Assessing LPWANs for IoT: NB-IoT, 4G LTE versions, 5G Redcap and LoRa WAN
Introduction – LPWANs (Low Power Wide Area Networks):
In April 2011, Cisco soothsayer Dave Evans predicted there would be 50 billion IoT devices in the world by 2020. Yet in 2025, there were only 22.3 billion IoT devices worldwide, according to the Ericsson Mobility Report. Evans forecast was off by 27.7 billion five years after the date for the 50 billion to be realized (2020). The real shocker was just how few of those devices used cellular networks for connectivity – 4.5 billion (again, Ericsson is the source). To add insult to injury, between 600 and 700 million of those cellular IoT connections – according to a chart in Ericsson’s Mobility report – used much older 2G and 3G systems. The rest were split between what Ericsson calls “massive” IoT [NB-IoT]-and a separate category of broadband and critical IoT [5G RedCap, LTE Cat-1 and Cat-1 bis standards, and LoRaWAN].
Narrow Band IoT (NB-IoT):
NB-IoT was first introduced in 3GPP Release 13 in June 2016. 3GPP Release 18 & 19 (2024–2026) provided physical layer enhancements for NB-IoT NTN, focusing on uplink capacity expansions and time-division duplex (TDD). Also, NB-IoT is included in the ITU-R M.2150 5G RIT/SRIT standard. According to the Ericsson report, NB-IoT accounted for only 1.3 billion IoT connections last year. Most are in China, where the government appears to have mandated rollout of NB-IoT to support Huawei, one of the technology’s original backers. In mid-2024, Omdia analysts, said that China was responsible for 90% of all NB-IoT connections worldwide! Regarding NB-IoT, Japan’s NTT Docomo switched off an NB-IoT network in 2020. AT&T decommissioned its NB-IoT network last year, saying it preferred LTE Cat-M.
NB-IoT still has meaningful deployments in Europe and other regions, especially in utilities and smart-metering, and roaming partnerships have improved its prospects. Omdia also reported that NB-IoT and LoRa together dominated LPWAN connections, with NB-IoT’s growth driven heavily by China but still expanding elsewhere. So the technology itself is alive; the issue is that its non-China commercial traction has been much weaker than its original promise.
The main problem was not radio performance so much as ecosystem economics. Outside of China, carriers often did not price NB-IoT aggressively enough for low-ARPU sensor use cases, and the lack of seamless roaming made global deployments harder than they needed to be. In practice, that left LoRaWAN with a freer runway in many regions, especially for private networks. In conclusion, NB-IoT has been a commercial underperformer outside China, but not a dead LPWAN. Its non-China adoption has been real but fragmented, while China became the scale engine that kept the standard commercially relevant.
4G LTE Versions for IoT:
- LTE Cat 1 / Cat 1 bis: For medium-speed IoT (10 Mbps) like telematics, e-bikes, and POS terminals.
- LTE Cat M1 (LTE-M): For low-power, mobile IoT (1 Mbps) supporting voice and firmware updates.
5G RedCap:
As with earlier forms of IoT, one of the big problems with 5G RedCap (5G reduced capability) is module pricing. Any IoT connection generates far less in revenues than any smartphone, and component prices must also be sufficiently low for connecting large volumes of objects to be economical. Even if RedCap is reserved for higher-end devices, such as the Apple Watch, the hardware for it remains too expensive, according to Omdia. In a research note emailed this week, it complains that “adoption has been limited by high module prices.” Omdia says the real constraint is the slow rollout in some countries of 5G SA, without which RedCap won’t work. Numerous mobile network operators (MNOs) have not even launched 5G SA, explaining the displeasure of Ericsson’s CEO. While take-up is especially low in Europe, it is also down at 50% in the U.S., compared with 98% in China, according to Ericsson’s Ekholm.
A new report on 6G by the NGMN Alliance, a club of prominent telcos said, “It should be noted that some MNOs may struggle to identify clear value to migrate to 5G SA, after initially deploying 5G NSA.” Omdia has a much gloomier outlook for IoT than Ericsson. The Swedish vendor currently forecasts the overall number of cellular IoT connections will grow to 7.8 billion by 2031. Omdia does not expect the figure to reach 5.9 billion until 2035, by which stage 6G networks are likely to have been in commercial operation for some five or six years.
……………………………………………………………………………………………………………………………………………..
Proprietary LPWANs Using Unlicensed Spectrum:
1. Sigfox pioneered the concept of LPWANs for IoT. It was founded in 2009 by French entrepreneurs Ludovic Le Moan and Christophe Fourtet in Labège, near Toulouse, France (an area later dubbed “IoT Valley”). While the telecom industry chased high-speed 4G LTE and WiMax, Sigfox did the exact opposite. They built a lightweight protocol optimized for extremely tiny, infrequent messages (just 12 bytes per upload). This allowed endpoints to be extraordinarily cheap and operate on a single battery for over a decade. Sigfox adopted a “top-down” model. It acted as a global network operator, partnering with local companies (Sigfox Operators) to build physical cell towers worldwide. Starting in 2019, Sigfox faced intense competition from LoRaWAN (which offered an open, decentralized model where companies could build private networks) and cellular standards like NB-IoT and LTE-M, which were backed by massive telecom operators. In 2022, the firm filed for bankruptcy protection, before its assets were acquired by Singapore’s UnaBiz. However imperfect, its business model was at least geared to IoT, which is not the case for nearly every mainstream mobile network operator (MNO). Financial reports from some major telcos have shown IoT contributing as little as 1% or 2% of total service revenues. When the returns are so minuscule, the commercial incentive remains weak.
2. LoRaWAN has been a success in its target markets, though it is not a universal mass-market wireless technology. A February 2026 LoRa Alliance report says LoRaWAN reached 125 million deployed devices globally, grew at a 25% CAGR, and is now used at scale in utilities, smart buildings, agriculture, and critical infrastructure. The evidence points to sustained adoption rather than a one-off technology spike. The LoRa Alliance says the ecosystem reached 360 members, surpassed 625 certified devices, and supports multi-million-device networks from multiple operators and vendors. Its 2024 report also cites more than 350 million end nodes and 6.9 million gateways with LoRa ICs deployed worldwide, plus use by major brands such as Starbucks, Volvo, Chevron, Chick-fil-A, and Logitech.
LoRaWAN appears strongest in low-power, long-range, intermittent-data use cases. The 2025 report says utilities remain the largest vertical, smart buildings are a leading segment, and NTN/satellite-enabled LoRaWAN is now commercially available from three service providers. That pattern suggests strong product-market fit for massive IoT, not for high-throughput consumer connectivity.
“Success” depends on the benchmark. If the bar is “won the IoT LPWAN market and built a durable ecosystem,” then yes; if the bar is “became a giant mainstream telecom platform,” then no. Independent commentary has also noted that some LoRaWAN companies have had uneven revenue growth and market acceptance even as deployments expanded.
In conclusion, LoRaWAN has been a commercial success in LPWAN/IoT, especially for private and enterprise deployments, but it remains a specialized niche rather than a broad consumer-wireless winner. The biggest adoption difference is that LoRaWAN has tended to win in private, operator-independent deployments, while NB-IoT has tended to win where carrier coverage and cellular integration matter. LoRaWAN adoption is often driven by enterprises, municipalities, and industrial users building their own networks or using community/public LoRa networks, whereas NB-IoT adoption is usually tied to mobile operators and SIM-based service models.tektelic+2
Here is a compact adoption-oriented comparison.
From Iain Morris, International Editor, Light Reading:
With 6G slowly approaching, the market is still held back by high module prices, inadequate coverage and support, and technological bewilderment, as the options continue to mushroom. Adoption is unsurprisingly not at the level that cellular IoT’s enthusiasts likely hoped for years ago. Yet the 6G story, paradoxically, seems to be all about new device types, from smart glasses to humanoids and other such physical AI.
Hence talk in Ericsson’s latest Mobility Report of a “supercycle, an unusually strong multi-year upgrade wave” that will carry smart glasses and other “connected physical AI device form factors” with it. Chipset vendors, says Ericsson, “are interested in being in the next generation from the start.”
All this, however, sounds a world apart from the humdrum IoT of smart meters and temperature sensors. The connected humanoid or AI drone must look more exciting and potentially lucrative to the average big telco hunting for sales growth in a saturated smartphone market. If telcos can extract service fees from consumers for smart glasses, they will prioritize that over anything in LPWA.
Much of the IoT probably didn’t, doesn’t and won’t ever need 4G, 5G or 6G, explaining the delta between the overall number of connections and the cellular quantity in Ericsson’s report. By 2031, Ericsson thinks we’ll have 47.1 billion total connections, including 38.8 billion that use short-range technologies such as Bluetooth, ZigBee and good old Wi-Fi. The farm of the average AgBot owner might enjoy ubiquitous 6G by then, making the vehicle even more versatile. For the far less dazzling stuff that might benefit from cellular, the risk is of ending up forgotten.
…………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………
References:
https://www.lightreading.com/iot/the-internet-of-things-was-5g-s-big-fail-6g-risks-a-repeat
LoRaWAN® Enters Its Next Growth Phase as Massive IoT Scales Globally
From LPWAN to Hybrid Networks: Satellite and NTN as Enablers of Enterprise IoT – Part 2
Enterprise IoT and the Transformation of UK Telecom Business Models – Part 1
GSA: 102 Network Operators in 52 Countries have Deployed NB-IoT and LTE-M LPWANs for IoT
5G Americas: LTE & LPWANs leading to ‘Massive Internet of Things’ + IDC’s IoT Forecast
LoRaWAN and Sigfox lead LPWANs; Interoperability via Compression
Semtech LoRa® PHY technology enables Amazon Sidewalk to expand while supporting fixed and mobile IoT endpoints
