When DSSS and FHSS fail: Avoiding interference on the 2.4GHz wireless desktop

The global availability of the 2.4GHz band has its advantages and disadvantages. On the plus side, manufacturers of RF products are fortunate that their 2.4GHz designs can be sold virtually worldwide without modification.

On the downside, manufacturers have to deal with the band's heavy traffic. And it's only going to get worse (which raises the spectre that a product that works today could fail in the field tomorrow and become an expensive warranty issue).

Contemporary IEEE.802.xx standards-based wireless technologies such as Wi-Fi, Bluetooth and ZigBee all crowd into the 2.4GHz band, along with various forms of wireless Ethernet and USB. In addition, there are many proprietary manufacturers, including Nordic Semiconductor, that also employ 2.4GHz wireless technology.

This is not to say the 2.4GHz band can't cope. It extends from 2.4 to 2.483GHz providing manufacturers plenty of opportunities to search for clear narrowband frequency channels within the allocation should their transceivers experience interference from other devices.

There are several established techniques for efficient use of the 83MHz width of the 2.4GHz band - ranging from the bullish approach of simply repeating the transmission on high power until it finally gets through to the elaborate direct sequence and frequency hopping spread spectrum schemes used by Wi-Fi, Bluetooth and ZigBee.

These latter schemes work well, but when it comes to ultra-low power wireless connectivity—where the unit's current can't exceed the peak current of coin cells or other small batteries typical of these applications and furthermore the batteries are expected to last for months or even years—these schemes demand power resources that just aren't available.

Meeting these exacting constraints requires innovative design unencumbered by the restrictions imposed by today's IEEE-based standards.

A good example is the wireless peripherals for PC. End users expect to use several 2.4GHz peripherals simultaneously (for example, keyboard, mouse, game pad and for the new generation of media centres, RF remote controls) often in very close proximity to Wi-Fi and Bluetooth-equipped products not least the PC itself.

But also others such as cell phones, headsets and cordless phones. Consumers demand uninterrupted communications and peripherals that fail to perform have no chance of selling in high volumes.

To meet consumers' expectations 2.4GHz peripherals must communicate with no loss of user data and fast (low latency) response in the presence of other aggressive 2.4GHz RF sources.

Unlike a more forgiving wireless application such as a wireless sensor application monitoring temperature, where undelivered data results in a missed reading refreshed with the next data packet perhaps a half-second later, a missed keystroke from a wireless keyboard causes intense user frustration.

Worse still, there may be a strong temptation for the consumer to switch to another vendor's product, or even return to a wired connection. Put simply: poor performance in the presence of interference can make or break a 2.4GHz consumer product.

So, what are the key issues in providing a wireless product with immunity from 2.4GHz interference? This article provides some answers with reference to a wireless desktop (although the fundamental principles are applicable to many 2.4GHz wireless application).

Understanding interference avoidance
The three popular IEEE-based wireless technologies, Wi-Fi/ZigBee and Bluetooth employ Direct Sequence Spread Spectrum (DSSS) and Frequency Hopping Spread Spectrum (FHSS) schemes respectively to maintain the integrity of the link.

(See sidebar "Interference free communication" for a summary of the basics.)