As the pressure to reduce development costs drives the re-use of hardware and software platforms across multiple form factor variants, reconfigurability is proving of huge benefit, as Robert Hatfield, Principal Solutions Architect of Mobile and Headsets at Wolfson Microelectronics explores
Mixed-signal audio components have not traditionally been the most flexible of devices with usage-mode-specific hardware preventing cross-platform re-use. A new generation of audio hubs is now bringing a new level of flexibility to audio in mobile devices while simultaneously simplifying system integration. There are many positive side effects, which can enable new application features, reduce BOM cost, extend talk time and improve performance capabilities.
Tablet PCs provide the most recent illustration of hardware platform re-use. Most tablet manufacturers tend to recycle as much as possible from their smartphone platforms.
However from an audio perspective, a tablet has defining characteristics, which can make this approach non-ideal when using legacy architectures and can lead to wastage and performance compromises.
The larger form factor of a tablet provides space for larger loudspeakers and more of them, including lower impedance speakers, which provide higher output power for the same speaker driver supply voltage.
Advanced compression techniques can further improve loudness when optimised for the acoustics of the playback device (as opposed to compression at source, which takes no account of the playback environment).
Many tablets now include two or three loudspeakers with the ability to deliver stereo audio, improving the video playback or gaming experience when tuned effectively. The extra size and weight of tablets makes them a little less portable however. A preference for indoor use is the result, where the browsing experience is more pleasant on a larger screen anyway.
The availability of wi-fi networks in the home, office or on the high street has led to many lower-cost tablets having no cellular radio capability at all. This means, voice communication is more often IP-based.
However, during a voice call it is typical for many smartphone platforms to implement speech pre- and post-processing algorithms within a cellular modem, which means that in non-cellular tablets these algorithms must be ported to a different core (usually the application processor) or in some cases implemented on additional hardware. This can complicate software re-use and seriously impact development time and cost.
Many of today’s audio hubs are already used as a transducer management platform, which can simplify integration by allowing tuning to take place at a late stage of development without needing to change a stable host software platform, for each form factor variant’s unique acoustics.
In modem architectures, audio hubs are increasingly becoming the natural home for voice pre- and post-processing and this can help to reduce time to market in many mobile applications while maintaining consistency of the audio ‘signature’ across different uses.
The larger display of a tablet amplifies the importance of graphics and touchscreen performance for the user.
Some recent innovations in tactile feedback mechanisms have the potential to make future touchscreen more natural and interactive as multiple transducers are embedded within the device and their control signals become more sophisticated, even using audio-like streaming.
The right audio architecture can now bring some of these new experiences to tablet users without significantly changing the system and without impacting the cost of more basic devices.
Tablets invariably have no earpiece speaker, with voice calls taking place in speakerphone mode or via headsets. Most tablets, which re-use phone designs today will nevertheless contain a mobile phone CODEC with a power amplifier dedicated to driving an earpiece, either by virtue of its amplifier design or related analogue signal routing (for example, the assumption of mutually exclusive loudspeaker and earpiece operation).
This is a redundant amplifier in practice in most tablets and it cannot easily be re-tasked to more useful functions in a tablet in legacy architectures.
The lack of earpiece call capability and the high-resolution display make video call and speaker-phone performance particularly important for tablets.
With high bandwidth wi-fi networks more commonly used, a richer and more natural-sounding video call is now theoretically possible due to the wider audio frequency range, which is transmitted in a wideband voice call.
However the extra distance from the mouth to the microphones during a video call increases the level of ambient noise relative to the signal of interest to the remote caller. The performance bottleneck for the overall call experience therefore moves away from network bandwidth to local signal processing capabilities, in particular the transmit-path noise cancellation and echo cancellation performance.
Since tablets can be held in portrait and landscape modes the stereo context of a loudspeaker can change dynamically as the tablet orientation changes.
Tablets containing three or more loudspeakers can bring improved stereo performance if the left and right channel mapping is orientation-sensitive, particularly when stereo widening algorithms are also employed.
This is only possible however when three or more speaker amplifiers are available in the system. Smartphones typically integrate just one loudspeaker and audio platforms must provide for this expansion in a way which minimises software and hardware impact and which doesn’t impact the cost of more basic smartphones built from the same platform.
The recently-released WM5100 from Woolfson pioneers a new type of audio hub architecture, designed to maximise flexibility and cross-platform re-use while enabling many future capabilities such as advanced multi-channel haptics, high performance noise cancellation technologies and a more scalable loudspeaker footprint.
The inclusion of PDM outputs provides robust and low cost digital connectivity for up to four external power amplifiers in addition to the on-chip class D and class W drivers. This allows the same audio platform to support a greater range of numbers and types of loudspeakers and other transducers, enabling re-use in more diverse applications.