In This Section
Journal of the AES
2013 April - Volume 61 Number 4
Guest Editors’ Note
Although modern WiFi technology has enough bandwidth to handle multiple audio channels, the current 802.11 standard has numerous defects that prevent widespread adoption for professional live audio. By augmenting the MAC algorithm used in the standard, many of the inherent problems can be addressed. For example, the use of CTS-to-Self was modified in order to distribute in broadcasting mode, the time the network is occupied for each transmission. This reduces the number of backoff counts and dropped packets. An Exclusive Backoff Number Allocation algorithm was also added in order to decrease the probability of collision while maintaining fairness and limiting delay to an acceptable level. A simulation of the modified 802.11 MAC algorithm showed drastically improved performance.
Efforts to create low-latency protocols to ensure quality of service (QoS) for audio networks often ignore the importance of the operating system (OS) in the computers at the ends of the network chain. An experimental OS called Tesselation, developed in the Parallel Computing Laboratory at UC Berkeley, was tailored to multicore processors with such features as guaranteed resource allocation and customizable user-level runtimes. To demonstrate performance isolation and service guarantees, the authors tested Tessellation under various conditions using a resource-demanding network application. This OS enables network audio applications to meet their time requirements.
The Open Control Architecture (OCA) is a proposed standard for control and monitoring of professional media networks. OCA is intended for system control and monitoring only, and may be integrated with any streaming program transport control scheme that affords suitable connectivity for OCA traffic. A descendant of the AES24 scheme developed by AES Standards Subcommittee SC-10 in the 1990's, OCA was developed by an informal collaboration of nine audio companies in 2011. This collaboration has now become a trade association named the OCA Alliance, and the standardizing of OCA is now the task of the AES standards project known as X210.
AES64, which is an IP-based peer-to-peer network control protocol, integrates the connection management and control capabilities across digital audio networks. This enables the streaming of all types of media content among devices from different manufactures because it provides a common approach for device control. This paper describes this protocol. Each parameter of a device can be addressed with a 7-level hierarchical structure that reflects the functional layout of the device. The protocol has been implemented in many different types of networks.
The lack of a universally accepted standard control protocol, as well as a lack of interoperability among existing protocols, limits how audio software can support a wide range of networked devices. To address this issue, the authors propose an abstract standard audio device model that is independent of any particular technology. This standard device model supports service discovery and enumeration, while also specifying device parts, the representation of parameter addresses, and the visual appearance of controls. The discussion explores the model in the context of a self-configuring discovery and control environment. Within this environment, controllers do not “pull” descriptions from a device; but rather, devices “push” full-service implementations to controllers.
Audio/Video Bridging (AVB) on Ethernet refers to a suite of standards that allows for deterministic and guaranteed delivery of audio and video content on Virtual Local Area Networks (VLAN). A requirement for the networking of audio devices is to allow for remote establishment and destruction of audio stream connections. There are a number of sound control protocols that utilize AVB transport, and this paper describes an approach for common control and interoperability among them. To demonstrate this approach, this paper describes the design and implementation of a command translator that enables communication between AES64 and OSC-networked AVB devices. Results from a quantitative analysis reveal that when a proxy is used for connection management, noticeable delays are not added to a user’s visual and auditory perception.
Audio Network-Based Massive Multichannel Loudspeaker System for Flexible Use in Spatial Audio Research
In order to provide a flexible means for exploring various spatial audio algorithms in sound field synthesis, a massive multichannel system consisting of an array of 640 loudspeakers was created. It is based on an audio network that distributes discrete audio signals to a grid of amplifiers. Given the massive size of the loudspeaker array, special means for configuring, routing, control, reliability, flexibility, redundancy, economics, and cabling had to be developed. The resulting system should prove to be a powerful tool for research various presentation formats.
A great deal of attention is now focused on how to achieve satisfactory or even good audio quality with mobile devices and other acoustically-challenged products such as laptops. There are also emerging standards such as HTML5 that promise a range of audio features for web-based devices. We consider the signal processing and audio architecture requirements for such systems.