Abstract: A matrix compression/decompression accelerator (MCA) system/method that coordinates lossless data compression (LDC) and lossless data decompression (LDD) transfers between an external data memory (EDM) and a local data memory (LDM) is disclosed. The system implements LDC using a 2D-to-1D transformation of 2D uncompressed data blocks (2DU) within LDM to generate 1D uncompressed data blocks (1DU). The 1DU is then compressed to generate a 1D compressed superblock (CSB) in LDM. This LDM CSB may then be written to EDM with a reduced number of EDM bus cycles. The system implements LDD using decompression of CSB data retrieved from EDM to generate a 1D decompressed data block (1DD) in LDM. A 1D-to-2D transformation is then applied to the LDM 1DD to generate a 2D decompressed data block (2DD) in LDM. This 2DD may then be operated on by a matrix compute engine (MCE) using a variety of function operators.

Abstract: An integrated circuit (IC) package includes a molding having a first surface and a second surface, the first surface opposing the second surface. An interconnect is encased in the molding. The interconnect includes pads situated at a periphery of a side of the IC package. A portion of the pads are exposed at the first surface of the molding. A die pad is situated proximal to the second surface of the molding. The die pad has a first surface and a second surface, the first surface opposing the second surface, and the second surface is circumscribed by the second surface of the molding. A die is mounted on the first surface of the die pad. A heat spreader is mounted on the second surface of the molding and the second surface of the die pad. The heat spreader extends between edges of the second surface of the molding.

Abstract: A method for encoding a video sequence in a video encoder to generate a compressed video bit stream is provided that includes binarizing a plurality of syntax elements, wherein each binarized syntax element comprises a string of one or more binary symbols (bins), wherein a bin is one selected from a context-coded bin and bypass bin, encoding the context-coded bins of the binarized syntax elements using binary arithmetic encoding, and adding the bypass bins of the binarized syntax elements to the compressed video bit stream with no encoding.

Abstract: A method of de-blocking filtering a processed video is provided. The processed video includes a plurality of blocks and each block includes a plurality of sub-blocks. A current block of the plurality of blocks includes vertical edges and horizontal edges. The processed video further includes a set of control parameters and reconstructed pixels corresponding to the current block. A boundary strength index is estimated at the vertical edges and at the horizontal edges of the current block. The set of control parameters, the reconstructed pixels corresponding to the current block and partially filtered pixels corresponding to a set of adjacent sub-blocks are loaded. The vertical edges and the horizontal edges of the current block are filtered based on the boundary strength index and the set of control parameters such that a vertical edge of the current block is filtered before filtering at least one horizontal edge of the current block.

Abstract: A system includes a memory configured to store a two-dimensional data structure that includes radar data arranged such that radar data of a first transmitter is separated from radar data of a second transmitter by a Doppler offset in the two-dimensional data structure. The system also includes a data fetch mechanism that includes a lookup table (LUT) applied on either of two dimensions. The lookup table is configured to store a data fetch location in the two-dimensional data structure, where the data fetch location indicates a location from which to fetch a subset of the radar data from the two-dimensional data structure and the data fetch mechanism is configured to fetch the subset of the radar data from the two-dimensional data structure based on the LUT. The system includes a processor configured to perform a fast Fourier transform (FFT) on the fetched subset of the radar data.

Abstract: This disclosure describes a test architecture that supports a common approach to testing individual die and dies in a 3D stack arrangement. The test architecture uses an improved TAP design to facilitate the testing of parallel test circuits within the die.

Abstract: A system includes a processor and a memory set coupled to the processor. The system also includes a repair circuit coupled to the memory set. The repair circuit includes a first repair circuit and a second repair circuit. The repair circuit also includes a test controller configured to select between the first repair circuit and the second repair circuit to perform an in-field self-repair of the memory set.

Abstract: A wafer chip scale package (WCSP) includes a substrate including a semiconductor surface including circuitry electrically connected to die bond pads exposed by a passivation layer, and a top dielectric layer over the passivation layer. A dielectric layer bounded (DLB) cavity formed in the top dielectric layer includes a first cavity being a center through-cavity bounded by a second cavity being a partial through-cavity, the DLB cavity is lined with a seed layer. A capping dielectric layer that covers the DLB cavity except for an aperture over the first cavity. A cavity metal that is generally configured as an integral structure of continuous metal material having no interfaces is for filling the DLB cavity to form a metal filled cavity including over the aperture that has an electrical connection to the die bond pads. A solder ball over the cavity metal is positioned over the aperture.

Abstract: A method is provided that includes performing, by a processor in response to a vector matrix multiply instruction, multiplying an m×n matrix (A matrix) and a n×p matrix (B matrix) to generate elements of an m×p matrix (R matrix), and storing the elements of the R matrix in a storage location specified by the vector matrix multiply instruction.

Abstract: An integrated circuit is fabricated by forming transistors having gates of orthogonal orientations and implanting, at two first rotations, a first pocket implant using a first dopant type with a masking pattern on a substrate surface layer, the two first rotations respectively forming two first pocket implantation angles and two first pocket implantation beam orientations, and implanting, at two second rotations, a retrograde gate-edge diode leakage (GDL) reduction pocket implant using a second dopant type with the masking pattern on the substrate surface layer, the two second rotations respectively forming two GDL-reduction implantation angles and two GDL-reduction implantation beam orientations.

Abstract: A semiconductor device includes a first semiconductor structure. The first semiconductor structure includes a first semiconductor material having a band-gap. The first semiconductor structure has a first surface. An insulating layer has first and second opposing surfaces. The first surface of the insulating layer is on the first surface of the first semiconductor structure. A second semiconductor structure is on the second surface of the insulating layer and includes a second semiconductor material having a band-gap that is smaller than the band-gap of the first semiconductor material. A floating electrode couples the first semiconductor structure to the second semiconductor structure.

Abstract: An integrated circuit (IC) is provided with a plurality of diode based mm-wave peak voltage detectors (PVD)s. During a testing phase, a multi-point low frequency calibration test is performed on one or more of the PVDs to determine and store a set of alternating current (AC) coefficients. During operation of the IC, a current-voltage sweep is performed on a selected one of the PVDs to determine a process and temperature direct current (DC) coefficient. A peak voltage produced by the PVD in response to a high frequency radio frequency (RF) signal is measured to produce a first measured voltage. An approximate power of the RF signal is calculated by adjusting the first measured voltage using the DC coefficient and the AC coefficient.

Abstract: A caching system including a first sub-cache, a second sub-cache, coupled in parallel with the first sub-cache, for storing cache data evicted from the first sub-cache and write-memory commands that are not cached in the first sub-cache, and a cache controller configured to receive two or more cache commands, determine a conflict exists between the received two or more cache commands, determine a conflict resolution between the received two or more cache commands, and sending the two or more cache commands to the first sub-cache and the second sub-cache.

Abstract: A motor system with an input for coupling to a motor control signal that, when presented in a predetermined state, indicates a motor receiving power should be disabled from rotating. The system also includes controller circuitry for providing a disabling signal to motor rotation, independent of processor software control signaling and the power, in response to the control signal.

Abstract: An integrated circuit (IC) die includes a substrate with circuitry configured for at least one function including metal interconnect levels thereon including a top metal interconnect level and a bottom metal interconnect level, with a passivation layer on the top metal interconnect level. A scribe street is around a periphery of the IC die, the scribe street including a scribe seal utilizing at least two of the plurality of metal interconnect levels, an inner metal meander stop ring including at least the top metal interconnect level located outside the scribe seal, wherein the scribe seal and the inner metal meander stop ring are separated by a first separation gap. An outer metal meander stop ring including at least the top metal interconnect level is located outside the inner metal stop ring, wherein the outer stop ring and the inner stop ring are separated by a second separation gap.

Abstract: In examples, a package comprises a semiconductor die having a device side and a bond pad on the device side, a conductive terminal exposed to an exterior of the package, and an electrical fuse. The electrical fuse comprises a conductive ball coupled to the bond pad, and a bond wire coupled to the conductive terminal. The bond wire is stitch-bonded to the conductive ball.

Abstract: A PLC network system and method operative with OFDM for generating MIMO frames with suitable preamble portions configured to provide backward compatibility with legacy PLC devices and facilitate different receiver tasks such as frame detection and symbol timing, channel estimation and automatic gain control (AGC), including robust preamble detection in the presence of impulsive noise and frequency-selective channels of the PLC network. A PLC device may include a delayed correlation detector and a cross-correlation detector operating in concert to facilitate preamble detection in one implementation.

Abstract: A method of operating a mesh network is disclosed. The method includes joining a network as a child of a parent node and receiving a downlink broadcast channel from the parent node. The method further includes setting the downlink broadcast channel as an uplink broadcast channel in response to the step of receiving.

Abstract: A circuit includes a filter, a comparator, and converter. A first input of the comparator couples to the output of the filter. A second input of the comparator is configured to receive ramp signal. An input of the converter couples to the output of the comparator. The circuit also includes a dual minimum pulse generator having an input coupled to the output of the converter. The dual minimum pulse generator is configured to, responsive to an input pulse on the input of the dual minimum pulse generator having a pulse width less than a predetermined delay time period, generate a pulse on the first output of the dual minimum pulse generator that has a pulse width equal to a sum of the pulse width of the input pulse and the predetermined delay time period. A driver is coupled to the output of the dual minimum pulse generator.

Abstract: A radio communications device includes a RTC configured to run even during sleep for receiving from a coordinator node (CN) in an asynchronous channel hopping WPAN an asynchronous hopping sequence (AHS) frame that includes the CN's hopping sequence. A processor implements a stored sleepy device operation in asynchronous channel hopping networks algorithm. The algorithm is for determining a time stamp for the AHS frame and the CN's initial timing position within the hopping sequence, storing the time stamp, going to sleep and upon waking up changing a frequency band of its receive (Rx) channel to an updated fixed channel. A data request command frame is transmitted by the device on the CN's listening channel that is calculated from the CN's hopping sequence, time stamp, CN's initial timing position and current time, and the device receives an ACK frame transmitted by the CN at the updated fixed channel of Rx operation.