Evolving Private IoT Networks: LoRa and CBRS

Public Versus Private Networks

Before we explore LoRa, CBRS, and how private LTE IoT networks can benefit solutions providers, let’s make sure we are clear on what a public network is. We are all somewhat familiar with the difference between public and private networks on an everyday user level.

An example of a public network is the Internet your computer or phone is using right now to access IoT For All. The public Internet enables internetworking a global array of subnetworks, public and private, using the standard  TCP/IP protocol suite to route data through the various subnets that comprise it, which, in turn, route traffic to and from their respective subnets, etc.

On the other hand, a simple example of a private network is a Virtual Private Network (VPN). Many of us are working remotely due to COVID-19, and our companies need internal data and access to remain as secure as if we were all in the same in-building secure LAN. Think of a VPN as a private network within a public one—generally, the Internet.

For argument’s sake, let’s say you are using a VPN right now. Your connection is routed through a VPN server which ensures anyone who sniffs your traffic would see only encrypted data. Through a VPN connection, your phone or computer is able to access the Internet and look up our domain name—iotforall.com—on a public Domain Name Server (DNS). The matching IP address the DNS sends back to you tells your browser how to reach the server hosting this website. And here you are.

Private IoT Networks

In the context of the Internet of Things (IoT), where our objective is to internetwork things beyond mere phones and computers, using public networks alone as the backbone/core of a thing-network would result in a number of complications.

For many IoT solutions providers, a private network is often used to network devices together before backhauling over a more public network—or, in mission-critical or sensitive Applications, a public network could be skipped entirely. Private networks can also be physically independent of customer public networks such as WiFi to allow segmentation of IoT traffic, more granular access and control, and enhanced security.

Here is an exemplary private LTE architecture diagram from Qualcomm: 

Image credit: Qualcomm

What are the Key Considerations When Choosing Public and Private Network Components?

When figuring out which network components should be public, private, and/or which components can operate in a licensed spectrum versus unlicensed spectrum, you should think about a few key factors. 

1. What is the per-device network usage cost? If your devices are talking straight to a commercial cellular network, you have to think about your carrier’s SIM and usage fees, which at scale can be prohibitive. 
2. What is the cellular coverage like in my area? If you are operating in a remote area, like an agricultural zone, or a resource extraction site with underground and overground buildings, are you able to reach public carrier infrastructure? Potentially not. 
3. Am I concerned about having a relatively public network backbone acting as the middleman between my devices and my solution architecture? If data privacy, security, regulation, or ownership is a primary concern, using a public network backbone may well be an issue. Say you are designing an IoT solution for a smart hospital, for example, over which vital patient data will be transmitted. Is a public network even an option? 
4. Is your solution mission-critical, making the added upfront cost of deploying a private network infrastructure a necessary one? Taking on the responsibility of managing your own network’s uptime is a double-edged sword. 
5. What is the RF environment like? If your new IoT solution is a smart hospital, and you’re considering building a LoRa-based campus/corporate area network (CAN) that will occupy the same RF environment as a university CAN that makes heavy use of the ISM band frequencies, you may want to think carefully about whether your solution can handle the potential competition using unlicensed bands engenders. You may need to license some spectrum locally so that you can rely on it being fully available to your private network. 
6. What kinds of physical obstacles will my IoT network need to overcome? If you are combining indoor and outdoor asset tracking, be sure to investigate how device connectivity will hold up as it encounters obstacles across the indoor/outdoor—and potentially also underground/overground—boundaries.

What Kind of Private Network is LoRa/LoRaWAN?

If your IoT solution has tight power constraints, a low data rate, is not mission-critical, and/or is operating in an area where having your devices interface directly with cellular or hardline public networks is not feasible, building a private IoT network that uses unlicensed ISM bands may be the best option.

The de facto choice for many Low-Powered Wide-Area Networks (LPWAN) implementers is LoRa. LoRa chipset technology, and the popular LoRaWAN open standard developed on top of it, has for many years now been enabling a robust and growing IoT ecosystem. By 2023, experts predict LoRaWAN, along with its roughly similar cellular alternative, NB-IoT, could represent more than 80 percent of the LPWAN market. 

Image credit: Yitaek Hwang (original source: Ubidots)

The Changing Landscape of Private IoT Networks

Until quite recently, LoRa-based solutions seemed like pretty much the only option for private LPWAN IoT networks. With the recent commercialization of much of the Citizens Broadband Radio Service (CBRS) in the US, private LTE is an increasingly viable backbone for IoT networks. CBRS is opening the field. To build a private LTE network up until now, you would either have had to go through a lot of hassle to get a Distributed Antenna System (DAS) that also locks you into a specific cell carrier, or use MulteFire’s unlicensed 5GHz LTE offering within a relatively small ecosystem. 

CBRS promises to solve a number of issues both for cellular consumers, who often experience only spotty coverage in buildings and also for in-building and indoor/outdoor-based IoT solutions providers, who were stuck between a rock and a hard place. They may be tempted to use a pre-existing WiFi backbone for their CAN, which has numerous security, range, consistency, and reliability drawbacks from the perspective of an IoT system.

Or, on the other hand, developers may try to stitch together BLE, LoRa, WiFi, and/or cellular into an ad-hoc assemblage that tries to cover all the blindspots and achieve campus-wide network coverage. 

The Citizens Broadband Radio Service (CBRS) Promises New IoT Solutions and Strategies

CBRS is a section of the radio spectrum between 3550 MHz and 3700 MHz (“3.5 GHz”), which until quite recently was reserved by DOD incumbents, primarily for seaborne and coastal radar systems. In an admirable partnership between the FCC, military, and industrial/commercial entities, this valuable spectrum is being unlocked, representing a 16 percent expansion in the commercially available sub-4GHz spectrum.

The CBRS band has been dubbed the “innovation band” by the FCC and others. CBRS Alliance, which oversees the rising OnGo CBRS standard, estimates the commercialization of CBRS will generate tens of billions in new economic value.

CBRS spectrum is exposed to users in a three-tiered licensure system that enables enterprises and agencies to stand up private LTE networks in the 3.5GHz spectrum, provided their network can ensure—via Environmental Sensing Capability (ESC)—that it will not interfere with incumbent (e.g. military) operations in the CBRS band. Spectrum sharing operates on the innovative Spectrum Access System (SAS) model, which automatically coordinates frequency access on the following tiered scheme:

Image credit: (Counterpoint Research)

Although LoRa-based solutions operating in the sub-GHz ISM band will remain vital to the LPWAN market, and the IoT industry more generally, CBRS will enable new IoT strategies and unlock higher bandwidth Applications. For example, as Brian Ray (former CTO of Link Labs) argues, “For high-reliability Applications or those in environments already plagued by complex IT systems and crowded unlicensed band usage, like hospitals, CBRS could open up a new capability to install IoT solutions.” 

One reason for the excitement around CBRS opening up is this: cell carriers used to dominate a lot of the good spectrum, excluding cable and tech companies from competing. Now, with neutral-host LTE becoming viable, cable and tech companies such as Comcast, Cox, Charter, Google, etc., can build even bigger monopolies. When cell carriers are no longer the sole gatekeepers of so much of the “sweet spot” of commercially available spectrum, giants like Google and Comcast will be able to reach end customers more directly. At least in the short term, until new monopolies form, this could benefit consumers.

 Neutral host connectivity could enable that hypothetical smart hospital to avoid paying per-device cellular carrier fees if the hospital provided its own connectivity components. The 4G/5G devices our smart hospital deploys could see three-to-four times the range it would have seen if attempting to use the hospital’s WiFi—without any of the headaches of managing WiFi security since that problem is now solved with SIM cards.

The smart hospital solution operators would not have to concern themselves with managing the low-level network issues they would have had to handle with IoT-specific network infrastructures like LoRa or Zigbee. They would have offloaded many of those concerns onto the cellular host currently backhauling their network.

Perhaps most importantly, since in-building consumer cellular rather than IoT is the primary driver behind CBRS commercialization, general in-building connectivity could improve dramatically, with smart applications piggybacking on top of that new CBRS backbone.

The Shared Future of IoT and CBRS-Enabled Smartphones

Commercial CBRS is only starting to become viable. And although it could be several years before deploying IoT solutions on top of CBRS infrastructure becomes frictionless, all signs are pointing to the band being a major enabler of IoT adoption. As Brian Ray concludes, the likeliest scenario in which IoT systems will leverage CBRS is not the deployment of dedicated CBRS networks for an IoT Applications, but rather IoT systems being layered on top of CBRS-enabled private LTE networks that get installed in buildings to amplify in-building and campus-wide cellular access at the enterprise level. 

Google, Apple, Sony, Commscope, and Federated Wireless, to name a few, were recently SAS-certified, meaning they can start building devices—e.g. smartphone modules—that can use the CBRS band. The commercialization of CBRS rests on the innovative idea of sharing spectrum. It is perhaps only fitting that CBRS-enabled IoT solutions share the stage with (currently) more dominant devices: smartphones. Expect to see IoT solutions thrive in the wake of mobile device drivers. This new, robust kind of shared private network will hopefully propel the IoT sector forward.