If you are a proximity solutions provider considering whether or not to create your own beacons, understanding the different chipsets you can use is essential. That being said, those wanting only to write an app that uses beacon APIs don’t really need to understand anything about the chipset inside the Bluetooth beacon. Therefore, this chapter is targeted at people who are selecting which type of beacon to deploy, or who want to convey to their colleagues the sense that they understand the beacosystem. For them, it will be a good idea to learn about the chipsets that are the engines driving this ecosystem.

The choice of chipset in a beacon plays a role in determining important characteristics, such as power consumption and signal strength. Understanding whether the beacon vendor has used a generic module or if they have invested in adding value with enhanced features, such as a more robust antenna, can be critical. It is also often worth noting where the beacon vendor has chosen to invest their limited capital, to help you consider their viability as a long-term supplier.

Competitively priced, highly functional chipsets have enabled small startups to develop their own beacons at relatively low cost. These Bluetooth Smart System on Chips (SoCs) can cost around $2.50, if you buy them in volumes of 1,000 or more. If you’ve noticed how many companies have created their own brand of beacons, it’s because the chipset vendors made it cheap and surprisingly easy to do so.

In the past, a beacon manufacturer had to assemble and integrate a range of chips that each performed discrete functions: a CPU, random access memory (RAM), read only memory (ROM) , flash memory, a signal processor, interface controllers, and a radio transceiver. This is a costly, time-consuming process with an element of risk for each additional integration step. Now these components are integrated into a single SoC, which forms the main building block for most beacons.

The SoC is integrated onto a small printed circuit board (PCB) the size of a coin, with an antenna, additional sensors, and any buttons or LED lights required to create a Bluetooth beacon module. Some beacon manufacturers create their own modules when their volumes or their unique requirements justify the expense. As the beacon volumes scale from 10,000 to 20,000 units, it may become economical to produce your own module. Beacon manufacturers that create their own modules will incur overheads ranging from additional design complexity and risk, testing overhead, calibrating/adjusting the output of the signal baseband chip, certifications, and managing a more complex supply chain. Hence, only those that will be producing enough bulk should consider creating their own module, especially considering how inexpensive generic ones are.

Chip manufacturers, such as Nordic Semiconductor, provide either the SoC or a version of the SoC packaged in a module. Their nRF51822 module/kit retails for about $30 for quantities of one. Chinese manufacturers offer their own modules based on the same SoC for around $5. In other cases, such as with Broadcom, the SoC manufacturer will partner with a third-party company that will design, manufacture, and sell an approved module based on the SoC.

A basic beacon is likely to use a small (2 cm) antenna that is printed on the PCB of the module. Some beacons, such as the Gimbal S20, have two larger external antennae. One is directional and the other is omnidirectional. The quality of the antenna can enhance the range of the beacon as well as control the shape of the signal that is transmitted.

Once a module has been sourced, adding a battery and a case is required to produce a basic beacon.

Many of the beacon chipset vendors provide an example software stack that includes drivers that will enable the beacon to transmit iBeacon or Eddystone conformant radio packets or frames. The beacon OEM still needs to join Apple’s MFi (Made for iPhone) program in order to be able to claim the beacon is iBeacon compliant.

We won’t enumerate the model number of every chip produced for the Bluetooth beacon market. (You’re welcome.) These product names are typically made up of a concatenation of the initials of the manufacturer followed by digits that give a clue to the logical relationship of the product relative to its ancestors in that product family. Listing them all would render this book obsolete the week it was published and make its pages resemble a phone book. It is, however, useful to understand the lineage of a few of the dominant players and why they have been successful.

Nordic Semiconductor was the first of the Bluetooth chip manufacturers to focus on the beacon market. Their Bluetooth Smart nRF51822 offering was particularly attractive to the burgeoning market of small beacon OEMs. Nordic’s approach was different than the larger chip vendors such as Broadcom, which focused on supporting a small number of customers who could commit to purchasing very large numbers of chips, like major PC or phone vendors. These bigger vendors provided their VIP clients with personal service and support via a limited number of systems engineers.

Nordic , on the other hand, provided their support via an online portal that could service the needs of an unlimited number of smaller developers. They also provided modules that could be easily leveraged with a minimum of additional work, as well as example apps and the software drivers required to support the iBeacon standard. Their low-cost developer kits included a module along with connectors and a power supply—a complete beacon—but it lacked a case and the cloud services to manage it. Nordic was enabling “the long tail”. As a result, their nRF51822 product gained the largest market share in the early stages of the beacosystem. Among their customers were companies such as Estimote and Kontoct.io, small companies that rapidly grew to be among the leaders in this emerging market.

More recently, the other chip vendors, including Broadcom, have moved to adopt the same support strategy with its Wireless Internet Connectivity for Embedded Devices for Bluetooth SMART (WICED SMART ) . It’s argued that they still have a legacy of restricted access to WICED and a way to go before addressing large numbers of smaller developers.

In this initial phase of the development of the market, Texas Instruments (TI) achieved second place in market share. Its product was used in Gimbal beacons such as the S10, which was produced in very large numbers and was one of the lowest priced products ($5) on the market. Gimbal’s S20, a higher-end product, was used by Apple in its retail stores. This win by TI was not publicized but became generally known in the industry. Qualcomm , which owned Gimbal at the time, competed with TI and produced its own Bluetooth chips via its QCA (Atheros) subsidiary. These chips, however, were Bluetooth Classic and Dual mode chips, unsuited for use in beacons, and didn’t require the Bluetooth Classic protocol stack. It would have been embarrassing for Qualcomm to have one of its subsidiaries selling beacons with a competitor’s Bluetooth chip inside.

This is a great example of why beacon vendors are reluctant to disclose whose chip they may be using in their beacons. The chipset vendors themselves are typically bound by non-disclosure agreements and so are unable to confirm the details of who their customers are. This can be a convenient but legitimate excuse when they have very few wins in the market. The truth can typically be unearthed by speaking to competing chip vendors, who may not be keen to disclose where they lost a deal, but are not typically barred by an NDA from talking about business they didn’t win.

Another source of information about which beacon vendor is using a particular chip are the “tear down” analyses published on blogs such as www.aislelabs.com/reports/beacon-guide/ , where new beacon models are dismantled and the chipset is listed.

Third place went to Cambridge Scientific Research (CSR) .

In a push to grow revenue and strengthen its position in the Internet of Things market, Qualcomm acquired CSR in 2015. Prior to the acquisition, CSR had been exploring a strategy of marketing their own beacons to pull through a cloud service offering and so had not been as aggressive in selling to the beacon vendors. With the acquisition by Qualcomm, that strategy has been dropped and they are competing for business as a provider of SoCs for beacon vendors.

Other vendors are now aiming to displace Nordic by focusing on value-added features such as lower power consumption (Dialog) and mesh networking (CSR), which provides an alternative approach to management of beacons.

Silicon Labs offerings to beacon manufacturers came from its acquisition of Bluegiga, a smaller Scandinavian company that brought its Blue Gecko BGM111 Bluetooth® Smart Module to CSR’s portfolio. Bluegiga used TI silicon but with its own custom stack (which in several cases was reported to be more reliable than TI’s). Power consumption of this product was higher than Nordic’s product, but newer generations became competitive.

Dialog was the first Bluetooth Smart vendor to reduce their transistor size to 55 nanometers (55/1,000,000,000 meter). This has enabled them to reduce their power consumption, which in turn extends the battery life and makes power harvesting viable. When a beacon’s appetite is modest, its power can be harvested from interior lighting and sunlight using photovoltaic cells, from heat, kinetic movement, and even from the magnetic radio signals of other sources.

Table 6-1 lists some of the more notable Bluetooth Beacon chip vendors .

This section discusses a few attributes to consider as you evaluate the choice of chipset for a beacon.

The platform provided to the makers of beacons by manufacturers of Bluetooth semiconductors is impressive. As solution designers, we are standing on the shoulders of giants. These giants have spent years developing very functional building blocks for us to use at a remarkably low cost. But there are still tradeoffs and decisions to be made and they will drive the choices of silicon vendor and product. While the details may be technical, the decisions will be driven by business strategy and priorities—the geographies where the beacon will be sold, the features where differentiation will be required, and the performance that is expected.

Having studied what’s inside the beacon, the next logical step is to look at the beacons themselves.