‘Smart integration’ design

‘Smart integration’ design

Combining wireless functions tames design problems

BY MATTHEW PHILIPS
CSR, Cambridge, U.K.
http://www.csr.com

The number of technologies that could go into a mobile phone is huge and growing all of the time from short-range wireless technologies to almost ubiquitous functions such as cameras and Internet capability. These technologies are changing not only how people use their phones, but the fundamental ways in which handsets must be designed.

The challenge that designers now face is which of these technologies to select to offer a handset that is suitably differentiated. It is also crucial that the technologies selected do not have an unwanted impact on the key design elements, particularly on the crucial constrains of portable design; space, cost, and power.

A range of technologies

Standard features in mobile handsets include Bluetooth, music players, and to a lesser extent, Wi-Fi. Bluetooth is already embedded in over 50% of mobile phones and this is expected to reach 55% or 60% by the end of 2008. The combination of MP3 music players and phones is a logical one and you don’t have to spend very long on public transportation to witness the growing number of people now using their phones in this way.

Also technologies such as FM radio with radio data services (RDS) functions, or GPS for navigation and the added benefits of location-based services (LBS) are attractive functions that are set to find their way into a large number of handsets. Other short-range wireless technologies are also lining up to nudge their way into the tightly packed mobile phone. Wireless standards such as NFC, Bluetooth low energy, and UWB build on the success of the adoption of Bluetooth, but take wireless connectivity to new levels and open a wide range of new applications.

Consumers are hungry for these technologies to be integrated into mobile handsets, but only if there is no adverse effect on its usability, design or performance. And although some current high-end handsets have rewritten the rules on how much a mobile phone can sell for, cost is still a limiting factor for the large volume market segments.

The integration challenge is therefore a delicate balancing act. For example, if a mobile phone maker wishes to include FM, GPS, UWB, and NFC, the average handset does not have the space to house an additional four radio ICs.

One way to approach this is to adopt the integrate-everything-now approach. This ignores any negative effects that combining multiple technologies might create, including whether or not multiple radios will “play nicely” when side by side.

Perhaps even more fundamental is the question of whether handset designers actually want certain technologies to be integrated on a single piece of silicon. If two technologies have wildly different attach rates in mobile phones, combination chips will waste money and power.

With the recent launch of its BlueCore 7 product range, CSR has opted for a strategy of what it calls “smart integration.” A highly integrated chip, BlueCore 7 is a good example of the way in which functions can be combined that makes sense for the handset designer (see Fig. 1 ).

Fig. 1. Smart integration refers to the ability of different systems to share resources, and to be aware of what the other on-chip elements are doing.

This single-chip CMOS IC measures less than 4 x 4 mm, yet includes Bluetooth v2.1 + EDR, Bluetooth low energy, enhanced GPS (eGPS), an FM transmitter, and a FM receiver with RDS. Smart integration combines complementary technologies in a single chip with performance and power that are at least equal to, if not better than, separate chips. There are also substantial savings in terms of cost and footprint.

Smart integration also refers to the ability of these different systems to share resources, and to be aware of what the other on-chip elements are doing. This minimizes the need for external components, and the overall cost and size of the final implementation is reduced even further. This approach also makes it easier to design multiple radios into any device by having a single chip that is pretested and supplied with the appropriate firmware for easy integration.

To see an example of dumb integration, you only need to look at the early attempts at integrating two 2.4-GHz wireless technologies (Wi-Fi and Bluetooth) into a single chip. Gluing the two technologies together is straightforward enough, but making them work well together, especially if the two radios need to be active at the same time, is a completely different challenge.

Instead, a module integrating separate Bluetooth and Wi-Fi chips can offer performance levels far in excess of single-chip solutions, with a similar physical footprint. In addition, coexistence software is available to ensure trouble-free operation of both radios.

Importance of attach rates

It is also very important to consider the attach rates of any technologies you are thinking about combining. The attach rate of Bluetooth in mobile phones, a figure which describes how many phones have Bluetooth, stands at well over 50% today (see Table 1 ).

Table 1. Worldwide Attach Rates in Mobile Handsets

In other words, Bluetooth is well established; it has a large installed base and substantial industry-wide momentum. It therefore makes a lot of sense to use Bluetooth as a building block around which other connectivity technologies can be based.

This “connectivity center” could be a Bluetooth hub on which value-added complementary technologies are integrated. This strategy can drive much wider and faster growth in the attach rate of developing technologies.

As an example of how this works, let’s look at GPS. Many consumers see the benefit of GPS and have become acquainted with the personal navigation devices (PNDs) in their cars. It’s therefore easy to see the value of having this technology in a handheld form factor to help them navigate the streets of New York to, for example, find a restaurant or theater.

Building this capability into a mobile phone — without adding to cost, size, or power consumption would be a highly desirable feature. Table 1 shows In-Stat’s research for recent and projected attach rates for several popular wireless technologies.

By reusing portions of the Bluetooth receive signal chain, and integrating only modest LNA and GPS specific functions, less than $1 is added to the overall cost of each chip. And since there is no separate chip for the GPS radio, there is also a substantial saving of valuable PCB space.

This customer-centric approach to combining technologies is key to the concept of “smart integration.” The high attach rates of Bluetooth will allow designers to save money and resources through a single-chip solution, but will also drive the attach rates of other wireless technologies like GPS within mobile phones.

In keeping with this “smart integration” philosophy, silicon vendors should only integrate technologies when it makes sense for the customer in terms of balancing added cost, space, power, and attach rate.

The mobile handset industry is fast moving toward a point where integrating ever more functions inside a single chip seems the sensible way to go. To integrate successfully, chip designers must ensure they have an intimate understanding of both their technologies and their market, and make sure their customers’ concerns are central to their thinking. ■