Introduction

CD-Digital Audio (CD-DA) is the standard format for audio compact discs (CDs). CD-DA is a prerecorded read-only format.

Physical characteristics of compact discs

Disc	Description
Playing time	74 minutes, 33 seconds maximum
Rotation	Counter-clockwise when viewed from readout surface
Rotational speed	1.2-1.4 m/sec.
Track pitch	1.6 µm
Diameter	120 mm
Thickness	1.2 mm
Center hole diameter	15 mm
Recording area	46 - 117 mm
Signal area	50 - 116 mm
Material	Any transparent material with 1.55 refraction index. Usually polycarbonate
Minimum pit length	0.833 µm (1.2 m/sec.) to 0.972 µm (1.4 m/sec.)
Maximum pit length	3.05 µm (1.2 m/sec.) to 3.56 µm (1.4 m/sec.)
Pit depth	Approximately 0.11 µm
Pit width	Approximately 0.5 µm
Number of channels	2 channels (four channel recording possible)
Quantization	16-bit linear quantization
Quantizing timing	Concurrent for all channels
Sampling frequency	44.1 kHz
Channel bit rate	4.3218 Mbps
Data bit rate	2.0338 Mbps
Data-to-channel bit ratio	8:17
Error correction code	Cross Interleave Reed-Solomon Code (CIRC) (with 25% redundancy)
Modulation system	EFM

Source

Ken C. Pohlmann, The Compact Disc: A Handbook of Theory and Use. Madison, Wisconsin: A-R Editions (1989): 49.

Layers of CD-DA

Image source

Ken C. Pohlmann, The Compact Disc: A Handbook of Theory and Use. Madison, Wisconsin: A-R Editions (1989): 50.

The following table describes the process of creating a pre-recorded CD:

Image source

Replication of CDs. Image source: J.M. Eargle. The Compact Disc (CD). In Handbook of Recording Engineering. Boston: Springer (1996): 465.

Technical specifications

Specification	Description
Channels	2 channel (stereo)
Bit-depth	16-bit
Data encoding	Linear PCM
Sample rate	44.1 kHz

The sample rate of 44.1 kHz was inherited from the widespread practice of using a PCM adaptor to convert digital audio into an analogue video signal for storage on a U-matic videocassette. This was the most economical way of transferring audio from the recording studio to CD manufacturers.

Technical specifications for CD-DA are defined in the "Red Book" (IEC standard 60908:1999).

Emphasis

Some CD-DAs produced in the early 1980s are encoded with a form of signal boosting called pre-emphasis. High frequencies on the master tape are boosted slighty. On playback, the pre-emphasis process is reversed ("de-emphasis") so the output accurately represents the original input. Emphasis produces slight improvement in the signal-to-noise ratio.

Pre-emphasis is denoted by a flag in the subcode. Emphasis should be considered during forensic imaging or "ripping" files from CD-DAs so digital audio files do not play back with a distorted frequency response.

Data structure and storage

Data frame structure

Data on CD-DAs is encoded in "frames" with a defined structure that allows the CD player to distinguish the types of data on the disc. Each frame includes 24 bytes of audio data, and some additional overhead data.

Prior to EFM modulation, the information contained in one CD-DA frame includes 291 bits:

27 bit synchronization word
8-bit subcode
192 audio data bits
64 parity bits

Image source

Ken C. Pohlmann, The Compact Disc: A Handbook of Theory and Use. Madison, Wisconsin: A-R Editions (1989): 128.

Frames are assembled through a complex series of steps:

Step	Description
Group audio data	Six 32-bit PCM audio sampling periods are grouped into a frame. Each sampling period is derived from alternating 16-bit left and right audio channels. The sampling period is divided to produce four 8-bit audio symbols.
Error correction encoding	Error correction encoding takes place using the Cross Interleave Reed-Solomon Code (CIRC). CIRC uses a combination of interleaving and parity bits.
Subcode	An 8-bit subcode symbol is added to each frame. Subcode bites are designated as P,Q,R,S,T,U,V, and W. Only the P and Q bits are used for audio encoding and include infromation such as the total number of tracks on a disc, beginning and end points, and other timing information. The other six bits are available for encoding other information.
EFM modulation	After the audio, parity, and subcode data are assembled, the bit stream is modulated using eight-to-fourteen modulation (EFM). 8-bit blocks are translated into 14-bit blocks known as "channel bits." EFM modulation helps achieve greater storage density and provides a form of error corretion because it allows for 16,384 unique bit patterns (rather than 256 patterns allowed for with 8-bit encoding).
Merging bits	14-bit blocks are linked by three merging bits.Two 0-bits are always used to prevent violations of the coding scheme and a third merging bit (either a 0 or 1) is added to assist with clock synchronization.
Pit length	The signal is converted from a "non-return-to-zero" (NRZ) signal to a "non-return-to-zero-inverted" (NRZI) signal. NRZI signals simplify the pit structure of the disc. The final NRZI signal produces pits and lands between three channel bits (720 kHz) and eleven channel bits (196 kHz) long.
Frame assembly	24-bit synchronization word is inserted between each frame. The synchronization word delineates each frame.

After data encoding, each frame contain 588 bits:

24 synchronization bits
336 data bits (12 x 2 x 14)
112 error correction bits (4 x 2 x 14)
14 subcode bits
102 merging bits (34 x 3)

Source

Adapted from Ken C. Pohlmann, The Compact Disc: A Handbook of Theory and Use. Madison, Wisconsin: A-R Editions (1989): 72-80.

Data storage and transmission

Note: Data on a CD-DA is written as a linear pulse-code modulation (PCM) stream. The ISO 9960 file system is not used in CD-DAs.

CD-DA data is addressed on the compact disc using a timecode system of minutes, seconds, and frames, expressed in mm:ss:ff. Timecode frames (i.e., sectors) are the smallest addressable time intervals on a CD-DA and correspond to 1/75th of a second of audio. Timecode frames are distinct from the data frames described above. Each sector can hold 98 data frames, or 2,352 bytes of audio data. The CD is played at 75 timecode frames per second, or 176,400 bytes per second (1,411,200 bit/s). In other words, CD-DA data transfers at a bit rate of 1.4 Mbit/s.

CD-DAs hold more data per sector than CD-ROMs, but the increased storage capacity also introduces a higher error rate. As a result, traditional forensic imaging techniques that involve a single pass over the disc are not suitable for CD-DAs. Instead, software designed for CD-DA extraction is used to convert tracks into standard computer.

The volume of storage required for CD-DA data can be calculated using the bit rate. The total size in bits of a CD-DA stream can be calculated as: 2 x 16 x 44,100 x length of time in seconds

The total number of bytes can be calculated by dividing the size in bits by 8 (the number of bits in one byte): size in bytes = size in bits/8

Example: calculate storage requirements for 30 minutes of CD-DA data

CD-DA with the following characteristics:
2 channels
16 bit
44.1 kHz sample rate
30 minutes (1,800 seconds)
2 x 16 x 44,100 x 1,800 = 2,540,160,000 bits of storage
2,540,160,000 bits divided by 8 = 317,520,000 bytes
317.52 MB

Technical standards

Red Book - IEC standard 60908:1999. Audio recording - Compact disc digital audio system.