Revolutionizing Audio: The Game-Changing Impact of LE Audio Technology

This article talks about LE Audio, the reasons for its rising popularity and the benefits it brings. It also details how it solved the challenges of existing Classic Bluetooth Audio, the status of implementation and prospects for adaptations in various application domains. Anyone with an interest in Bluetooth audio and related fields could be the audience of this article.

INTRODUCTION

In the two decades since its inception, Bluetooth? technology has emerged as the preferred solution for a wide range of wireless audio streaming applications. If you've ever enjoyed wireless music through a headset or your car's infotainment system, chances are you've experienced the convenience of Bluetooth technology.

The main Bluetooth audio (Classic Audio) standards, HFP and A2DP, were initially designed for specific use cases without foresight for future needs. Consequently, they exhibit distinct behaviors. HFP prioritizes low-latency, bidirectional, mono voice transmission, while A2DP focuses on high-quality music streaming to a single device without a return audio path. Both profiles lack easy extensibility, limiting their adaptability.

We're increasingly relying on wireless audio in our daily lives for communication and personal space. To adapt to this shift, we must move beyond focusing on connections between individual devices, such as headsets and phones. Instead, we should consider a broader audio ecosystem with various ear-worn devices seamlessly interacting throughout the day. To achieve this, control mechanisms need enhanced flexibility, prompting the development of Bluetooth LE Audio.

The hearing aid legacy

Surprisingly, much of this innovation was spurred by the hearing aid industry, addressing challenges like audio quality, latency, battery life, and broadcast transmissions. With users wearing hearing aids for an average of nine hours a day, battery life is crucial. The desire to connect to phones and Bluetooth devices was hindered by the power consumption of traditional HFP and A2DP solutions.

Constraints and proprietary enhancements

True Wireless

In 2013, Cambridge Silicon Radio developed a stereo streaming solution for two earbuds, later rebranded as TrueWireless by Qualcomm after the acquisition. The term TWS (True Wireless Stereo) quickly became the norm for new products, irrespective of the chip inside.? CSR's solution is known as the replay or forwarding approach. In this design, the Primary earbud must manage two audio connections, resulting in increased power consumption. A2DP faced challenges for separate earbuds due to its single point-to-point design.

With both earbuds directly receiving the same A2DP stream, this is a potentially more robust Bluetooth-only scheme, offering significantly better latency where no audio stream relaying is involved.

It's nearly impossible to mix earbuds from different manufacturers due to likely differences in TWS schemes. This might not be an issue for consumer TWS earbuds sold as paired sets, but it poses challenges for hearing aids and speakers, where different types may be needed. Proprietary solutions can boost market growth initially but often hinder long-term development. This is where Bluetooth LE Audio comes in.

Shared listening

A Paris based Bluetooth software company developed a phone-based audio OS called Dual-A2DP, generating separate left and right streams for two earbuds simultaneously. Though limited and adopted by few, Bluetooth LE Audio surpasses this, offering a scalable solution for transitioning from one to many listeners.

INTRODUCING BLUETOOTH? LE AUDIO

Building on 20 years of innovation, LE Audio enhances the performance of Bluetooth audio, adds support for hearing aids, and introduces Auracast? broadcast audio, an innovative new Bluetooth use case with the potential to change the way once again we experience audio and connect with the world around us.

In the early stages of Bluetooth LE Audio development, four waves of use cases drove its evolution, beginning with those from the hearing aid industry. These were centered on topology, power consumption, and latency.

Hearing aid requirements added features to the core, primarily for performance and power saving. This includes the flexibility for the new codec to be implemented in the Host or the Controller, with the latter being more power-efficient for hardware implementations.

Advancing beyond the capabilities of HFP and A2DP

Nowadays, VoIP calls being common, the need for more sophisticated media control, including handling multiple calls on a single device, emerged. Additionally, products required the ability to support voice commands without interrupting a music stream.

In today's phone and conferencing apps, users frequently switch between calls and audio streaming across different devices and applications. The inherent architectural differences in HFP and A2DP led to a set of best practices in the Multi-Profile specification. Bluetooth LE Audio had to surpass this by incorporating built-in multi-profile support, ensuring robust and interoperable transitions between devices, applications, and unicast/broadcast scenarios.

THE BLUETOOTH? LE AUDIO ARCHITECTURE

The Bluetooth LE Audio architecture has been built up in layers, as has every other Bluetooth specification before it. This is illustrated in the below figure, which shows the main new specification blocks relating to Bluetooth LE Audio, (with key existing ones greyed out or dotted).

LE Audio is the next generation of Bluetooth audio that is designed to enhance the performance of standard Bluetooth audio and takes advantage of Bluetooth Low Energy radio.

Isochronous Channels

LE Audio incorporates a new, high-quality, power-efficient audio codec named LC3, which uses LE Isochronous Channels for low-power data transport and includes new middleware designed for the delivery and control of audio content.

The Core proposal suggested implementing Isochronous Channels for audio streams in Bluetooth LE, alongside the existing ACL channel. ACL handles setup, control, and generic information, while Isochronous Channels support unidirectional or bidirectional audio streams, allowing multiple channels with multiple devices. This separation enhances the flexibility of Bluetooth LE Audio.

A significant improvement in Isochronous Channels is the capability to simultaneously stream audio to multiple devices and synchronize the rendering perfectly.

Unicast

In a CIG, CIS Events are scheduled for each CIS, such as one for the left earbud and one for the right earbud in a pair. These events form a CIG event. The Core defines a CIG reference point, usually slightly before the Anchor Point of the first CIS, and a CIG synchronization point after the last receive event for the last CIS in that CIG. Each device is informed of its CIS Sync Delay, calculated from the Instant for the associated CIS's ACL link. This allows calculating the common timestamp for CIG synchronization. With this knowledge, devices can determine when to start decoding audio. BAP introduces Presentation Delay, guiding the Acceptor on when to render the decoded stream.

Broadcast

As with Connected Isochronous Streams, individual receiving devices, typically a pair of earbuds or hearing aids, don't necessarily know about each other's existence. At the BIG Synchronization Point, all devices receiving Broadcast Isochronous Streams within the BIG are assured that every other device has received the data, establishing a fixed time for applying the Presentation Delay. This delay, defined higher up the stack, indicates when audio should be rendered. Notably, audio rendering does not occur at the BIG Synchronization Point but initiates at the end of the Presentation Delay, starting from the BIG Synchronization Point. In broadcast, where transmissions are unidirectional from a Broadcast Source, there is no Presentation Delay applied to captured data returning from an Acceptor.

Extended Advertising

Broadcasting in Bluetooth LE introduced new concepts, especially for devices to discover broadcasts without a connection. Bluetooth LE utilizes advertisements for devices to announce their presence.

In Bluetooth LE, the 2.4 GHz spectrum is segmented into 40 channels, each spanning 2 MHz. Among these, channels 37, 38, and 39 are designated as the Primary Advertising Channels, reserved for fixed-frequency advertising. However, the substantial volume of information intended for transmission by the Broadcaster surpasses the capacity of these three primary advertising channels, risking overload. To address this constraint, the Extended Advertising feature of Core 5.1 permits the transfer of advertising information to the general-purpose channels (channels 0 to 36). To achieve this, a Broadcaster incorporates an Auxiliary Pointer (AuxPtr) in its primary advertisements, notifying devices that additional advertising details are available in the general-purpose channels, now serving as Secondary Advertising channels.

Extended Attribute Protocol (EATT)

The variety of control profiles in Bluetooth LE Audio led to the EATT enhancement in the Core. The Attribute Protocol (ATT) used for communication assumes a single command at a time, causing delays for concurrent commands due to its blocking nature. The Core 5.2 release introduced the Extended Attribute Protocol (EATT) to address this, enabling multiple ATT instances to operate simultaneously.

Generic Audio Framework (GAF)

In the Host, there's the Generic Audio Framework (GAF), serving as audio middleware with generic functions for various audio applications. The Core and GAF form the core of Bluetooth LE Audio, offering flexibility. At the stack's top are 'top level' profiles adding application-specific details to GAF specifications.

THE LC3 CODEC

The role of an audio codec is to compress the audio stream at the source in preparation for transmission and then decompress it at the receiving end for playback.

During Bluetooth LE Audio development, it was evident that existing Bluetooth codecs faced challenges in meeting requirements. They posed limitations in quality, latency, and efficiency. SBC, despite its low complexity, proved inefficient for earbuds and hearing aid designers, impacting battery life. This was particularly concerning for hearing aids with small zinc-air batteries sensitive to peak current and burst length during reception or transmission. To overcome these issues, the Bluetooth SIG embarked on a codec search, leading to the adoption of LC3.

Bluetooth LE Audio allows manufacturers to use other codecs, but LC3 is mandatory for all devices. The reason for this is to ensure interoperability, as every Audio Source and every Audio Sink has to support it.

The LC3 is one of the most advanced audio codecs available today, providing enormous flexibility and covering everything from voice to high quality audio.

The codecs operate at low latency, low computational complexity, and a low memory footprint.

The LC3 specification stands out as a successful endeavor to encompass diverse audio quality and latency requirements for wireless audio in a single codec. Optimized for a 10ms frame size, it accommodates new applications, particularly in public broadcast, while also supporting a 7.5ms frame size for compatibility with Bluetooth Classic Audio applications. Additionally, it extends to a 10.88ms frame to accommodate legacy 44.1kHz sampling, with a reduced 8.163ms variant for systems supporting both Bluetooth Classic Audio and Bluetooth LE Audio implementations.

Some of the technical properties of LC3 include:

·???????? It is a block-based transform audio codec

·???????? It provides a wide range of usable bitrates

·???????? It supports a frame interval of 10 ms and 7.5 ms

·???????? It supports the following bit depths: 16, 24, and 32 bits per audio sample

·???????? It supports an unlimited number of audio channels

·???????? It supports the following sampling rates: 8 kHz, 16 kHz, 24 kHz, 32 kHz, 44.1 kHz, and 48 kHz

Listening test performance

The LC3 codec achieves a high compression ratio without compromising audio quality, with low latency suitable for both conversation and music streaming. The X-axis represents various bitrates for each SBC and LC3. The Y-axis represents subjective methods for assessing minor audio system impairments.

Packet Loss Concealment (PLC)

To enhance the listening experience, the industry has devised Packet Loss Concealment algorithms to predict and conceal missing audio. These algorithms work well with voice and reasonably well with music, though challenging for segments near attack transients. The LC3 specification features a matching Packet Loss Concealment algorithm activated by the Bad Frame Indication flag for lost or corrupted frames, ensuring a consistent listening experience. It is recommended to always use this or an alternative PLC algorithm.

LE AUDIO SECURITY

??Pairing connected devices provide link security for unicast audio

??Broadcast Codes may be used for the encryption/decryption of broadcast audio

??Broadcast audio may be completely open and public

??Coordinated Set membership secured with a Set Identity Resolution Key (SIRK)

LE AUDIO APPLICATIONS

Auracast? Broadcast Audio

Auracast? broadcast audio is a new Bluetooth? capability that will deliver life-changing audio experiences. It will let you share your audio, unmute your world, and hear your best, enhancing the way you engage with others and the world around you.

How Auracast? broadcast works

With an Assistant

1.?An Auracast? transmitter advertises the availability and details of a Standard Quality, Auracast? broadcast stream.

2.?An Auracast? assistant scans for advertisements and provides a user interface (UI) to join an Auracast? broadcast stream.

3.?Once an Auracast? broadcast is selected, the Auracast? receiver to joins the broadcast, directly.

The Broadcast Assistant is your reliable companion and commander in the realm of Auracast. Easily integrated into familiar devices such as your phone, tablet, or smartwatch, this feature establishes a seamless connection with your receiving device, handling the intricate details.

By scanning for accessible broadcast sources, the Broadcast Assistant presents a user-friendly interface, enabling you to effortlessly choose your preferred audio source. It equips your receiving device with all the necessary information to seamlessly acquire and play the audio broadcast. To streamline the process further, it can even scan QR codes or read NFC tags.

Without an Assistant

1.?An Auracast? transmitter advertises the availability and details of a Standard Quality, Auracast? broadcast stream.

2. An Auracast? receiver scans for advertisements and provides a mechanism (button, swipe, switch) to join an Auracast? broadcast stream.

3. Once an Auracast? broadcast is selected, the Auracast? receiver to joins the broadcast, directly.

UNICAST MEDIA STREAM EXAMPLE SEQUENCE

REFERENCES

https://www.bluetooth.com/learn-about-bluetooth/feature-enhancements/le-audio/

https://www.bluetooth.com/auracast/

https://www.bluetooth.com/blog/a-technical-overview-of-lc3/

https://www.bluetooth.com/learn-about-bluetooth/feature-enhancements/le-audio/le-audio-specifications/

https://www.youtube.com/watch?v=dEz2CC0bZZc

https://www.youtube.com/watch?v=s6i6t2i2wcE

https://www.bluetooth.com/specifications/specs/core-specification-5-4/

https://www.bluetooth.com/specifications/specs/telephony-and-media-audio-profile-1-0/