-IoT is much cheaper than LTE-M, but it can't do all the same things. LTE-M has higher bandwidth, can support two-way communication (helpful for updates and security patches, for example), and supports speech, which NB-IoT can't. Basically, the two standards take slightly different approaches to the same goal, which is whether a cheaper, lower-capacity solution can provide enough flexibility to meet not only current IoT use cases, but also use cases that haven't been thought of yet.

Original title: 5 reasons the IoT needs its own networks, author: Fredric Paul

[Translated by 51CTO. Please indicate the original translator and source as 51CTO.com when reprinting on partner sites]