Motion video signal processing for recording or reproducing – Local trick play processing – With randomly accessible medium
Reexamination Certificate
1998-06-26
2002-01-22
Tran, Thai (Department: 2615)
Motion video signal processing for recording or reproducing
Local trick play processing
With randomly accessible medium
C386S349000
Reexamination Certificate
active
06341198
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention generally relates to the field of digital communication, and particularly relates to the parsing of packets from a data stream.
2. Description of the Related Art
Digital data are generally transmitted in contiguous groups called packets. A packet in a data stream typically starts with a number of bits called a header that provides information on the packet's characteristics such as its length, the type of data it carries, and perhaps error-checking information and the destination of the data, among other characteristics. The header of a data packet is typically followed by a payload section of data, which comprises the data to be communicated by the packet. The packetization of data provides several advantages. Primary among these is the robustness of the data transfer. The finite length packets are typically independent of each other, therefore, errors in one packet do not propagate to cause errors in data from other packets. Packetization also permits multiplexing, allowing multiple users to share a single communications resource. This flexibility also allows a single data stream to carry more than one type of data. For example, some data formats in common use for the transfer of video and audio information are the MPEG standard formats (such as MPEG-1 and MPEG-2) defined by the Moving Picture Experts Group, a working group of ISO.
FIG.
1
: Packetized Data Stream
FIG. 1
shows a sample data stream comprising audio and video packets from an MPEG data stream. Although this discussion particularly describes an MPEG data stream, the structure described in this figure is equally applicable to other interlaced digital data formats.
As shown in the figure, a packetized elementary stream (PES)
100
comprises a stream of audio and video PES packets. The PES data stream is thus a single data stream comprising two interlaced elementary streams (ES): an audio ES and a video ES. The PES stream comprises a series of audio PES packets and video PES packets. Each audio PES packet comprises a header portion and an audio data portion, and each video PES packet comprises a header portion and a video data portion. When read in sequence, the audio data portions can be recombined into the audio elementary stream
110
. Similarly, the video data portions can be read in sequence and recombined into the video elementary stream
120
. Although only two types of elementary streams (audio and video) are shown in this figure, the MPEG-2 specification provides for four different types of PES packets: MPEG video (a compressed video format), MPEG audio (a compressed audio format), private stream
1
, and private stream
2
. The two private streams allow the MPEG-2 definition to be used in specialized systems that communicate information other than the standard MPEG audio and video. Each of the four types of PES packets has a predefined format for its header. A parameter in the PES header, called the stream_id, identifies the packet as being one of these four types of PES packet. Thus by reading the stream_id from the PES packet header, a receiving unit may rapidly identify which type of packet is being received in the data stream, and respond accordingly. For example, in the case of an audio-video decoder that does not use the private stream
1
or private stream
2
information, a parser in the decoder may read the headers of each PES packet received in the data stream, and after identifying the packet according to its stream_id, direct the payload data so that video data portions are sent to a video buffer, audio data portions are sent to an audio buffer, and private stream
1
and private stream
2
data are discarded.
A particular technology using the MPEG-2 data format is the digital versatile disk (DVD) for optical disk storage technology. DVD storage systems have widespread use in the storage and communication of audio and video data, particularly in home entertainment use. In addition to holding MPEG audio and video, the DVD standard specifies formats other types of information, such as other types of audio streams, sub-picture information for display at the bottom of the video screen, and navigation information to enable interactive usage. These other types of information are also transferred in packets in the MPEG 2 data stream, interleaved with the MPEG audio and MPEG video data packets. The additional data are placed in custom packets of various types. There are custom data packets for Dolby AC-3 audio, DTS audio, SDDS audio, linear PCM audio, PCI data, and DSI data. In addition, other types of custom data packets are reserved for future use, such as extended sub-picture data. All these different types of custom data packets must be formatted into either private stream
1
or private stream
2
data packets to be included in an MPEG-2 PES. To distinguish among these various types of data packets, identifying information on them is included in the beginning of each data portion of the private stream data packets. This identifying information, though it is included in the data portion of the packet, effectively comprises an additional header portion for the private stream
1
and private stream
2
data packets in the DVD specification.
FIG.
2
: AC-3 Audio Packet in a DVD Data Stream
An example of such a custom data packet is shown in FIG.
2
. As described earlier, the first part of the packet is a private stream
1
packet header
251
, and the following information is the private stream
1
data
252
. The private stream
1
packet header
251
comprises a number of fields. The first field is the “packet_start_code_prefix”
201
, a predefined 24 bit sequence of 23 0's followed by a trailing
1
(or in hexadecimal: 00 0001h). This unique sequence identifies the start of a PES packet, whether the PES packet is an MPEG audio, MPEG video, private stream
1
or private stream
2
packet. The next field is the stream_id
202
. As shown in the following table, the stream_id uniquely identifies which of the four types of MPEG 2 packets is being received. Note that in this discussion, a number with suffix “h” is a hexadecimal number, while a “b” suffix indicates a binary number.
stream_id
Stream coding
110x 0nnnb
MPEG audio stream (for Decoding Audio
stream number nnn)
1110 0000b
Video stream
1011 1101b
private_stream_1
1011 1111b
private_stream_2
By way of example,
FIG. 2
shows the stream_id
202
to be the binary number 1011 1101b, indicating that the packet is a private stream
1
packet. The field following the stream_id
202
is the “PES_packet_length”
203
, which indicates the total length of the PES packet. The following two bytes in the private stream
1
header comprise various flags
204
describing the packet. Following these flags is the “PES_header_data length”
205
, which indicates the number of remaining bytes before the end of the PES header. This field is a one byte field, so in principle between 0 and 255 bytes may remain in the packet header. According to the DVD specification, however, between 8 and 15 further bytes of data would be included at the end
206
of the header for this packet. After the last byte of the private stream
1
packet header begin the private stream
1
data bytes. Note that the private stream
1
packet header has identified the packet as being a private stream
1
packet. There has been no indication in the header
251
, however, of the type of data to be carried in the payload of the packet. This identification is carried in the first byte of the private stream
1
data
252
: the “sub_stream_id” field
210
. The sub_stream_id
210
, along with other payload description at the beginning of the private stream
1
data portion
252
effectively make up a sub header
253
of additional information for the private stream
1
sub-packet.
By way of example,
FIG. 2
shows a data packet carrying Dolby AC-3 audio data. Thus the one-byte sub_stream_id
210
for this field would be in the range between the hexadecimal numbers 80h and 87h, corresponding to a binary number of 10000
Gouzu Manabu
Mendenhall Todd C.
Conley Rose & Tayon
LSI Logic Corporation
Onuaku Christopher
Tran Thai
LandOfFree
System for byte packing multiple data channels in an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for byte packing multiple data channels in an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for byte packing multiple data channels in an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2831542