Abstract: A process of synchronizing two execution units sharing a common memory with a plurality of memory banks starts by assigning a first memory bank to a one of two execution units. The other memory bank is assigned to the other execution unit. Then a sequence of operations is processed within one of the execution units while another sequence of operations is processed within the other execution unit. When the first execution unit completes a sequence of operations, a synchronizing operation is performed which causes that first execution unit to suspend processing if a corresponding sequence of operations in the other execution unit has not been completed. When both execution units have completed their respective sequences of operations, the assignment of memory banks is swapped between the two execution units, thereby preventing erroneous reads and writes.

Abstract: The invention provides methods and apparatus for performing inverse telecine processing on an input video frame sequence to be encoded. A method well-suited for detelecine of film-only telecine in MPEG-1 or MPEG-2 applications includes the steps of attempting to fit known 3:2 pulldown pattern phases to the entire sequence, determining a likely edit point in the sequence if a known phase cannot be fit to the sequence, and recursively repeating the fitting and edit point determination steps on subsequences until the entire sequence is partitioned into subsequences which can be fit with one of the known phases. The phases fit to the sequence or subsequences are then used to generate field-match indicators which are further processed to generate MPEG-1 or MPEG-2 compliant repeat field codes.

Abstract: An improved current steering cell for a DAC which eliminates the need for an inverter reduces the noise at the common mode. The cell includes a first and a second current steering MOS transistor of a first polarity type, each having a gate and a pair of current passing terminals. The cell has an input terminal for receiving digital input signals coupled to the gate of the first of the pair of current steering transistors, and a common mode node for receiving an input current coupled to the same one of the pair of current passing terminals of each current steering MOS transistor. The current output terminal of the cell is coupled to the other of the pair of current passing terminals of the first of the current steering MOS transistors.

Abstract: A method and apparatus for providing memory arbitration which allows multiple hardware functions implemented in a single ASIC to utilize a single shared memory unit or multiple shared memory units. The memory arbitration technique establishes a priority among multiple memory access requesters and is particularly well-suited for use in a set top box processing system. A plurality of memory access requests are received from a plurality of processing elements in a set top box processing system. The processing elements include a transport stream demultiplexer, a host central processing unit and a graphics processor. The processing elements are permitted to access a shared memory in accordance with an established priority. The established priority assigns a higher priority to the graphics processor than to the host central processing unit, and may be in the order of graphics processor, transport stream demultiplexer, and central processing unit.

Abstract: A selectable reference voltage circuit for a digital-to-analog converter (DAC) which includes an input terminal for receiving an external reference voltage, a voltage comparator having two inputs and an output, one input for receiving the reference voltage and the other for receiving a predetermined voltage, the comparator providing one of two possible output voltages based upon the relationship of the magnitudes of the reference voltage and the predetermined voltage. The circuit includes a multiplexer having a control input coupled to receive as an input signal the output of the comparator, and having two inputs for receiving input voltage signals, one for receiving the voltage on the reference voltage input terminal and the other for receiving an on-chip generated reference voltage, the multiplexer selecting as an output voltage one of the two input voltages determined by the input signal at its control input.

Abstract: A rate control algorithm for an MPEG-2 compliant encoder is described. The rate control algorithm has embodiments useful for constant bit rate and variable bit rate encoding. In particular, the invention relates to a field frame macroblock encoding decision.

Abstract: A rate control algorithm for an MPEG-2 compliant encoder is described. The rate control algorithm has embodiments useful for constant bit rate and variable bit rate encoding. In particular, the invention relates to scene change detection.

Abstract: A decoder for compressed video signals comprises a central processing unit (CPU), a dynamic random access memory (DRAM) controller, a variable length code (VLC) decoder, a pixel filter and a video output unit. The microcoded CPU performs dequantization and inverse cosine transform using a pipelined data path, which includes both general purpose and special purpose hardware. In one embodiment, the VLC decoder is implemented as a table-driven state machine where the table contains both control information and decoded values.

Abstract: A method and structure including four video decompression structures and eight memory banks are provided for decoding high definition television (HDTV) signal. In this HDTV decompression structure, the 1920.times.1080 pixel display space is divided into four vertical sections of 480.times.1080 pixels. Each memory bank stores the values of pixels in one non-overlapping group of 240.times.1080 pixels. Each decompression structure decodes a 480.times.1088-pixel picture area with access to up to two additional 240.times.1088-pixel picture areas. The video decompression structures decode the vertical sections in lock-step to avoid the problem of the same bank of memory being accessed by more than one video decompression structure. In one embodiment of the present invention, a macroblock fetch can cross 1-4 DRAM page boundaries. So, in order to maintain the lock-step relationship of the video decompression structures, each page mode access is limited to fetching only an 8.times.

Abstract: A motion compensation structure and a method are provided for decoding interframe coded video data using motion vectors. The motion compensation structure includes a filter for resampling the pixel data in both vertical and horizontal directions, a prediction memory structure and a weighted adder structure. In one embodiment of the present invention, a weighted adder structure and a method are provided for performing bilinear interpolation of two values using multiplexers and an multiple-input adder.

Abstract: A rate control algorithm for an MPEG-2 compliant encoder is described. The rate control algorithm has embodiments useful for constant bit rate and variable bit rate encoding. In particular, the invention relates to fade detection.

Abstract: A rate control algorithm for an MPEG-2 compliant encoder. The rate control has embodiments useful for constant bit rate and variable bit rate encoding of video frames. The present invention relates to statistical multiplexing, virtual buffers and virtual buffer verifiers.

Abstract: A system and method for removing motion artifacts from an interlaced image is disclosed. The interlaced image comprises an odd and an even field. The system and method includes providing one of the odd and the even fields on every other line of the display and then providing a set of constant signal level lines to the remaining lines of the display. The method and system further includes shifting the location of the constant signal levels lines by a scan line responsive to a timing signal from the display, and providing the other of the odd and the even field to the display responsive to the shift.

Abstract: A structure and a format provide a video signal encoder under the MPEG (Motion Picture Experts Group) standard. In one embodiment, the video signal interface is provided with a decimator for providing input filtering for the incoming signals. In one embodiment, the central processing unit (CPU) and multiple coprocessors implements discrete cosine transform (DCT) and inverse discrete cosine transform (IDCT) and other signal processing functions, generating variable length codes, and provides motion estimation and memory management. The instruction set of the central processing unit provides numerous features in support for such features as alpha filtering, eliminating redundancies in video signals derived from motion pictures and scene analysis. In one embodiment, a matcher evaluates 16 absolute differences to evaluate a "patch" of eight motion vectors at a time.

Abstract: A structure and a method capable of both packing data into and unpacking data from either the little endian or the big endian format are provided. Under the structure and method of the present invention, the packed or unpacked data, as the case may be, is only shifted in one direction. During a packing operation, a stream of n-bit data is packed into a stream of m-bit words. During an unpacking operation, a stream of m-bit packed data is unpacked into a stream of n-bit words. n.ltoreq.m.

Abstract: A method and a structure are provided to decode intraframe and interframe coded compressed video data. In one embodiment, a decompression structure having a processor is provided with a global bus over which a decoder coprocessor, an inverse discrete cosine transform coprocessor and a motion compensation coprocessor communicate. The decompression structure communicates with a host computer over a host bus and with an external dynamic random access memory over a memory bus. The processor provides overall control to the coprocessors by reading and writing into a plurality of data and control registers, each register associated with one of the coprocessors. A structure including four of the decompression structures and a method are provided for decoding high definition television (HDTV) signals. In this structure for decoding HDTV signals, each decompression structure decodes a 480.times.1088-pixel picture area with access to up to two additional 240.times.1088-pixel picture areas.

Abstract: In accordance with the present invention, a structure and a method for asynchronously interfacing a master processor and a slave processor is provided by receiving from and providing to the master device control signals of a polling protocol, and receiving from and providing to the slave device control signals of an interrupt type protocol. In a first embodiment of this invention, the master processor provides WR (write request), RD (read request), OE (output enable) signals, and receives a BUSY (busy) signal. The slave processor receives an "int" (interrupt) signal, and provides "intack" (interrupt acknowledge), "outs" (output), and "ins" (input) signals. In a second embodiment of this invention, instead of the RD signal of the first embodiment, the read request signal is the AND product of an AS (address strobe) signal and the most significant bit of the read address.

Abstract: A digital video compression system and an apparatus implementing this system are disclosed. Specifically, matrices of pixels in the RGB signal format are converted into YUV representation, including a step of selectively sampling the chrominance components. The signals are then subjected to a discrete cosine transform (DCT). A circuitry implementing the DCT in a pipelined architecture is provided. A quantization step eliminates DCT coefficients having amplitude below a set of preset thresholds. The video signal is further compressed by coding the elements of the quantized matrices in a zig-zag manner. This representation is further compressed by Huffman codes. Decompression of the signal is substantially the reverse of compression steps. The inverse discrete cosine transform (IDCT) may be implemented by the DCT circuit. Circuits for implementing RGB to YUV conversion, DCT, quantization, coding and their decompression counterparts are disclosed.

Abstract: A digital video compression system and an apparatus implementing this system are disclosed. Specifically, matrices of pixels in the RGB signal format are converted into YUV representation, including a step of selectively sampling the chrominance components. The signals are then subjected to a discrete cosine transform (DCT). A circuitry implementing the DCT in a pipelined architecture is provided. A quantization step eliminates DCT coefficients having amplitude below a set of preset thresholds. The video signal is further compressed by coding the elements of the quantized matrices in a zig-zag manner. This representation is further compressed by Huffman codes. Decompression of the signal is substantially the reverse of compression steps. The inverse discrete cosine transform (IDCT) may be implemented by the DCT circuit. Circuits for implementing RGB to YUV conversion, DCT, quantization, coding and their decompression counterparts are disclosed.

Abstract: A method and a structure provide discrete cosine transform (DCT) and its inverse (IDCT) using digital FIR filters in a filter bank. The filter bank of the present invention forms a structure of cascaded filters, in which data are communicated only between filters having "parent-child" relationships. Each filter in the filter bank is required only to communicate with at most two other filters in the filter bank. Consequently, in any implementation, both the communication overhead between filters in the filter bank, and the circuit size are minimized. Therefore, the filter bank is particularly suited for integrated circuit implementation. In one embodiment, the filter bank is implemented in an image compression and decompression integrated circuit using a structure which includes pipeline registers, adders and multipliers. In that embodiment, the filter bank provides an 8-point DCT in each of the two passes of a 2-dimensional DCT used in a data compression operation.