Coded data generation or conversion – Digital code to digital code converters
Reexamination Certificate
1998-06-16
2001-03-20
Tokar, Michael (Department: 2819)
Coded data generation or conversion
Digital code to digital code converters
C341S067000
Reexamination Certificate
active
06204780
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and circuitry for compressing and decompressing digital data.
More particularly, the present invention relates to a method and circuitry for respectively compressing and decompressing digital data to and from a frequency domain representation.
One particular application, inter alia, or the present invention is in MPEG-2. While this document will focus on MPEG-2 as an application of the present invention, it should be understood that the MPEG-2 application is to be considered as an exemplification of an application of the present invention and that there is no intention of limiting the applications to which the present invention can be applied.
2. Background Art
Digital video decoders such as MPEG-2 require up to two previously decoded frames to generate the frame currently being decoded.
Each frame consists of three color components: luminance Y and two chrominance U, V. Each color component may be stored as a single frame or two fields, depending upon the implementation.
It should be noted that throughout this entire text, unless otherwise stated, one Y, U or V component pixel is represented by one byte and 1 KByte represents 1024 bytes.
The luminance component Y for each ‘PAL standard’ frame comprises 720×576 pixels per frame, which therefore requires 405 KBytes of memory per luminance component Y per frame. Each of the two chrominance components U and V for each ‘PAL standard’ frame comprises 360×288 pixels per frame, which therefore requires a total of 2×101.25 KBytes (202.5 KBytes) of memory per chrominance components U, V per frame. Therefore, each ‘PAL standard’ frame requires 607.5 KBytes of memory in order to store the luminance Y and chrominance U, V information.
It should be understood that even though the ‘PAL standard’ has been referenced above, the principle also applies to other standards, for example NTSC.
As stated earlier, in MPEG-2 there is a requirement for storing up to two previously decoded frames, i.e. a total of 1215 KBytes, in addition to the frame currently being decoded, i.e. 607.5 KBytes. Therefore, there is a requirement for storing up to 1822.5 KBytes of data in memory.
MPEG-2 as well as many other video/image compression methods and apparatus, for example, JPEG, H.261 and H.263, divide each of the three colour components Y, U and V into 8×8 blocks and then transform each block into a representative frequency domain block using an 8×8 two-dimensional Discrete Cosine Transform (DCT). Each frequency domain coefficient block is then quantized with the higher frequency coefficients being quantized to a greater extent than the lower frequency coefficients. This therefore results in a loss of information with many of the high frequency coefficients typically having values that become zero. The quantized coefficients are then run-length encoded using a zig-zag pattern and finally Huffman encoded. The resulting compressed block is thus not of a fixed size.
One disadvantage of the state of the art MPEG decoders is that they require two mega-bytes (2 MBytes) of memory in which to store the data. Another disadvantage of these MPEG decoders is that they have high memory bandwidth requirements. The more the memory requirements are, the more expensive, both in terms of cost, space and efficiency a system becomes.
OBJECTS & SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to overcome the disadvantages and inconveniences of the state of the art.
Another object of the present invention is to reduce the amount of memory required for storing digital data.
Another object of the present invention is to compress digital data as it is written to memory, or transmitted, and decompress compressed digital data when read from memory, or received, using simple operations.
Another object of the present invention is to provide a variable compression/decompression ratio.
Other objects of the present invention are to keep to a minimum the read/write latency and to reduce the memory or transmission bandwidth requirements.
Another object of the present invention is to fix the size of the compressed data so as to enable simple address calculation for random memory access.
In order to achieve these objects, the present invention proposes circuitry for processing digital data items. The processing circuitry could, for example, be incorporated as part of an MPEG decoder for processing digital video/image data items. The circuitry further comprises compression circuitry and decompression circuitry, said compression circuitry comprising: a circuit for receiving and transforming M first data items into N second data items, where M is an integer greater than or equal to two and where N is an integer less than or equal to M; a circuit for receiving and quantising P second data items and producing Q third data items, where P is an integer less than or equal to N and where Q is an integer less than or equal to P; and a circuit for appropriately storing in memory and/or transferring R third data items; where R is an integer less than or equal to Q, said decompression circuitry comprising: a circuit for appropriately retrieving from memory and/or receiving S first data items, where S is greater than or equal to R; a circuit for receiving and dequantising T first data items and producing U dequantised second data items, where T is greater than or equal to S and U is greater than or equal to T; and a circuit for receiving and inverse transforming V second data items into W third data items, where V is greater than or equal to U and where W is greater than or equal to V, said W third data items being representative of said M first data items.
According to another embodiment of the present invention, the compression circuitry further comprises a circuit for truncating the N second data items and producing N′ second truncated data items, where N′ is an integer less than or equal to N and where P is an integer less than or equal to N′.
According to another embodiment of the present invention, the compression circuitry further comprises a circuit for range clamping the Q third data items and producing Q′ third range clamped data items, where Q′ is an integer less than or equal to Q and where R is an integer less than or equal to Q′.
According to another embodiment of the present invention, the decompression circuitry further comprises a circuit for range clamping the W third data items and producing W′ third range clamped data items, where W′ is an integer less than or equal to W, said W′ third range clamped data items being representative of said M first data items.
According to another embodiment of the present invention, the number M of first data items equals 2
Y
, where Y is an integer greater than or equal to one.
According to another embodiment of the present invention, the P second data items are quantised, i.e. compressed, independently of each other and the T first data items are dequantised, i.e. decompressed, independently of each other.
The present invention also proposes a method for processing digital data items. The processing method could, for example, be incorporated as part of an MPEG decoder for processing digital video/image data items. The method comprises the steps of compressing and decompressing the digital data items. The step of compressing further comprises the steps of: receiving and transforming M first data items into N second data items, where M is an integer greater than or equal to two and where N is an integer less than or equal to M; receiving and quantising P second data items and producing Q third data items, where P is an integer less than or equal to N and where Q is an integer less than or equal to P; and appropriately storing in memory and/or transferring R third data items; where R is an integer less than or equal to Q. The step of decompressing said decompression comprising further comprises the steps of: appropriately retrieving from memory and/or rec
Galanthay Theodore E.
Jean Pierre Peguy
Morris James H.
SGS-Thomson Microelectronics Limited
Tokar Michael
LandOfFree
Method and circuitry for compressing and decompressing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and circuitry for compressing and decompressing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and circuitry for compressing and decompressing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2439587