Mentioned Companies: ARMH, BRCM, GOOG, KRX, NXPI, QCOM, SLAB

The Brewing IoT Standards Battle

November 27, 2015
9 min read

Kevin Ashton coined the term Internet of Things (IoT)[1] in 1999. In January 2014, Cisco’s (CSCO) then CEO, John Chambers, pegged its global financial impact at $19 trillion by 2020. Now in the role of Executive Chairman, Chambers went on to say at Fortune’s 2015 Global Forum that he expects 500B internet connected devices by 2025. For the IoT to harness and fulfill its potential, however, unified communication standards must facilitate add-on connection of devices, creating an IoT standard that allows for fluid data transmission.

It is important for a thematic investor to discern which communication standards will prevail, because they will have a strong network effect, boosting associated companies. ARK Invest predicts that Bluetooth Low Energy (BLE)[2] will be the predominant network IoT standard for small sensing devices, which will provide tailwinds to chipmakers, device makers, and more that are heavily focused on BLE.

Currently, the standardized stack of Wi-Fi,[3] BLE and national cellular networks serves as the IoT communications backbone. With time, this networking IoT standard will have to become more flexible to accommodate additional end-uses. For example, Wi-Fi is great for connecting a computer to high speed Internet, but too energy intensive for many small sensing devices. BLE is low energy, but historically has been hampered by limited range. Cellular networks are widespread, but bandwidth is limited and data rates are expensive.[4]

Hubs currently act as the glue for IoT communications, the most common of which is a smartphone, as pictured below. By definition, a hub is a central device that can control smaller low-power connected devices, can aggregate information, and can perform local processing, before relaying to the compute and storage of the cloud.


A hub must often be near the devices it’s controlling, potentially limiting the flexibility and range of IoT deployments. Mesh networks are a key high-efficiency solution to this problem. Through a process called hopping, mesh networks allow small sensing devices to pass signals to each other, ultimately relaying information back to central hubs. This digital chain reduces power consumption, because it’s easier to send a signal an arm’s length than across a football field. A standardized mesh network becomes more failure-tolerant and capable of greater range, the more sensors it has.

Currently, four mesh networks are vying to dominate the US connected home space: Z-wave, ZigBee, Thread, and BLE.

  • Z-wave has been on the market for ten years, supported by companies like General Electric, AT&T and Verizon. Sigma Designs owns the intellectual property, and although it purports interoperability through its Z-Wave alliance of 325 companies, its proprietary ownership is hindering adoption.
  • ZigBee is an open, non-profit organization, but has suffered from the forking of its technology and resultant interoperability issues. Its release of ZigBee 3.0 late in 2014 aims to solve these issues.
  • Thread, one of the most recently released standards, was developed by ARM [ARMH], Google via Nest Labs [GOOG], Samsung [KRX], Silicon Labs [SLAB], Yale Security, Freescale Semiconductor (now owned by [NXPI]) and Big Ass Fans. Thread can support a mesh network of 250 devices, is focused on certification to avoid ZigBee’s missteps, and also works on the same protocol as ZigBee (IEEE 802.15.4). Therefore, all ZigBee chips could be updated to work with Thread through a software push.
  • BLE, while not typically thought of as a mesh network, has undergone rapid innovation led by Cambridge Silicon Radio (CSR) to become mesh network capable.

As with most standard wars, the battle will be won by the solutions that achieve widest distribution as they integrate into popular existing infrastructure. For the IoT standard, key hubs need to have modems that can communicate with the chosen mesh networks, which will require broad cooperation amongst many corporations.

It will be particularly important to have tier 1 cellular modem manufacturers, like Qualcomm [QCOM] and Broadcom [BRCM], on the same page. Both believe BLE is preferable as the low energy solution. At one point, Chris Boross, President of the Thread group, argued that BLE wouldn’t work because it is not a true mesh network. Qualcomm countered by acquiring CSR and its CSRmesh technology in late 2014 for $2.5B.

CSRmesh represents a breakthrough in BLE because it can connect 65,000 devices in a single network. Given Cisco’s estimate of 25B worldwide connected devices in 2015, and with 12B of them projected to be connected via BLE by the end of 2015, ARK Invest believes roughly half of the IoT is already connected by BLE. Mesh network capabilities may push BLE well past this midpoint.


Despite Qualcomm’s bias for BLE, in 2011 it did announce the AllJoyn open source project, focused on creating an interoperable IoT stack. It passed the project over to the Linux Foundation under the umbrella header of the AllSeen Alliance in December 2013. With more than 185 major members, the AllSeen alliance projects itself as an open project with support for Z-wave, Zigbee and the protocol underlying Thread. That said, with roughly 50% share and CSRmesh’s added capability, ARK Invest believes BLE is in the driver’s seat to connect the vast majority of tiny sensing devices. The feature that may save Z-wave, ZigBee or Thread is the greater range they support from a single signal, as shown below. Greater range enables a network to work with fewer sensing devices.


While all four standards are classified as very low power, BLE has four times the throughput of Thread and ZigBee, and 25 times that of Z-wave, displayed below.

iot-standard-research-Signal-range-iot-maximum speed

Considering the many examples of prolonged IoT standard disputes, which have inhibited progress in the technology space (LTE vs WiMAX, VHS vs Betamax, Ethernet vs InfiniBand, and so on), the IoT players would be wise to avoid another painful iteration of an overplayed story. The Bluetooth Special Interest Group was put in place to avoid such disputes within BLE, and is vigilant in its aim to innovate. It has fast-tracked approval of a standardized mesh technology within the BLE protocol, which ARK believes will likely include much of CSR’s work. Given Qualcomm’s acquisition of CSR, as well as its long standing dominance in wireless communications, and vested interest in creating integrated OEM connectivity solutions, its grip will be formidable. That said, the likes of ARM Holdings, Google, NXP Semiconductor, Samsung and more have put their bet on Thread, setting the scene up for another rendition of standardization smack-down, relegating the communication layers that lack corporate firepower to IoT niches or graveyards.

You are leaving
By clicking below you acknowledge that you are navigating away from and will be connected to ARK Investment Management LLC manages both web domains. Please take note of ARK’s privacy policy, terms of use, and disclosures that may vary between sites.