Design and implementation of high performance digital audio transmission system

With the rapid development of electronic technology and digital technology, digital audio has been widely used in various application fields such as recording, broadcasting, and transmission of radio and television. In many cases, analog audio can no longer meet the most basic requirements of the entire sound reinforcement system. The most critical issue in the design of a large stadium sound reinforcement system is how to solve the long-distance and high-quality transmission of weak audio signals. For large venues, the distance to be transmitted is usually several hundred meters. Using traditional analog transmission methods, it is difficult to solve problems such as signal loss, electromagnetic interference, and ground interference. The various performances of digital audio are far superior to analog modes, so the digitization of radio and television equipment has become an inevitable trend. The advantage of using digital signals for transmission and processing is that digital signals are not sensitive to interference, and the signal-to-noise ratio and distortion of the entire system are independent of the transmission distance. For long-distance transmission, its excellent performance indicators are unmatched by analog transmission.

At present, both radio and television studios are moving towards digitalization. As the main functional means of digital TV production, the basic theory, interface mode, audio format and system design of digital audio have also become major issues in the field of radio and television program production. However, most of the current high-performance digital broadcasting and transmission equipment are imported equipment and are expensive. This paper studies and designs a high-performance digital audio transmission system for use in this field.

2 Digital Audio Interface Standards

At present, the commonly used digital audio interface standards include AES/EBU (AES3 - 1992) interface, S/PD IF interface, MADI interface, etc. S/PD IF is mainly used as a civilian digital audio format standard, and MADI interface is based on the dual-channel AES/EBU interface. In the field of professional digital audio, the AES/EBU interface standard is mainly used.

The full name of AES/EBU is Audio Engineering Society/Eu2ropean B roadcast Union (Society of Recording Engineers/European Broadcasting System Union), which has become a popular standard for professional digital audio. A large number of civilian products and professional audio digital equipment such as CD players, DAT, MD players, digital mixing consoles, digital audio workstations, etc. support AES/EBU.

The AES/EBU standard is a digital audio transmission standard developed by AES and EBU. It is a digital device interface protocol for transmitting and receiving digital audio signals. It stipulates that audio data must be encoded in 2's complement. The transmission medium is a cable, which allows high bandwidth capacity and serial transmission of parallel data bytes generated by A/D converters. When transmitting 16-20 bit parallel bytes serially, the least significant bit is transmitted first. A byte clock mark must be added to indicate the start of each sample value. The final data stream is encoded in a biphase mark code. In addition, the clock information is also embedded in the AES/EBU signal stream.

AES/EBU transmits digital audio data over a single twisted pair of wires using a serial bit transmission protocol that can transmit data over a distance of up to 100 m without equalization. It provides two channels of audio data (up to 24 bits quantized), and the channels are automatically timed and self-synchronized.

It also provides transmission control methods and status information (channel status bit) and some error detection capabilities. Its clock information is controlled by the transmitter and comes from the AES/EBU bit stream.

Common physical connection media for AES/EBU are: (1) Balanced or differential connection, using a three-core shielded microphone cable with an XLR (Cannon) connector, with an impedance of 110Ω, a level range of 0.2 to 5 Vpp, and a jitter of ±20 ns. (2) Single-ended unbalanced connection, using an audio coaxial cable with an RCA plug. (3) Optical connection, using a fiber optic connector.

Since its revision in 1992, AES/EBU has been widely used in recording production, digital cinema and broadcasting and television industries, becoming the most common digital audio format. Related equipment, interfaces, cables, accessories, etc. are all available and inexpensive.

3 System Circuit Design

3.1 Overall plan of the system

The entire digital audio transmission system is divided into three parts: the transmitter, the receiver and the transmission medium (cable), as shown in Figure 1. The transmission medium mainly includes twisted-pair shielded cable, coaxial cable, optical fiber and wireless transmission (such as PDH or SDH digital microwave), which is selected according to the specific occasion and transmission distance.

Figure 1: Schematic diagram of digital audio transmission system.

The transmitter is mainly responsible for completing signal access, A/D conversion, format encoding, clock generation, etc. In order to increase the dynamic range of the signal and prevent aliasing distortion in A/D conversion, signal conditioning circuits and anti-aliasing filters should be set in the analog input channel.

The receiving end is mainly responsible for receiving and decoding AES/EBU format data, recovering the main clock signal and synchronization signal, and then performing D/A conversion on the audio data.

3.2 Transmitter circuit design

According to the system solution described in the previous section, we use Cirrus Logic's CS5381 and CS8406 to complete the A/D conversion and AES/EBU format encoding and transmission of analog signals respectively. The circuit principle is shown in Figure 2.

Figure 2 Schematic diagram of the transmitter.

CS5381 is a 120 dB, 192 kHz high-performance 24-bit stereo analog-to-digital conversion chip launched by CirrusLogic. CS5381 can work in master and slave modes. The mode selection can be made through pin 2 (M/S). This design works in master mode. The sampling rate of CS5381 can be controlled by the logic levels of the three pins MD IV, M0 and M1. The master clock selection can be selected according to the selected sampling frequency and MD IV pin. In this design, a single-speed sampling rate of 48 kHz is selected, and a 12.288MHz active crystal oscillator is used as the clock source. The conversion result of CS5381 is 24-bit two's complement serial data, and the left and right channels are output alternately, which can be distinguished by the high and low levels of LRCK. There are two formats for output data, namely left-aligned and I2S format. This design uses the I2S format.

The digital audio format encoding and transmission is completed by Cirrus Logic's digital audio transmitter CS8406. The CS8406 can support 192kHz sampling rate and meet the next generation audio format. It can receive and encode audio and digital data, and then transmit them to the cable/optical fiber interface after multiplexing and encoding.

The device's operating mode is selected as hardware mode (H /S = 1), the input data format is I2S (SFMT1 = 0, SFMT0 = 1), and the main clock frequency OMCK is selected as 256 × FS (HWCK1 = 1, HWCK0 = 1). IL2RCK, ISCLK, and SD IN are the left and right clock signals, serial clock signals, and audio data from CS5381; TXN and TXP are serial data output terminals, and the encoded AES/EBU format data is sent out through these two pins.

[page]

3.3 Receiver circuit design

For the receiving circuit, we use the digital audio receiving circuit CS8416 corresponding to CS8406 to complete the reception and decoding of AES/EBU format audio data, and use CS4397 to complete the conversion of digital audio signals to analog signals. The circuit principle is shown in Figure 3.

Figure 3 Schematic diagram of the receiving end.

The CS8416 is an industry-leading 192 kHz digital audio receiver with extremely low jitter performance of 200 ps. The CS8416 receives and decodes digital audio data sampled up to 192 kHz and uses an extremely low jitter clock recovery device to generate a clean recovered clock from the incoming audio stream. An 8:2 input multiplexer allows up to eight digital audio input sources, and the second output of the multiplexer provides an SPD IF pass-through feature for added system flexibility. The CS8416 integrates automatic detection of compressed audio input streams and CD-Q subcode decoding, and allows the signal to be selectively routed to three general-purpose output (GPO) pins.

Working in software mode, 8 digital audio signals can be connected to CS8416 at the same time. When SDOUT is connected to ground with a 47 kΩ resistor, the device works in hardware mode. At this time, RXP4, RXP5, RXP6, and RXP7 will work in the second function, and they are used to set the selected RXP0, RXP1, RXP2, and RXP3 as receiving pins. In this design, there is only one synthesized left and right channel digital audio signal at the receiving end, so we choose RXP0 and RXN as receiving pins (correspondingly set RXP4= 0 RXP5 = 0), and other unused receiving pins are left floating; AD0 is the signal reception confirmation pin, which is connected to a light-emitting diode. When no signal is received, the light-emitting diode is on, and when a signal is received, the light-emitting diode is off. OLRCK, OSCLK, and SDOUT are the left and right clock signals, serial clock signals, and audio data extracted from AES/EBU data.

AUD IO is the non-audio data stream indicator pin, and is also the input data format selection bit SFSEL1; C (pin 19) is the channel status indicator bit, and is also the input data format selection bit SFSEL0. These two pins can determine the output data format by connecting them to ground or high level through a 47kΩ resistor. The connection method in Figure 3 selects the I2S 24-bit data format. U is the user data bit, which is connected to ground through a 47kΩ resistor and selects the restored master clock frequency MRCK to 256 ×FS.

The CS4397 is a complete, high-quality 24-bit 48/96/192 kHz stereo digital-to-analog conversion chip launched by Cirrus Logic.

In the figure, LRCK, SCLK, and SDATA of CS4397 are the left and right clock signals, serial clock signal, and audio data pins, respectively, which are directly connected to the corresponding pins of the digital audio receiving circuit CS8416 to receive the decoded digital audio signal. The C/H pin is connected to a low level when the device works in hardware mode. The M4~M0 pins are used to set the sampling frequency and input data format. The connection method in the figure selects a 48 kHz single-speed sampling frequency and I2S24bit data format input. A INL +, A INL -, A INR +, and A INR- are the output terminals of the in-phase and anti-phase signals of the left and right channels after D/A conversion, respectively, and are connected to the low-pass filter circuit composed of NE5532 to filter out high-frequency components above 20 kHz.

3.4 Transmission Media

As mentioned above, there are four main ways to transmit digital audio signals: twisted shielded pair cable transmission, coaxial cable transmission, optical fiber transmission and wireless transmission. The first three methods are standard transmission methods recommended by AES/EBU. Wireless transmission can use frequency modulation or dedicated digital microwave channels, such as the PDH digital microwave E1 interface. However, due to the inconsistent transmission rates of the E1 interface and the AES/EBU standard, the AES/EBU digital audio signal needs to be adjusted to make it suitable for the E1 interface.

At present, there is a new way to transmit digital audio, which is to use audio embedding technology to transmit through TV channels, that is, to insert digital audio signals into the line and field synchronization pulses (line and field blanking) of the video signal and transmit them simultaneously with the digital component video signal. Audio embedding technology can combine audio and video signals that had to be transmitted separately in the past into one video channel for transmission, thereby greatly simplifying the amplification and switching processing equipment required for audio and video interconnection in the studio, and can realize the synchronous transmission and playback of audio and video, which is also an important application of digital audio in the field of digital television.

4 System Testing and Conclusion

Figure 4 is a physical picture of the digital audio transmitter and receiver. The transmitter and receiver are connected by coaxial cable, and the entire transmission system is tested using the HP HP8903B audio tester. The main technical indicators are as follows:

Figure 4: Digital audio transmitter and receiver circuit diagram.

(1) Frequency response: Unflatness < ±0.1 dB within 20 to 20 kHz.

(2) Signal-to-noise ratio: >90 dB in the entire frequency range, >94 dB at 1 kHz.

(3) Distortion: <0.1%.

(4) Dynamic range: >90 dB.

At present, the digital audio transmission system has been used in a radio station, which is connected to PDH digital microwave through E1 interface to replace the original analog audio system. It has stable performance and remarkable effect. The digital audio transmission system will be further widely used in the digital transformation of radio and television broadcasting transmission equipment and related sound reinforcement systems at the city and county levels.