Paul Gough, Principal Corporate Strategy, u-blox

As the market for trendy smartphones heads towards maturity, the next consumer “must have” appears to be a wearable device. Often designed for use with a smartphone, using its short and long-range gateway connectivity capabilities to cloud-based applications, the demand for wearables is forecast to grow significantly.

The term wearable is a broad classification that comprises smartwatches, Bluetooth headsets, wristbands, chest straps, sports watches, smart garments and head mounted displays such as those used for gaming. According to a recent report from Gartner, 274.6 million wearable devices will be sold in 2016, an 18.4% increase from 2015, and will generate revenue of $28.7 billion, of which $11.5 billion will be from smartwatches.

Of the forecast 274.6 million devices to be sold in 2016 the largest unit sales are forecast as Bluetooth headsets (128.5 m), smartwatches (50.4 m) and wristbands (35 m). While Apple has set the smartwatch benchmark with the launch of its Apple Watch, there is also an increasingly strong growth from fitness wearables such as sports watches, fitness bands, and vital signs monitors that are used by runners, cyclists and water sports enthusiasts.

Fig 1 – Forecast for Wearable Devices Worldwide, source Gartner

According to Gartner, this particular category is set to maintain its average retail price over the next years thanks to the special application-specific user interfaces used, the need for environmental durability and the continued advances of sensors and analytics. Keen to offer a broader set of functionality that competes with those of a smartphone, such as mobile payments, the wristband manufacturers are working hard to take market share away from the smartwatch sector.

These manufacturers are also keen to develop the premium paid-for cloud-based services that analyze the data generated by the device. Another application of the fitness band and chest strap sectors is those devices used by wellness programs. Initially driven by health initiatives established in the United States, the positive link between an individual’s activity levels and general health is gaining popularity with health professionals around the world. Many of these programmes pay the individual for maintaining a regular exercise regime as opposed to the future high costs of providing health care resulting from lack of exercise.

One of the key requirements for any wearable device is that of connectivity. The popular methods include Wi-Fi, Bluetooth, ZigBee and cellular, each having their merits. When faced with developing, for example, a new fitness band the engineer needs to think about how much data will need to be transferred, how frequently and over what range it would typically need to be sent.

Pretty much for every application there will be a trade off between range, data rate and use case to be considered. Use case questions such as, will the fitness band communicate to a smartphone that then collects data and forwards it to the cloud, and will the smartphone application perform local analysis of the data or will that be done in the cloud, all have an impact.

Wearable devices will always be battery powered, so this will also influence which connectivity method is used. Bluetooth Low Energy (BLE) is designed for low power requirement and is ideal for sending relatively small amounts of data. Virtually any smartphone can support this method of communication. However, if a higher quantity of data, say a few Mbytes needs to be transmitted then the designer might best consider using Bluetooth Classic or Wi-Fi. Then the consideration of range needs to be taken into account. BLE typically can communicate over 30 meters in line of sight.

For the fitness band example it will be assumed that the wearer will also have their smartphone with them so distance is not an issue. However, some wearable applications will dictate the use of cellular communication since independence and reliance on other communication methods are needed.

An example might be the use of tracking bands used for workers in isolated locations, so called lone-workers. Other such examples include child and pet trackers that give locational data in near real time. An example of a module that provides cellular connectivity is the SARA-U2 from u-blox. This miniature LGA-sized package measures just 16 x 26 x 3 mm and weighs under 3 g making it ideal of use in space constrained wearable designs. It offers high speed 5.76 Mb/s (HSUPA) and 7.2 Mb/s HSDPA cellular data rates but still manages to have a low idle mode current consumption down to 0.9 mA

Figure 2 – SARA-U2 module from u-blox provides cellular connectivity in space constrained wearable designs

In addition to connectivity many wearable devices also need to track and record the wearer’s location. Sports performance monitors and cycling watches use this to overlay the wearer’s heart rate to the actual latitude, longitude and elevation. Incorporating positional capabilities into the design can be achieved in one of two ways.

The most obvious one is through the use of a global navigation satellite system (GNSS) receiver which naturally, for any wearable application, needs to be the smallest and most power efficient means possible. The EVA-M8M is an example of a 43pin LGA packaged GNSS module that would suit use in any wearable design. Measuring 7 x 7 x 1.1 mm this surface mount device weighs just 0.13 g and only consumes up to 25 mA in full continuous operation but down to as low as 5.5 mA in the power save mode, where the GNSS data is updated every second.

Fig 3: Block diagram of the EVA-M8M highlighting the comprehensive capabilities squeezed into such a small package.

Like any GNSS system achieving a reliable position “fix” is reliant on the antenna being able to “see” the satellite. Achieving this indoors or where the satellite signals have been reflected by large buildings, such as in dense city centres or in any area of marginal signal conditions represents a major challenge.

Some wearable devices might need reliable indoor reception more than others, for example, the lone worker application mentioned earlier. Should this be a design requirement, and the primary communications method is through cellular means then a mobile network-based positioning approach can complement the GNSS data. By maintaining a database of the positions of cellular network towers a cellular service, such as that of CellLocate from u-blox – Figure 4 – can estimate a coarse location of the device based on previous observations from other CellLocate-enabled modules.

Figure 4 – Block diagram of mobile network positioning technology

In meeting a wearable design challenge what are the key steps an engineer needs to review before commencing development? One of the first steps in this process will be a through understanding of the device requirements. What will the wearable application monitor, what sensors need to be incorporated and what are the potential use cases? Reviewing the use cases is a crucial aspect since it will highlight the key factors such as product size, available space envelope and the duty cycle expectations.

These will shape the space available for a battery, and the direct impact on the battery capacity that will of course have a direct correlation to the time between charge cycles and operational duty cycle. The various use case scenarios will also identify the type of communications required. Does it need to use a smartphone as a gateway to a cloud application or will it communicate directly using its own cellular data connection. An increasingly important consideration for many Internet of Things (IoT) devices is the ability for the device firmware to be updated over the air (OTA) rather than requiring user intervention to download new device images to a PC from the manufacturers site and upload to the wearable device. The specification of the host processor and the amount of memory needed to achieve OTA might need careful review should this be the case.

Most wearable devices are also likely to experience the same environmental influences as the wearer. Rain, moisture, dust, and wide temperature variances all need to be taken into consideration in how the product enclosure is designed. Will it require an ingress protection (IP) rating in order to satisfy the marketing specification together with balancing these factors with the experience for the user.

Ending up with a wearable device that the user finds uncomfortable to wear due to its size, weight and shape are all critical factors for the future success of the product. The engineer needs to look not only at the electrical specifications of the components selected but their physical attributes too. As highlighted above the SARA-U2 cellular module and the EVA-M8M GNSS module together weigh just 3.13 grams making their combination ideal for any wearable design.

The marketing requirements analysis of any product will also estimate the anticipated volumes possible. These will shape a number of production decisions and greatly influence the overall BOM goal. From the connectivity aspects this might prompt many engineers to review whether a discrete design is better than using a module. The difference of cost against price is a hard-learnt lesson. The BOM for a discrete approach might be slightly cheaper but factor in the test and certification costs then there is little difference. Being able to put a module that is pre-certified to most worldwide wireless regulatory bodies on a PCB not knowing where the device might be sold is a huge timesaver. Also, RF design is a specialist skill and requires equally specialist test equipment and facilitates. Having to spend many weeks and potential PCB redesign due to encountering EMI problems resulting from a poor track layout would negate all the cost benefits of a discrete design.

Making the available power budget last as long as possible is a skill that many embedded engineers know through trial and error. When selecting connectivity and GNSS modules for use in a wearable design engineers need to carefully review the module’s technical documentation for the methods that can both keep power consumption to a minimum while not impacting the device’s responsiveness, particularly that concerned with user interaction.

Most microcontrollers and modules available today will offer several different power saving modes and the engineers needs to diligently review these to find the scheme best suited to the application needs. Typically such modes will selectively turn on or off parts of the module’s functions. For example, the u-blox EVA-M8M uses a power save mode to reduce system power consumption by turning parts of the GNSS receiver on and off. This process is illustrated in the state diagram shown in Figure 5.

Figure 5 – Power saving mode state diagram of u-blox EVA-M8M GNSS module

The power saving mode is based around five different states defined as inactive (awaiting next fix and next search), acquisition, tracking and power optimized tracking. The power consumption profiles differ with each state with the acquisition consuming most power down to inactive where most parts of the receiver are switch off. By taking full advantage of these power saving modes in the GNSS module, and similar methods available within the connectivity module, will considerably enhance the battery life of the wearable device and the user experience.

Ultimately all these factors will determine the product’s commercial success and serve to reinforce the manufacturer’s brand reputation.

About the author:

Paul has over 25 years of experience in managing research, developing new concepts and commercializing newtechnologies and applications. His areas of focus include leading edge technologies, GPS, Geotagging and Semiconductors.

Paul was an early pioneer of wearable technology and believes in the concept of “marrying” electronics and clothes!

Receive latest electronics news and products in your inbox
Weekly alerts and exclusive articles, products and free webinars
Your email will not be shared with anyone else