Proprietary 2.4-GHz wireless audio design

Market demand for greater mobility has led to a migration of traditionally wired audio entertainment products to wireless. As product manufacturers look to cut the cord for entertainment electronics, engineers face significant challenges in maintaining signal range, sound quality, and maximum play time in battery-powered wireless audio designs. To meet these requirements, engineers can draw on a wide range of available products from multiple manufacturers, including (in alphabetical order) Analog Devices, Cirrus Logic, CSR, Freescale Semiconductor, Linear Technology, Linx Technologies, Maxim Integrated Products, Microchip Technology, NXP Semiconductors, Quickfilter Technologies, and Texas Instruments.

In a typical wireless entertainment system (Figure 1), the source signal is transmitted to a player system, such as a stereo headset or speaker, via a wireless RF interface with an optional range extender. Within the player, the corresponding wireless RF interface receives the signal for processing by a codec, audio processor, or DSP to create the final analog signal that drives the speaker. An appropriate power supply, typically consisting of a battery and charge management circuitry, completes the system.

Figure 1: A typical wireless audio system relies on reliable wireless communication between a transmitter and receiver, as well as efficient audio processing on the receiving end. (Courtesy of Texas Instruments.)

When creating wireless entertainment designs, engineers face unique challenges in wireless communications and audio processing. Reliable RF communications with high throughput and signal integrity are critical to ensuring that users receive an uninterrupted audio experience even at extended ranges from the audio source. At the same time, powerful audio signal processing is essential to maintain audio fidelity and deliver the range of audio features the market expects. Available integrated solutions for each subsystem can help engineers meet these requirements while reducing design complexity and cost.

frequency band

Legacy issues, market acceptance and the availability of unlicensed bandwidth often drive the choice of radio communication frequencies. At the same time, meeting requirements for increased operating range and extended battery life sets boundaries for useful frequency bands.

The need for increased power is a natural consequence of the desire to operate at greater distances, but the choice of RF wavelength plays a critical role in balancing range and power. The relationship between RF wavelength and range is described in the Friis transmission equation:

in

Pt = Transmitted power

Pr = Received Power

Gt = Transmitter antenna gain

Gr = Receiver antenna gain

λ = wavelength

d = distance between transmitter and receiver

For unity, distance becomes a simple linear function of wavelength, so longer wavelength radios equate to greater range. Of course, longer bands face issues including interference and payload bandwidth. In this case, the 2.4 GHz ISM band offers a good balance between practical range limitations and useful bandwidth.

The appeal of 2.4 GHz solutions lies in their ability to provide useful range at low power consumption, as well as their global availability. 2.4 GHz designs based on frequency hopping spread spectrum (FHSS) used in standards such as Bluetooth have the advantage of being relatively immune to interfering signals in highly active radio environments. These systems also provide enough data bandwidth to allow digital transmission of high-quality stereo sound, compared to the lower ISM bands, which are typically limited to analog or lower data rate digital audio.

The Bluetooth standard operates at 2.4 GHz, making it a good fit for consumer connectivity requirements. Its widespread use makes it likely that Bluetooth-based wireless audio players will find compatible audio hosts such as computers, laptops, tablets, and smartphones. However, using Bluetooth requires using data throughput rates supported by Classic Bluetooth or Bluetooth Smart Ready devices, which operate in dual mode and support both Classic Bluetooth and low-bandwidth Bluetooth Low Energy (LE). Bluetooth Low Energy is designed to deliver short bursts of data from low-power devices and is not designed to deliver the kind of continuous data streams required for Classic Bluetooth support and wireless digital audio.

Engineers can implement Classic Bluetooth by running a properly configured Bluetooth software stack on an embedded processor such as the Freescale Semiconductor KineTIs family, Microchip Technology PIC24 family, and Texas Instruments Stellaris family. This embedded software approach provides a variety of Bluetooth services, including the Advanced Audio Distribution Profile (A2DP), which provides a standard for streaming stereo audio. A2DP is a familiar option in most Classic Bluetooth devices, and it provides audio that is considered psychoacoustically acceptable by most listeners.

In addition to general-purpose embedded processor solutions, dedicated devices such as the CSR BC57G687C integrate audio signal circuitry to reduce component count and improve power efficiency. The BC57G687C, a member of the CSR BlueCore5-Multimedia family, combines a Bluetooth stack 16-bit RISC MCU with a Kalimba 64 MIPS 24-bit DSP coprocessor and on-chip memory (Figure 2). The device's Bluetooth modem completes the wireless solution.

Figure 2: Integrated multimedia SoCs such as the CSR BlueCore5 series combine Bluetooth radio functionality with extensive on-chip processing capabilities used in wireless audio applications. (Courtesy of CSR.)

Sub-GHz Alternatives

Bluetooth's ubiquity and familiarity to users and most designers give it a clear advantage as a wireless audio solution. However, standards such as Bluetooth, designed to support connectivity between any-to-any products, impose overhead in the communications to ensure authorized, reliable communication between heterogeneous wireless nodes. For engineers seeking greater range or greater audio throughput, other options have their own advantages.

While the first choice is usually the 2.4 or 5.8 GHz ISM bands, sub-GHz bands remain a viable option. Modulation technology improvements have increased interference immunity and alleviated coexistence issues that existed in early sub-GHz approaches. Most importantly, using sub-GHz ISM bands means longer range and lower power requirements, both of which are of course core issues in wireless audio design.

As the Friis equation shows, a low frequency signal translates into a correspondingly longer range. Given this, sub-GHz wireless communications remain attractive for applications where maximum range is a primary concern and audio requirements are less intensive.

For example, Linx Technologies quotes a 1,000-ft. communications range between its TXM-900-HP3 transmitter module and RXM-900-HP3 receiver module. Designed as an embedded solution for wireless applications, these modules require only a single antenna to complete a 900-MHz multichannel RF design capable of transmitting both analog FM and digital FSK information. The TXM-900-HP3 transmitter uses its precision 12-MHz voltage-controlled crystal oscillator (VCXO) to drive a PLL to form a frequency synthesizer that is managed by an onboard microcontroller, allowing software-based channel selection (Figure 3). The complementary receiver module in the RXM-900-HP3 demodulates the signal to provide analog and digital data, as well as an RSSI value that can be used for a user range display indicator.

Figure 3: The TXM-900-HP3 transmitter signal path uses an MCU-controlled PLL to provide software-controlled channel selection in the 900 MHz band. (Courtesy of Linx Technologies.)

Engineers can also find a variety of RF ICs to build custom wireless audio solutions that operate in the 900 MHz band. For example, the ADF7025 from Analog Devices is an ISM transceiver IC that operates in multiple bands, including the 900 MHz ISM band. The ADF7025 includes an on-chip ADC, eliminating the need for a separate ADC to collect basic data such as temperature, battery status, or RSSI. As a result, the device requires very few external components to provide a cost-effective solution. In addition, engineers can tune the device to balance power and sensitivity requirements. The signal chain of the ADF7025 includes multiple programmable options for receiver linearity, sensitivity, and filter bandwidth (Figure 4).

Figure 4: Engineers can set several programmable options in the ADF7025 front end to trade off requirements for power, linearity, sensitivity, and filter bandwidth. (Courtesy of Analog Devices.)

Proprietary 2.4-GHz solution

While Classic Bluetooth is ubiquitous and sub-GHz offers extended range, the use of 2.4 GHz ISM and proprietary communication protocols opens the door to the highest quality multi-channel audio at acceptable range and power. With its lightweight, application-specific stack, the proprietary protocol reduces overhead and saves maximum bandwidth for the payload. On the other hand, proprietary protocols also mean that the host needs a bridge device as well. However, for wireless audio systems, a specialized bridge device can simply be built into both the player device and the host "console" unit, for example, which can be used for user display of speaker status in any case.

Designed for wireless streaming of high-quality multichannel audio, the Texas Instruments PurePath wireless platform uses this proprietary approach in its transceiver and range extender set. TI PurePath devices include the two-channel CC8520 and four-channel CC8530, as well as CC8521 and CC8531 versions that also offer USB audio support. The CC85xx SoC integrates the complete signal path from RF to digital output, enabling engineers to build wireless audio designs with few additional components (Figure 5).