Texas Instruments Corporation

Qualcomm Incorporated

Njr Corporation

Lsi Logic Corporation

Toshiaki Yoshino

Kwok K Chau

Yiwan Wong

Louis G Johnson

Kai Wang

Hiroyuki Nishimura

Qiong Wu

Yiliang L Bao

Jian Wei

King Pang

The subject matter of this specification can be implemented in, among other things, a method of decoding video data that includes decoding compressed input video data using a first entropy coding technique to obtain first decoded data. The method further includes re-encoding the first decoded data using a second entropy coding technique that is different from the first entropy coding technique to obtain re-encoded data. The method further includes storing the re-encoded data in a storage device. The method further includes decoding the re-encoded data using the second entropy coding technique to obtain second decoded data.

Two-stage entropy decoding

Techniques for optimizing the Context-based Adaptive Binary Arithmetic Coding (CABAC) bitstream decoding are disclosed. In one configuration, a device has a first processing circuit operative to decode a Context-based Adaptive Binary Arithmetic Coding (CABAC) bitstream into an intermediate signal having a CABAC decoded standard format and a decoded order. A second processing circuit decodes the intermediate signal using a non-CABAC decoding standard. A buffer is provided between the first and second processing circuits to improve processing speeds.

Architecture for multi-stage decoding of a CABAC bitstream

A decoder is disclosed that provides dynamic pipelining of an incoming compressed bitstream. The decoder includes decoding logic modules capable of decoding an incoming compressed bitstream, and memory storing logic in communication with at least one of the decoding modules. Preferably, the memory storing logic is capable of determining whether a memory operation is complete that stores the uncompressed video data to memory. In addition, the decoder includes halting logic in communication with the decoding logic and the memory storing logic. The halting logic halts the decoding of the incoming bitstream during a specific time period, which includes a time period wherein the memory operation is incomplete. Finally, initiating logic is included in the decoder that is in communication with the decoding logic and the memory storing logic. The initiating logic of the decoder restarts the decoding when the memory operation is complete.

Method and apparatus for dynamic pipelining

A decoder for decoding a compressed incoming interruptible bitstream is disclosed. The decoder includes an input register that is capable of receiving a latch command, and that is further capable of latching stored data in a particular state upon receiving the latch command. The decoder also includes decoding logic, such as variable length decoding logic, or run-length decoding logic, in communication with the input register. Further included in the decoder is an output register in communication with decoding logic, that is also capable of receiving the latch command, and that is further capable of latching stored data in a particular state upon receiving the latch command. Finally, the decoder includes a register controller in communication with both the input register and the output register. The register controller is capable of receiving a halt command from the system and, upon receiving the halt command, the register controller sends the latch command to both the input register and the output register.

Method and apparatus for bitstream decoding

A circuit for calculating a sum of absolute errors for use in full block search matching in a motion estimation processor is disclosed herein, the circuit being easily implemented and capable of running at 54 Mhz. The circuit accesses search window and reference data from memory and loads the data into rows of laterally interacting processing elements having an architecture capable of fast data processing. A sum of absolute errors between all elements of each row of search data and all elements of each row of reference data is calculated, and the absolute error for all rows of processing elements is totalled. From this total sum of absolute error, the motion vector may be predicted for the next frame.

Motion estimation processor architecture for full search block matching

An nth degree function computing device having a low-cost, small scale circuit in which no multipliers are present and which allows high-speed computing operations. The nth degree function computing device comprises two series operators 32 and 38 connected in series, with an adder 44 inserted between them, and with an adder 48 inserted between the output terminal of the second-stage series operator 38 and device output terminal 46. A constant 2a.sub.2 is sent from constant generator 50 to the first input terminal of an adder 34 of the first-stage series operator 32. A constant (a.sub.1 -a.sub.2) is sent from constant generator 52 to first input terminal of adder 44. A constant a.sub.0 is sent from constant generator 54 to the first input terminal of adder 48. With respect to variable x (integer), a clock circuit 56 sends (x+1) synchronized clock pulses CLK.sub.i and x clock pulses CLK.sub.k to registers 36 and 42 of first-stage and second-stage series operators 32 and 38, respectively.

Computing device for nth degree functions

A digital filter, in which the operation time can be kept short even when the order of the filter is high. The digital filter comprises a first transversal filter TF.sub.1 which has an input terminal connected to a signal input terminal 10, first delay means DR.sub.1 which has an input terminal connected to the output terminal of first transversal filter TF.sub.1, a second transversal filter TF.sub.2 connected to the output terminal of the first delay means DR.sub.1, second delay means DR.sub.2 which has an input terminal connected to the output terminal of the second transversal filter TF.sub.2, adder AD which has one of the input terminals connected to the output terminal of the second delay means DR.sub.2, third delay means DR.sub.3 which has an input terminal connected to the output terminal of the adder AD and an output terminal connected to the signal output terminal, and a third transversal filter TF.sub.3 which has an input terminal connected to the output terminal of the aforementioned third delay means DR.sub.3 and an output terminal connected to the other input terminal of the aforementioned adder. In this digital filter, the chain configuration of the operator that determines the time needed for one cycle of operation is a multiplier-adder chain including one multiplier and one adder in each transversal filter. Consequently, its operation time is equal to the sum of the operation times needed by one multiplier and one adder and is unrelated to the order of each transversal filter, that is, the order of the digital filter.

Infinite impulse response digital filter with plural processing units

A tri-state driver circuit is disclosed which provides rail-to-rail output swings and does not consume a significant amount of d.c. power.

BICMOS tri-state circuit with full output voltage swing

A pipelined Fast Fourier Transform (FFT) architecture includes a memory for storing complex number data. A pipelined data path is coupled to the memory for accessing R complex number data therefrom, for computing an FFT butterfly, and storing R results from the FFT butterfly computation in the memory during one pipeline cycle.

Pipelined Fast Fourier Transform (FFT) architecture

A method for densely packing a complex multiplier for multiplying two complex numbers in the form of (A+jB) and (X+jY) is provided. The complex multiplier consist of two multipliers which perform (A*X), (A*Y), (B*X) and (B*Y) multiplications, and each multiplier has a plurality of partial product generating stages and a partial product summing stage. The method comprises interleaving the plurality of partial product generating stages of the (A*Y) multiplier between the plurality of partial product generating stages of the (A*X) multiplier, forming the partial product summing stage adjacent the interleaved partial product generating stages of the (A*X) and (A*Y) multipliers. Further interleave the plurality of partial product generating stages of the (B*Y) multiplier between the plurality of partial product generating stages of the (B*X) multiplier, and form the partial product summing stage adjacent the interleaved partial product gneerating stages of the (B*X) and (B*Y) multipliers.

Method for densely packing a complex multiplier

Apparatus for generating memory addresses for accessing and storing data in an FFT (Fast Fourier Transform) computation is provided. The FFT computation is typically performed by computing a plurality of FFT butterflies belonging to a plurality of ranks. The apparatus includes a butterfly counter for determining the current FFT butterfly being computed. The butterfly counter produces a plurality of butterfly carries. A rank counter for determining the rank of said current FFT butterfly being computed produces a rank number. Coupled to the rank and butterfly counters is incremental curcuitry, which generates an incremental number in response to the rank number and the butterfly carries. An adder circuitry coupled to the incremental circuitry adds the incremental number and a plurality of memory addresses to produce the FFT data memory addresses.

Growing community of inventors

Fremont, CA, United States of America

Toshiaki Yoshino