Abstract: Pixel data is stored and subsequently read from a random access memory of a video decompression processor in a manner that reduces the number of times different rows of the RAM must be addressed in order to retrieve portions of the pixel data therefrom. Pixel data from a video frame is stored in the RAM as a plurality of pages. Each page substantially fills a different row of the RAM and corresponds to a different section of the video frame. A motion vector is decoded to determine the location of a prediction area within the video frame. In the event that the prediction area encompasses more than one of the pages of the video frame, the pixel data is retrieved one page at a time, minimizing the number of row changes required when addressing the RAM to retrieve the data.

Abstract: Impulse purchasing is provided for packages of services offered to consumers via an information network. A plurality of services is offered for individual purchase on an impulse basis via a user terminal coupled to the information network. The plurality of services is offered for purchase as a package on terms that may be more favorable than when the services are purchased individually. Upon purchase of a package, a package record is established for access by the user terminal. The package record provides cryptographically secure access to services provided in the package during a defined time period over which the package is valid.

Abstract: In a Time Division Multiple Access (TDMA) communication network, a central office (CO) provides interactive communication between a service provider and a plurality of user terminals. To synchronize transmissions of the user terminals, a timebase timer in the CO generates modulo N-bit cyclical reference counts as time markers. In a downstream direction to the plurality of user terminals, time marker insertion units in the CO receive both (a) separate digital TDMA transport streams having a predetermined data rate including data packets and Media Access Control (MAC) packets that are interspersed between the data packets at predetermined intervals, and (b) the time markers generated by the timebase timer, and insert a currently received time marker count into a concurrently received MAC packet. The resultant downstream TDMA transport streams with the inserted time marker counts in the MAC packets are transmitted in continuous TDMA output transport stream to the remote user terminals.

Abstract: Disparity estimation between the right and left view pixel luminance values in a stereoscopic video signal is optimized by determining the minimum least-square-error between macroblocks of the right and left view pictures. Affine transform coefficients and disparity vectors which correspond to the minimum error are also determined and transmitted in the data stream for use by a decoder in reconstructing the right view picture. The scheme can be implemented either locally, at the macroblock level, or globally, at the picture level. At the macroblock level, least-square-error optimization may occur for each individual macroblock in the right view picture. In this case, affine transform coefficients are provided for each macroblock. At the picture level, the sum of the least-square-errors is minimized after the blocks of the right view picture are matched to the left view picture. In this case, only one set of affine transform coefficients are required for the entire right view picture.

Abstract: A video signal is modified to include a supplemental burst signal on at least some of the scan lines for each frame. The supplemental burst signals are used to prevent a pirate decoder from descrambling horizontally sync suppressed scrambled signals by deriving a synchronization timing reference from the color burst signal. The supplemental burst signals can also or alternatively be used to carry data in an otherwise conventional analog video signal.

Abstract: Suppressed synchronizing signals of a scrambled standard TV picture are recovered and the picture unscrambled by a method including the steps of: separating a video portion having suppressed sync signals from an audio portion of a TV picture at an intermediate frequency level, amplifying the video portion through a gain controlled amplifier, selectively compensating for the suppressed levels of sync signals, passing the video portion with recovered sync signals of the TV picture through a demodulator, and providing a feedback signal from the demodulator to the gain controlled amplifier to automatically control its gain and thereby minimize artifacts in the TV picture after sync recovery. This new method is easily carried out by apparatus using only a few standard circuit components which can be manufactured as an application specific integrated circuit (ASIC).

Abstract: A high voltage silicon rectifier includes a substrate portion and an epitaxial mesa portion that is a frustrum of a pyramid with a substantially square cross section and side walls that make a forty five degree angle with the substrate portion. The mesa portion includes three germanium doped layers that introduce strain to speed up recombination of charge carriers.

Abstract: A method and apparatus are disclosed for interpolating pixels to obtain subpels for use by a video decompression processor. A prediction area is defined from which subpels are necessary to decompress a portion of a video image. Instead of reading all of the pixels from the prediction area and then processing them together to perform the necessary interpolation, portions of the pixel data are read and simultaneously averaged using in-place computation in order to reduce hardware requirements. Rounding of subpixel results is achieved using the carry input of conventional adders to add a binary "1" to the averaged pixels, which are subsequently truncated to provide the interpolated subpels.

Abstract: Secure point-to-point communication of information to specific terminals is provided via a shared network. Far in advance of the establishment of an information session with a particular one of a plurality of terminals, a unique session identifier is securely delivered to the terminal by a highly secure entity. Information to be provided to the terminal is subsequently encrypted under the session identifier of that terminal by an insecure connection manager. The encrypted information is inserted into designated locations in a signal multiplex. The signal multiplex is transmitted over a portion of the network serving the terminal that is to receive the information. The terminal is informed of the designated locations of the encrypted information in the signal multiplex and of a transmission frequency at which the signal multiplex is carried on the network portion.

Abstract: A rotationally invariant trellis coder is provided for encoding data to be transmitted using a two-dimensional symbol modulation. A precoder, provided at the transmitter, processes data such that a counterpart postcoder at the receiver will provide an output that is invariant to any multiple of a 90.degree. rotation. An encoder encodes the precoded data using a transparent binary convolutional code, which can be a punctured or unpunctured code. The encoded data is mapped to a two-dimensional signal space having a plurality of signal points. The signal points are labeled with unique binary codes in which the two least significant bits, denoted by (I.sub.j, Q.sub.j), are permuted and partially complemented to (Q.sub.j, I.sub.j) for each 90.degree. phase rotation around the signal space. The remaining most significant bits for each point, if any, are invariant to such rotation. The postcoder provided at the receiver inverts the precoder and is feedback-free, thus limiting error propagation.

Abstract: A method and apparatus are provided for changing the compression level of digital information, such as digital video information. After having been compressed by a first compression scheme that provides quantized transform coefficients, the digital information is partially decompressed to recover the transform coefficients as they were prior to quantizing. The transform coefficients are then requantized at a different compression level. Complexity and expense are reduced by only partially decompressing and requantizing the previously compressed information to modify its compression level without the need to provide components such as a motion compensation processor and frame store that would be required for full decompression of the information prior to recompression.

Abstract: Variable rate data is reformatted for communication to an external device at a fixed rate. Fixed length packets of the variable rate data are recovered from a multiplex of data packets. A packet start byte is added to the beginning of each of the recovered packets. Fill bits are added to the recovered packets as necessary to maintain the desired fixed output rate if the output rate is greater than an information rate of the variable rate data. The recovered packets with the added packet start byte and fill bits are provided as output to the external device at the fixed rate. The recovered packets with the added packet start byte and fill bits are advantageously encoded using alternate mark inversion (AMI) and binary eight zero substitution (B8ZS). The packet start byte and B8ZS substitution can comprise unique AMI violation sequences.

Abstract: Mesa diodes of improved mechanical properties are formed by providing a central depression in the regions of the chip from which the mesa is formed before the diffusion step that forms the rectifying junction in the mesa. In symmetric diodes, symmetric depressions are formed on both the top and bottom surfaces of the chip.

Abstract: Apparatus is provided for authenticating information using a double feedforward hash function to provide complementarity in the implementation of an encryption algorithm. A cryptographic processor has a first input for receiving plaintext, a second input for receiving a key and an output for outputting ciphertext generated by cryptographically processing the plaintext and key. A first circuit element is responsive to the ciphertext and plaintext for outputting a first ciphertext derivative. A second circuit element is responsive to at least a portion of the first ciphertext derivative and the key for outputting a second ciphertext derivative. The first and second circuit elements can be XOR gates. Alternatively, these elements can be provided using lookup tables.

Abstract: A scheme is provided for communicating high rate isochronous data in an MPEG-2 type transport stream in a robust manner. The syntax for the data bitstream also provides for the communication of a data clock increment value to a decoder. The increment value enables a simplified and cost effective approach to the direct digital synthesis of a clock frequency necessary to recover data from the packetized data stream at the decoder. The time resolution for presenting isochronous data presentation units at the decoder is increased by providing a presentation time stamp (PTS) extension to supplement a standard MPEG-2 PTS.

Abstract: Limited duration previews of program offerings available for purchase via a communication network are provided in a cryptographically secure manner at virtually any time during the service. The invention has particular applicability to the provision of video services on a pay-per-view basis. Such a video service is provided during a program epoch. A fixed period is defined during the program epoch when portions of the video service are available for viewing on a preview basis. A consumer is allowed to preview, without purchase, portions of the video service at any time during the fixed period for up to a maximum preview duration that is shorter than the fixed period. The consumer can then purchase the video service for viewing during the program epoch after previewing portions thereof. A plurality of records is maintained to service different previewable programs concurrently.

Abstract: A resilient snap fitting retainer provides a desired retention force despite tolerance variations. A pair of parallel resilient legs extends from a base. The legs are separated by a gap. A deformable rib is provided on at least one of the legs to partially crush when the legs are compressed with sufficient force. The partial crushing compensates for tolerance variations in the size of an opening into which the retainer is inserted or in the size of the retainer itself. The legs are elongated and terminate in wedge shaped feet at their distal end. The feet form a generally arrowhead-like shape for insertion into an opening. The structure of the feet and legs enables the legs to deflect toward each other (i.e., buckle) after the interior surfaces of the legs have made contact at their distal ends as the feet are pushed through the opening.

Abstract: A video syntax parser is provided for a video decompression processor. A compressed bitstream is examined until a start code is located. The start code is decoded to enable a particular parser state corresponding thereto to be initiated. The parser state is communicated to a plurality of subsystems of the video decompression processor. Each subsystem monitors the parser state to determine which parsed data following the start code, if any, is intended for that subsystem. The subsystem then retrieves the parsed data it requires via a data bus which receives the parsed data from the parser.

Abstract: A method and apparatus are provided for motion compensated coding of interlaced digital video signals. A first prediction error signal is obtained indicative of the difference between a current portion from a field of a current video frame and a similarly situated first portion in the same field of a prior video frame. A second prediction error signal is obtained indicative of the difference between the current portion and a similarly situated second portion in an opposite field of the prior video frame. A third prediction error signal is obtained indicative of the difference between the current portion and an average of the first and second portions. One of the first, second and third prediction error signals is selected to represent the current portion. Motion vector information is provided to enable the current frame portion to which the selected prediction error signal corresponds to be reproduced by a decoder.