The (peripheral) value of Bluetooth low energy

Despite a tough economic environment, Bluetooth wireless technology is expecting to do well in 2010. According to US-based analysts IC Insights, Bluetooth chip revenues could hit US$3.2 billion next year. If the forecast is realised it will mean that annual production of the low power, short range RF technology would have more than doubled compared to 2006's US$1.47 billion of shipments.

Despite this impressive success, Bluetooth misses one key market that is currently dominated by proprietary technologies: ultra low power (ULP) wireless connectivity. Though not formally defined, ULP RF is invariably regarded as wireless connectivity powered by coin cell batteries, such as a standard sized CR2032 with 3V output. These devices can withstand a peak current of 22mA, but for practical applications with reasonable battery life, an RF transceiver should draw a peak current of less than 15mA, have very short transmit times (in the microsecond range), and feature a very low power consumption sleep mode (in the nanoamp range). The combination of these operating parameters leads to an average operating current in the microamp range and coin cell battery life of many months or even years in typical applications.

Bluetooth technology is unable to run on coin cells because it wasn't designed for ULP operation. For example, Bluetooth's power consumption rises to around 35 to 45mA when using the full transmission capability between a single master and slave for a contemporary Bluetooth Version 2.0 Enhanced Data Rate (EDR) device. This drops back to 5 to 10mA when simply maintaining synchronisation and down to microamp levels when in a sleep mode.

With lack of interoperability between products from different vendors a major downside of proprietary ULP wireless connectivity, Nokia decided to develop an interoperable ULP wireless technology in 2001. Later, to encourage wide adoption, the Finnish company decided to form an open industry initiative. Consequently, in October 2006, a group of like-minded companies, including Nordic Semiconductor, formed the Wibree Alliance, in order to coordinate the development of a specification and then hardware.

Meanwhile, the Bluetooth Special Interest Group (SIG) – a not-for-profit trade association comprising 13,000 member companies including such industry heavyweights as Ericsson, Intel, Lenovo, Microsoft, Motorola, Nokia and Toshiba – faced pressure from its members for an alternative to so-called 'Classic Bluetooth' technology that was able to run on coin cells. The SIG's members were keen to extend wireless connectivity to everything from biomedical monitors, watches, toys, sports goods and thousands of other consumer products, removing inconvenient wires and connectors, and opening up entire new product categories.

Sharing a common interest, the two groups merged in June 2007. Since then Bluetooth SIG members have been working on the specification for Bluetooth low energy technology, a ULP variant of Classic Bluetooth. Version 1.0 of the standard is nearing release and is expected to be available in early 2010.

So much for the history, what designers are now interested in is what the Bluetooth low energy wireless technology specification will look like, what problems the technology will solve and when it will be available.

Proprietary ULP fills the niche

Before Bluetooth low energy was a gleam in a researcher's eye, design engineers seeking to add wireless connectivity to their products faced a bewildering choice of options. Just taking into account the technologies based on open standards, there were WiMAX (based on IEEE802.16d and e), Wi-Fi (IEEE802.11b, g and n), Classic Bluetooth technology (formerly based on IEEE802.11.15.1) and ZigBee (IEEE802.15.4). At first glance, these technologies appear to cover the entire wireless communications spectrum from long-range, high-bandwidth to short-range, low-power consumption (suitable for battery-powered portable devices). However, none of these are suitable for wireless connectivity between small personal portable products with extremely limited battery power, such as a sportswatch communicating with a heart rate belt.

The lack of such an open standard has left a lucrative niche for proprietary solutions to fill. For example, Nordic Semiconductor's nRF24xxx family of 2.4GHz transceivers has been used in millions of wireless mice, keyboards, health sensors and sportswatches. The nRF24LE1 transceiver (which consumes around 13.5mA when transmitting or receiving at 0dBm and 2Mbps) runs the company's Gazell protocol and provides a wireless mouse with a battery life of a year on two AA batteries (under normal usage) compared to a month for an equivalent Classic Bluetooth-equipped mouse.

While the lack of interoperability is not a concern for manufacturers making both ends of a peer-to-peer link (who seek to benefit from a proprietary solution's superior price/performance ratio) it does prevent use by manufacturers intending to wirelessly connect to other company's products, or those looking for a second source of transceivers. These latter groups are the target customers for Bluetooth low energy.

Extending wireless connectivity

Currently, the Classic Bluetooth chip embedded in all but the most basic mobile phones allows a handset to communicate with other devices such as PCs and headsets with ease. But the Bluetooth SIG and handset maker Nokia realised it would be useful if this communication could extend to sensors or other devices fitted with ULP wireless connectivity. How many new or enhanced standalone devices could the mobile phone support? Intelligent sportswatch with heart rate, foot pod or cycle cadence sensors, RF remote control functionality, health and wellness sensors come to mind, but the list is potentially endless.

Notably, current Bluetooth technology wouldn't be used for such connectivity because any mobile phone-based peripheral device would have to be compact and lightweight and therefore coin cell-powered. Moreover, handset makers aren't about to add yet another radio to a mobile phone that already has three or four. But if the wireless mobile phone peripherals employed Bluetooth low energy wireless technology, and handsets were equipped with suitably modified Bluetooth chips that integrated the ULP functionality into an existing Bluetooth die ” then much more becomes possible.
The Bluetooth low energy wireless technology draft specification∗ details a short-range RF communication technology featuring ULP consumption, a lightweight protocol stack and integration with Classic Bluetooth technology. According to the current estimate, the first commercial version of the interoperability specification will be available in January 2010.

Like Classic Bluetooth, the low energy variant operates in the 2.4GHz Industrial, Scientific and Medical (ISM) band. It features a physical layer bit rate of 1Mbps with a range of up to 15 metres. This may seem 'over-engineered' for sending relatively little information across a short-range wireless link, but this bandwidth has been carefully chosen because years of field experience with proprietary technology has shown that 1Mbps is the optimal tradeoff in the kind of wireless applications that Bluetooth low energy wireless technology will target. The tradeoff is between transmit power – which increases with increasing bandwidth – and duty cycle – which decreases with increasing bandwidth for a given amount of data.

The Bluetooth low energy wireless technology specification will feature two implementations, namely 'dual mode' and 'single mode'. In the dual mode implementation, Bluetooth low energy functionality is integrated into traditional Bluetooth circuitry. The resulting architecture shares much of Bluetooth technology's existing functionality and radio and results in a minimal cost increase compared to contemporary chips.

Figure 2: Bluetooth low energy wireless technology features dual mode and single mode implementations

To see a bigger version of this graphic click here.

Single mode chips will be highly integrated and compact devices. The simplified Bluetooth low energy wireless technology protocol stack features a lightweight Link Layer (LL) providing ULP idle mode operation, simple device discovery and reliable point-to-multipoint data transfer with advanced power-save and encryption functionalities. The LL provides a means to schedule Bluetooth low energy wireless technology traffic between Bluetooth transmissions. Profiles will include support for HIDs, sensors and sportswatches.

ULP consumption is critical to Bluetooth low energy's success. Single mode chips will typically operate with low duty cycles, entering ULP idle and sleep modes, to wake up periodically for a communication “burst”. In typical single mode Bluetooth low energy wireless technology operations – such as a sportswatch communicating with a heart rate monitor, for example – the chip's peak current consumption will be less than 15mA when transmitting or receiving, dropping to around 2A when in standby mode, and 900nA in sleep mode. Average power consumption will be in the microamp range, and hence single mode devices should be able to run for many months or even years on standard coin cell batteries.

Dual mode chips are targeted at handsets, multimedia computers and PCs. The dual mode specification is also advanced and it is envisaged chips will feature power consumptions of around 75 to 80 percent of Classic Bluetooth chips when operating in Bluetooth low energy wireless technology mode and cost just tens of cents more. These next generation dual mode Bluetooth chips will share much of Classic Bluetooth technology's existing functionality and radio in a single die. However, because dual mode devices will use parts of Classic Bluetooth technology's hardware, power consumption is ultimately dependant upon the Bluetooth implementation. Consequently, dual mode devices will not enjoy all of the benefits and possibilities outlined in the Bluetooth low energy wireless technology specification.

The huge applications potential

Initial applications for Bluetooth low energy wireless technology include leisure, healthcare, entertainment and office. So, for example, a person taking a workout could use their smartphone equipped with a Bluetooth dual mode chip as the centre of a Personal Area Network (PAN) comprising Bluetooth low energy-running shoes, heart rate belt and sportswatch. It's also possible that this data could be sent to a suitably equipped GPS unit that could then make predictions about where the user will be in the future based on their current rate of progress.

Figure 3: Sportswatches are a prime target for Bluetooth low energy

Alternatively, the sportswatch could communicate with a single mode Bluetooth low energy chip in a gym's rowing machine, and pass on the data to a mobile phone. Bluetooth low energy could also be used to monitor heart rate and blood pressure and then wirelessly connect to a mobile phone that could then send an SMS message to a physician. Or a runner could log heart rate, distance and speed and send it to friends' mobile phones for them to beat on their own runs. Or a winemaker could record temperature and humidity from sensors in a vineyard as he strolls around inspecting plants.

In the entertainment sector, Bluetooth low energy wireless technology could be used to replace IR remote controls. Replacing IR with RF endows a remote control with lots of new functionality. It's not hard to imagine a user browsing the web while on the move for the schedules of their favourite TV programmes. Then, with one press of a button when the user returns home, the Bluetooth dual mode chip in their mobile phone could connect wirelessly with the single mode chip in a set-top box and/or TV, and the week's viewing could be automatically programmed. If the user then wants to modify their selection, there will be no need to retrieve the traditional remote control from the back of the sofa, the user can simply make the changes by controlling the EPG directly from their mobile phone.

But these are just the obvious examples; with a dual mode Bluetooth chip in a mobile phone being able to communicate with other Classic Bluetooth-equipped devices and single mode Bluetooth low energy-equipped products, the opportunities are vast (See figure 4.) Single mode chips will also be able to talk directly to other single mode chips.

Figure 4: Bluetooth low energy extends wireless connectivity beyond the capabilities of Classic Bluetooth

To see a bigger version of this graphic click here.

A promising future

The Bluetooth low energy specification is still at the draft stage and engineers are often cynical about how long it takes standards-based specifications to be ratified and reach mass-market use. In the case of Bluetooth technology, this is said to have taken seven years. But, Bluetooth low energy is already at an advanced stage and moreover, silicon vendors are already finalising products. For example, Nordic Semiconductor plans to be the first company to release single mode Bluetooth low energy chips meeting the version 1.0 specification. The company recently announced the forthcoming release of its µBlue ('MicroBlue') products. The first product in Nordic's µBlue range will include the nRF8001 – a single mode slave chip for sportswatches, sensors and remote controls – scheduled to be generally available in early 2010 (after the official release of the version 1.0 of the Bluetooth low energy specification).

The envisaged market for Bluetooth low energy wireless technology is clear. It is targeted at those manufacturers who want to add a low cost, ULP, robust 2.4GHz wireless link to their product in order to transmit small volumes of data to another single mode chip equipped device or a central resource such as a mobile phone or PC. Because Bluetooth low energy can run from coin cell batteries, it can be integrated into thousands of low-power items to form PANs with dual mode Bluetooth chip-equipped devices.

As Nordic Semiconductor's CEO, Svenn-Tore Larsen puts it: “Once you've got a really cheap way to add an interoperable wireless link to anything that's battery powered, the potential is huge. Designers will come up with thousands of ways to use that link, especially if the information can be transmitted to a mobile phone and stored.”

∗Typical figure, check manufacturers' specifications for details of a specific device

Please note: Technical information is provisional and subject to change prior to the publication of the industry open standard.

Author Profile: Thomas Embla Bonnerud is product manager for ultra low power Wireless technologies with Nordic Semiconductor.

Related articles:
Low power Bluetooth set to be energized

Easing the challenge of RF design: Wireless Links and Antennas

Simulation of Proprietary Low Power Wireless Systems

What’s in an RF front end?

0 comments on “The (peripheral) value of Bluetooth low energy

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.