Wireless Audio and Data Using Frequency Hopping at 2.4 GHz
Advanced digital wireless networks, which are fundamentally changing the commercial and professional audio industries by intimately combining frequency hopping technology and digital audio
Wireless Audio and Data Using Frequency Hopping at 2.4 GHz
Advanced digital wireless networks carrying multiple channels of high fidelity music and voice multiplexed with data are in operation for the first time. The networks, operating from a PC or as stand-alone systems, are fundamentally changing the commercial and professional audio industries by intimately combining frequency hopping technology and digital audio. Wireless audio is a particularly difficult challenge that requires higher transmission reliability than wireless data networks because errors are audible and unacceptable. In wireless data transmission, the data can be retransmitted when errors are detected. In the commercial sound industry, these new digital wireless networks are providing a true design and operational paradigm shift. Advantages such as faster installations, flexibility in cable-free design, high quality multiple audio and data channels, built-in drivers for light-emitting diode (LED) visual signs for compliance with the Americans with Disabilities Act (ADA), data security with real-time supervision and lower costs are changing the industry forever. This article defines the technological contribution of the wireless distributed audio and data systems and compares it with the recent distributed (networked) computing revolution in the commercial public information systems industry. The wireless audio-visual and emergency system (WAVESª) is described, and the direction these wireless systems are heading is proposed.
David Manela and Alan Avidan
New York, NY
Can you recall the last time you heard an intelligible public address (PA) announcement? PA systems are installed in numerous facilities, including airports, train stations, industrial facilities, malls and stadiums. Clear PA messaging capability often can be the lifesaving difference in facility evacuations and other emergencies. Unfortunately, clear, intelligible PA announcements are not the norm. More likely, these announcements sound garbled, distorted or reverberant, or worse yet, though needed, they don't exist. These announcements often are disregarded because they are anticipated to be unintelligible. Over time, the poor quality of audio messages spewing from cheap (yet expensive to install) PA systems has been accepted. Audible messages alone, even if sufficiently clear, are difficult to understand. Integrated audible and visual PA systems, which are now required by the ADA and can enhance intelligibility substantially compared to audio messaging alone, are only now starting to be implemented (at very high retrofit costs) in transit systems, airports, manufacturing facilities and public gathering places. In the future, integrated audio-visual public information systems will be a major contributing factor in enhancing information recognition in our daily lives.
The answer to the question of how is it that, at the dawn of the 21st century and with all the current technological savvy, we are not yet able to deliver the audio goods rests with two observations: location and the law of averages. Just as in real estate, the right location is everything in audio and visual messaging distribution. The precise placement of properly selected speakers is critical in achieving satisfactory area coverage and good intelligibility results. In most commercial installations the location is scarified readily, often early in the design stage, for cost reasons. The cost of installing a PA system depends heavily on labor in relation to equipment. Running cables in conduit, drilling through walls, pulling wires, hanging speakers and finally adjusting the system in a facility are a major undertaking for both facility owner and the contracting installer. Obstacles such as asbestos in the walls, roads to cross and obtaining right of ways to lay cable increase the level of difficulty and expense. An average installation often involves weeks of construction work and business disruption. Thus, it is quite common to design and build systems in which the speakers are placed only in cost-effective locations vs. the acoustically correct locations.
The law of averages, as defined in this article, explains the second reason for the poor-quality output and low intelligibility of current PA systems. In a conventional wired PA system, a single cable feed (home run) typically drives all the speakers, which are daisy-chain tapped to it. The cable run carries the amplified source audio that is generated in the PA room (or delivered from a remote location) to the individual speakers. Any changes in the audio gain at the source immediately affect all speakers that receive their average portion of the audio. Although it is possible to change the impedance of an individual speaker such that the volume is controlled somehow, it is only a crude control mechanism. The second important setting of audio systems is the equalization. In order to achieve intelligible audio, the signal must be equalized (filtered) to match the environment. Here again, the law of averages takes precedence since the whole speaker line can be equalized at the source only. This average control scheme often produces an unsatisfactory result because each speaker may be located in an acoustically different location, requiring different equalization settings. The third adjustable parameter in the deployment of an audio system is the audio delay. Delay is used as a method to adjust local audio quality for distortions created by audio propagation differences from different sources. Because the delay unit is in the PA room, and because placing a delay unit at every speaker is cost prohibitive, the law of averages again applies. To summarize, the settings of three key parameters in audio system deployment (gain, equalization and delay) are averaged in nearly every installation because of the line feed control limitations. Each of these limitations can be alleviated by a smart wireless system.
The WAVES wireless solution was designed to bring the PC revolution to the audio and data distribution industry. The old mainframe implementation of computer networks and the dramatic changes toward distributed networks, where each unit on the network is an independent PC performing tasks locally, are all too familiar. This revolution has skipped the commercial installed sound industry where the old mainframe paradigm still dominates. All the horsepower is located at a single location (or a few single locations), driving audio power into cables toward the speakers. Generally, a typical PA installation is based in an air-conditioned room containing several 19-inch racks with all the audio processing and amplification equipment installed in it. Wires are pulled from this room, often within protective metal conduit, to distribution points for the audio (the speakers).
The WAVES system approaches the task from the same distributed approach that exists in PC networks. All WAVES transceiver units, shown in Figure 1 , in the private wireless network are wholly independent PA units that contain all the necessary audio-processing capabilities and control features (audio amplification, gain control, equalization and audio delay). Table 1 lists the system parameters. This independent node-level parameter adjustment capability enables system designers to adapt each unit to its immediate acoustic environment and operators to optimize the system's performance in real time and adapt to changing conditions. Consequently, the result is a much higher quality audio (and information) at the exact locations where it is needed.
Fig. 1: A WAVES transceiver.
Table I: System Parameters
Real-life stories on the shortcomings of existing conventional analog, hard-wired PA technology are abundant. At the main production plant of McDonnell Douglas Aerospace in St. Louis, the PA system has degraded to a tone information system. Numerous times a day, tones are blown through the system to indicate the start/stop of shifts and breaks. Audible messages are not possible because of poor intelligibility resulting from placement of the output horns high up near the ceiling (over 40 feet), where they are safe from the traversing gantry cranes. The personnel these messages are intended for work in concentrated areas of production, which are shifted around according to the plane orders being built. The concrete and steel environment further introduces reverberations that distort the sound. Visual messages on LED signs are desired but require all new wiring. McDonnell Douglas turned to WAVES for a solution. Wireless transceivers driving speakers and signs were placed in strategic locations close to the employees and attached to the manufacturing jigs. Now, specific zone-targeted messages produce clear sound and visual messaging. If the manufacturing lines are moved, so too is the PA system.
A shopping mall invested more than $250 K on a new PA system. The average audio settings were such that the audio level at the food court was blasting. After a month of operation, the owners of the food court restaurants demanded that the mall management either lower the audio to a level that allowed people to sit at the food court area, break the lease or use a wire cutter. The management, unable to resolve the law of averages, offered no solution and so the cutter was used.
Approximately four years ago, some unfriendly individuals attempted to blow up the World Trade Center in New York City. The official event investigation report concluded that the responsibility for the high number of injuries fell on the PA system. Surely, everyone at the explosion site was killed. The building's PA system collapsed immediately because the system's wires were cut. Chaos ensued and the uncontrolled spontaneous evacuation that followed resulted in the majority of subsequent injuries, primarily from smoke inhalation and broken limbs caused by descending many floors in smoke-filled, dark stairways. The report wisely concluded that a more fault-tolerant wireless system would have better served the facility and recommended its purchase.
The New York City subway is one of the largest facilities in the world, but its PA system has earned the reputation of being extremely poor in quality and intelligibility. The thinking within management regarding PA messaging was that producing quality audio is an impossible task due to the especially acute acoustic challenges in the subway. An acoustic consulting company was hired and various tests were performed on the premises, but the results where not encouraging. Then, on a chilly Monday morning in December 1995 at the 59th Street Columbus Circle station (a busy station where four active train tracks run in parallel), WAVES engineers deployed a wireless PA system at the different station levels within a few hours. The public event that followed included live speeches by a senior US representative, advocates for ADA and others using the WAVES system on the central platform (with trains rumbling by). The live (open microphone) and prerecorded messages demonstration produced clean, clear sound and visual messages despite the harsh environment and was more than a surprise to the attending dignitaries and the large number of dazed onlookers and media. Not only were voice announcements made, but the event also included playing of background music with surprisingly good quality. Selecting the best acoustic locations and remotely controlling the settings of the individual speaker units in the system produces excellent results even in harsh and noisy environments.
These examples lead to the conclusion that a little modern wireless technology can go a long way in the PA business. As a result, a focused effort was undertaken to develop digital wireless audio products and successfully implement spread spectrum frequency hopping (SSFH) wireless technology Ñ a technology familiar from the earlier days of military developments. The WAVES system was developed specifically to overcome the deficiencies of wired PA systems discussed previously.
The concept of spread spectrum was discovered in the 1940s as part of research in information theory. Actual demand for spread spectrum systems appeared in the 1960s when conventional communication systems could not survive reliably in the electronic battlefield.
The military requirements were stringent. The systems had to withstand jamming of communications (by all types of determined jammers), allow a large number of users to operate simultaneously, have a low probability of interception and a low probability of detection. The solution was to use SSFH. Recently, the need and the varied uses of spread spectrum systems have become obvious in the commercial and industrial markets. The effort received a major push and legitimacy when the Federal Communications Commission (FCC) and other regulating agencies worldwide recognized the potential of the technology and pre-approved its use at specific frequencies without requiring a user license. Today, spread spectrum systems are in use by the military as well as in civilian communication systems.
The WAVES product has been developed using SSFH technology. With this technique, the transmitted carrier frequency is changed rapidly in a pseudorandom manner. The receiver, with prior knowledge of the transmitting parameters, acquires the transmitting signal, synchronizes and locks to it. Thus, from the perspective of a viewer also shifting frequencies with the system, the transmitter and the receiver appear to operate as a stationary communication system. As such, the transmission and reception of the signals in reality are the familiar method used in ordinary communication systems. Although this technique may be easy to understand, it is not easy to implement within the constraints of a commercially viable system.
The WAVES advanced digital wireless communication network communicates audio and data for the purpose of PA systems. The system contains a base station and any number (from 1 to 16,384) of autonomous nodes that process the information received on the network or from local inputs and delivers it to input/output devices such as speakers, visual displays and other serial data devices. Figure 2 shows the base station's individual modules and interface capabilities. The entire system has a plug-and-play add/remove modular configuration, which uses an advanced mesh networking concept that permits any unit to also function as a relay. Thus, any size or shape facility can be covered reliably. This concept is called Total-Site-CoverageTM . The bidirectional communication between all the units is by SSFH. Efficient system design and the inherent capabilities of SSFH allow systems to operate properly even within a heavy RF environment. All the information is digital including the audio streams. The WAVES system is a real-time control system, which enables the operator to fully control all system audio and data parameters as well as to monitor each unit's operational parameters in real time.
Fig. 2: A WAVES base station.
The convergence of three technology developments has made the successful implementation of WAVES possible. In 1985, the FCC changed Part 15 regulations, enabling license-free use of spread spectrum wireless devices in the industrial, scientific and medical (ISM) band, including 2.4 to 2.4835 GHz. Once considered mission-critical, interference-immune military RF only, the technology has powerful capabilities unknown previously in commercial and professional audio business. Also significant was the advent of digital audio components of professional quality (brought on by the CD revolution), and the appearance of lower cost, digital signal processing (DSP) engines dedicated to audio. Finally, much like the PC revolution, powerful low cost microprocessors have decentralized the processing burden, with each node in the distributed network providing local intelligence and network services. Digital wireless networks have made possible powerful new command and control capabilities, such as node addressing and zoning functions in a heavy multi-user environment.
The WAVES SSFH wireless audio and data distribution system operates in the ISM band at 2.4 to 2.4835 GHz. In the system, each node acts as an independent PA system with all the necessary functions. The remote operator has full real-time control of all the audio and data parameters at each node independently. The control information is communicated to the nodes wirelessly, together with the audio and data channels.
From the base station, the WAVES system transmits two high fidelity audio channels multiplexed with an asymmetric bidirectional data channel. This unique feature allows the user to transmit two independent audio messages and synchronize either one with a visual display message running on the data channel. This feature was prompted by McDonnell Douglas' requirement for both voice messages to the manufacturing floor and background music to the offices. The system's ability to synchronize and output audio and visual messages simultaneously coincides with the ADA requirements for the hearing impaired (approximately 10 percent of the population who cannot benefit from any audible PA system).
Another unique feature of the WAVES system is its real-time control, monitoring and reconfiguration capabilities. Figure 3 shows how, in a typical network configuration, the operator can direct the information flow selectively by arranging the nodes in any required zone layout. Moreover, the base station can store a large number of presets, specific network layouts and settings that enable it to operate instantly at predetermined conditions and situations. Each preset also includes individual local unit audio parameters and can be activated by a built-in event scheduler.
Fig. 3: A typical network configuration.
One of the most important features in the WAVES system is its inherent fault-tolerance design, achieved by the fact that every unit communicates with two other units in the network. Thus, when a particular node is down it does not affect the flow of information, which is rerouted automatically, as shown in Figure 4 .
Fig. 4: A WAVES network showing frequency management with relays.
The system offers a robust solution to multipath problems by implementing a sophisticated antenna diversity system. In the subway example, despite the high ratio of received reflection signals, reception within the long tunnel-like structures was continuously good and was not affected when the 600-foot-long metallic electric trains rolled in and out of the station.
The WAVES system consists of two basic units. The base station, with the central unit (CNU), audio interface (ADU) and base transceiver (CRLU), controls the network, interfaces and processes the audio and data sources, and transmits and receives to the network. The programmable wireless transceivers can be programmed to function as receive only, relay with local output or relay only.
To achieve relay capabilities in the network structure with a relay approach, the unit's frequencies are organized such that collisions will not occur. Furthermore, because of the system's SSFH configuration, collision is guaranteed not to occur when different frequency hopping sequences are chosen properly. The blocks under each relay describe the offset increments between the transmit and receive frequencies needed to avoid collision. The fault-tolerance capability of the network can be recognized because each of the relays has the capability to receive from two separate sources.
Transferring information between zones or from the CNU to the zones is performed by the relay units. Each zone consists of various combinations of relays and receivers. According to the system's hierarchical structure, a zone is considered a grouping of units (relays and receivers) within a physical location and is defined by the user according to the information addressing requirements. A subzone is a physical location that is part of a zone, where system units are positioned and can be addressed as a single group. A unit is defined as a transceiver programmed as a relay (RLU) or receiver (RXU). Each unit is addressable individually. The network can address the units in one of several modes: individually, grouped in zones and subzones, or all (broadcast).
Certain applications require networking within the network, such as in airports with central and local gate announcements. This network with satellite base station configuration is shown in Figure 5 . At each satellite entry point, a central unit with audio-processing capability enables duplicating audio and data inputs just as in the network's base station. This unit is fully synchronized to the master network yet controls announcements made locally. The satellite base stations also function as gates to the network and control the local audio according to a programmable priority setting. This type of configuration can also be used as a fault-tolerant backup approach to emergency conditions where the satellite base station becomes the network's main base station. When this condition occurs, the network tree changes its configuration around the new trunk.
Fig. 5: A WAVES network with a satellite base station.
An emergency configuration of the WAVES system is shown in Figure 6 . The emergency layout can be triggered by the operator or activated from call boxes. When activated, the system switches to a preprogrammed emergency configuration that can include preprogrammed announcements, data messages and other control data. Activation can occur from any zone or from a preprogrammed schedule. For example, it is assumed a fire broke out in a certain location in the facility and was detected by a sensor monitored by a local WAVES unit. In turn, the WAVES unit communicates this information to the base station and satellites. This emergency condition triggers all the appropriate alarm and security actions sequences according to a preprogrammed response plan.
Fig. 6: A WAVES network with emergency call boxes.
The Base Station
The base station CNU is a PC running WindowsTM '95 or NT. Touch-screen panels and dual-tone multiple frequency control modules are also possible. The control software is an integrated database and a graphic user interface enabling the user easy and intuitive control of the system.
The Wireless Unit
The wireless unit's major elements are the transceiver and digital units, antennas, battery backup unit (optional), Motion Pictures Expert Group audio encoder (present only in base station or backup units), built-in test (BIT) unit, audio unit, ambient noise-sensing microphone (optional), PC 104 interface and space for an additional PCB. The transceiver's block diagram is shown in Figure 7 .
Fig. 7: A WAVES transceiver functional block diagram.
To make the case for wireless public information systems, the advantages of digital wireless distribution systems are summarized. Beyond the fact that the wireless configuration eliminates the wiring and conduit, saves on installation time and minimizes the disruption associated with a hard-wired installation, the wireless WAVES system has other significant advantages over conventional analog, hard-wired systems.
In wired systems it is almost always necessary to transform the audio signals to 70 V to prevent losing power excessively on the long wire runs. By code, this transformation requires the use of conduit for the wiring. Often, the resulting spaghetti of wiring and conduit and the high maintenance associated with them become the nightmare of every PA system operator. The cost of wiring in any installation is a major portion of the system cost, and the problems associated with it can be numerous. For example, while installing the new PA system in the Pittsburgh airport, a mistake was made by the contractor in the wiring and was discovered during a later stage. This mistake caused a delay of a few months in the completion of the work as well as high overrun costs.
Acoustics and audio quality are always major painful problems in any installation. Acoustic problems that exist in airports, train stations, malls and any other facility that has a PA system are well known. Clean, clear sound is the rare exception. The two main reasons for such poor results are the averaging equalization approach and the less than optimal choice of locations for the speakers.
Maintenance is always a major issue with PA systems. Because PA systems are distributed over large areas, they are difficult to maintain and problems are difficult to detect. The WAVES system incorporates a self-supervision system. Each node has a BIT function that tests continuously for the functionality of that node and reports back its status to the base station. Two kinds of BIT tests are performed. At level one, which covers most of the critical functions, tests are performed constantly. This test is initiated by the unit. A full BIT test that is performed routinely is triggered by demands from the base station. The reports of the system's functionality are made available in real time and any malfunctioning of a unit is pinpointed immediately for easy maintenance. This added reliability in maintenance is of major importance in the selection of a PA system.
Another advantage is the system's modularity. Generally, wired PA systems require that the whole system be installed in one (painful) session that may last weeks or months. PA systems are normally considered a capital expense. Only the WAVES system, with its inherent design modularity (since units can be added or their location changed as needed), permits the system designer to decide on a specific layout without the fear of making a major costly mistake. Many facilities often reconfigure the facility and its PA zones and can benefit significantly from the real-time wireless rezoning and layout reconfiguration feature of the WAVES system.
The ADA requires that people with disabilities be provided with comparable and equivalent accommodations in places of public gatherings (public and private). For hearing-impaired individuals, visual displays must be provided in place of an audible PA. WAVES provides this capability and enables the simultaneous transmission of voice and visual messages. The visual message is sent on the data channel and is synchronized with the audible message.
Facilities are becoming more and more computerized. This trend toward integrated systems requires sensors and transducers to be located around the facility with connection to a central location. This new type of facility requires new wiring for each system. The WAVES system allows the transfer of data through its data channel capabilities. Using an integrated PA system design can solve major present and future problems for the facility's manager.
PA systems are informative for regular daily operation and act as a life-safety tool in facility evacuating and emergency notification. Both uses require a PA system with intelligible audio. But in an emergency, the system's survivability is of major concern. Typically, daisy-chained wired PA systems collapse in emergency situations (explosions, fire, floods, earthquakes and storms). This system breakdown is because the weakest link in the system is the wire runs. The wireless system with fault-tolerance capability has a higher probability of surviving and continuing to function in these conditions.
The WAVES wireless audio visual and emergency system was designed with a complete system approach making it the first time that the cooperation of experts in the specific application areas was used in defining such a system. The resulting system provides practical solutions to a large number of applications from entertainment and emergency response to stationary and portable facility operations. The system integrates the most advanced technologies (wireless, SSFH, digital audio and DSP) to deliver a complete solution. The WAVES system also is helping to catapult this rather unsophisticated PA business into the 21st century. New applications are being discovered every day and the response from the industry is very positive.