No compromise for mobile audio
20 January 2008
In audio-focused devices skilled analogue design will be crucial to deliver hi-fi sound in the power, size and cost constraints
Almost as much of a lifestyle-accessory as a communication tool, mobile phone handsets are increasingly becoming differentiated in order to succeed in today’s competitive market. Vendors aim to create highly targeted new models that satisfy core requirements better than any competitor product aimed at the same market segment. A gaming phone, for example, must prioritise features that will deliver an outstanding experience for gamers. An MPEG4 video/camera phone, on the other hand, must provide a seamless image capture and audio playback capability for sharing. Users who enjoy listening to music may even wish to share music content with friends during telephone calls, creating yet another different set of usage scenarios.
Handset technologies, as well as approaches to integration, are continually evolving in order to meet these challenges. For example, to maximise media playback performance and streaming audio in next generation music-oriented handsets, integrators will demand a much higher performance and feature set than currently achievable from existing codecs.
Codecs plus
The integrated baseband chipset approach has evolved to cost-effectively address basic voice telephony in current 2G/2.5G networks. A mixed-signal voice codec provides the analogue front end to a low power DSP which performs the channel coding and compression/decompression algorithms in order to lower the data rate over the air interface. Balancing network capacity with speech channel performance basically dictates the functionality (sampling rate, bandwidth) of the codec, enforced by the applicable cellular type-approval standards. Although audio capability may be added by bolt-on integration of stereo audio DACs, constraints and limitations of the integrated approach typically yield up to 10dB signal-to-noise ratio degradation when compared to performance achievable with a standalone audio codec.
However, with consumer focus on richer multimedia applications supported by higher data bandwidth 3G networks, and the prospect of wireless broadband on the horizon, handset integrators need more capable codecs to support true ‘mobile hi-fi’ products. Indeed, due to ease of use and portability, tomorrow’s handset may be expected to perform the role of ‘mobile multimedia hub’ as consumers roam between house, office and automotive environments. Determining the optimum architectures to satisfy these emerging applications demands mixed-signal architectures that are capable of unlocking and further stimulating this digital revolution.
Music features
For example a suitable codec for music phone applications will blend digital features more usually performed in software, such as enhanced 3D surround sound capabilities, multi-band equaliser, auto level control and bass boost. Wolfson Microelectronics’ WM8983 stereo audio codec, for example, implements this feature set around a 24bit high-order oversampling ADC and hi-fi DACs plus offering additional digital application filtering functions such as wind noise and programmable notch filters. By implementing such processing in dedicated hardware these optimised codecs will allow handset integrators to reduce the processing demand on the device’s main multimedia processor with cost, memory and power savings.
Flexible digital and analogue interfaces add versatility by allowing the mixing of synthesised content (such as ringtones) with audio or voice all at different sample rates. Emerging audio scenarios include initiating normal mobile phone functions while continuing to listen to MP3 or FM radio, as well as audio playback during calls. Other functions now becoming attractive to music phone users include karaoke, with full mixing capability for headphone playback as well as recording, facilitating sharing of karaoke files via MMS or email.
Audio-centric
By adopting an application rather than baseband approach mobile codecs deliver audio-centric codec solutions at competitive power consumption and small form factor. Flexible interfacing makes for seamless integration into various baseband architectures. For example, the WM8753 supports PCM/I2S dual interfaces and WM9713 offers AC’97/PCM and touchscreen support. Dual interface products allow user applications (such as gaming and MP3 playback) to function with less baseband intervention or provide an additional interface for peripherals (such as Bluetooth headsets).
Integrating power management
Although the signal processing power dissipation of audio codecs continues to move downward, the market is now demanding louder and more powerful speaker drivers. Without lowering speaker impedance this necessitates high voltage supply operation coupled with more power efficient speaker driver topologies to maintain acceptable battery runtime between charges. Mobile TV and ‘video snacking’ are examples of emerging applications, demanding better speakerphone capability. This direct battery-voltage capability within the codec may be usefully exploited by co-integrating power management functions such as voltage regulators. But what challenges must be overcome?
In general, both mixed-signal and power management circuits use transistors as analogue devices rather than digital switches. Hence, both approaches focus on optimising analogue performance metrics, enabling integration to proceed with few compromises. Like voltage regulators, mixed-signal audio circuits often work with relatively large voltages and currents, for example to drive miniature loudspeakers.
Mixed-signal and power management circuits also have much in common in the way they are designed. Both need detailed analogue simulations (e.g. SPICE), based on accurate mathematical models of individual transistors.
Design challenges
Some analogue characteristics that differentiate products in the eyes (and ears) of end users, such as noise, harmonic distortion, and power efficiency, are notoriously difficult to predict through simulations. Analogue IC designers take this into account and often fine-tune the layout of circuit elements on chips by hand. Additionally, external components and PCB layout often determine the end product's performance as much as the chip itself, making a system-level approach indispensable.
However, combining mixed-signal with power management also raises new issues. The power efficiency of today's portable devices is only possible with judicious application of switched-mode power supplies, which naturally generate switching noise. Placing audio or video functions on the same die creates many coupling paths, degrading achievable performance and impacting the end user's listening or viewing experience.
Mixed-signal benefits
Fortunately, mixed-signal designers have already developed many techniques to deal with this issue. Mixed-signal ICs have always contained a great deal of digital circuitry, which also produces switching noise. The methods they have developed to protect sensitive analogue signals can deal with larger amounts of noise, too. Power management designers, on the other hand, are generally more adept at minimising noise/EMI produced at the board-level.
Giving mixed-signal designers control over the power supply has advantages for system-level design, too. Some of the most prevalent performance issues plaguing mixed-signal circuits today are directly related to the quality and stability of power supplies. For example, the switching noise problem can be addressed by making the power supply switch at those frequencies where the mixed-signal functions are least sensitive to interference. Tight integration of power management with audio functions also enables the supply to react more quickly, or even anticipate, sudden increases in the system's power consumption. Such power surges are often caused by volume peaks in audio signals, especially when played through a loudspeaker. Shortening the power supply's reaction time reduces the need for storing a reserve of electrical charge in large, costly capacitors. As a result, smaller, lighter and cheaper capacitors can be used.
Integration
Mixed-signal designers’ familiarity with ‘hand crafting’, and their experience in listening to audio performance and tweaking the circuit, even at the prototype silicon level, to achieve the best possible sound, are invaluable when creating an integrated mixed-signal audio and power management chip with no compromise on audio performance.
In an increasingly digital world the combination of feature-rich mixed-signal audio integrated with power management is increasingly becoming the architecture of choice for emerging multimedia handsets. Up until now the quest for higher levels of mixed-signal integration has been achieved by sacrificing audio capability in standards-driven baseband products. Future developments will address the applications-driven market where higher quality audio will be integrated with power and digital features.
ERIC HABER is a product architect, Wolfson Microelectronics plc
Contact Details and Archive...
Related Articles...
Most Viewed Articles...