Abstract:
An integrated circuit comprising a plurality of multiplier-accumulator circuits connected in series to form a linear pipeline to process first data, via performing a plurality of concatenated multiply and accumulate operations, and generate MAC output data, wherein each multiplier-accumulator circuit of the plurality of multiplier- accumulator circuits includes (i) a multiplier to multiply first data by a multiplier weight data and generate a product data, and (ii) an accumulator, coupled to the multiplier of the associated multiplier-accumulator circuit, to add second data and the product data of the associated multiplier to generate sum data. The integrated circuit further includes an activation circuit, connected to the output of the linear pipeline of the plurality of multiplier-accumulator circuits, to receive the MAC output data and process the MAC output data, via a non-linear activation function, to generate MAC pipeline output data.
Abstract:
Implementations disclosed describe a programmable analog subsystem (PASS) having a plurality of reconfigurable analog circuits. The PASS may be coupled to an input/output device to receive an input signal and to an interface to communicate data with a central processing unit. In a first PASS configuration, with the plurality of reconfigurable analog circuits having a first configuration setting, the PASS may process the input signal through the plurality of reconfigurable analog circuits to generate a first output value based on the input signal. Responsive to the first output value, the PASS may reconfigure the plurality of reconfigurable analog circuits into a second PASS configuration having a second configuration setting, such that the second configuration setting is different than the first configuration setting.
Abstract:
Techniques and mechanisms described herein include a signal processor implemented as an overlay on a field-programmable gate array (FPGA) device that utilizes special purpose, hardened intellectual property (IP) modules such as memory blocks and digital signal processing (DSP) cores. A Processing Element (PE) is built from one or more DSP cores connected to additional logic. Interconnected as an array, the PEs may operate in a computational model such as Single Instruction - Multiple Thread (SIMT). A software hierarchy is described that transforms the SIMT array into an effective signal processor.
Abstract:
Disclosed are various embodiments providing circuitry that includes camouflaged gates that each have multiple switches arranged in a predefined format. A switch at a specific position in one camouflaged gate can have a different threshold voltage than a switch at the specific position in another camouflaged gate. The logical function performed by the camouflaged gate can be based on which of the switches have a low threshold voltage and which of the switches have a high threshold voltage.
Abstract:
An embedded device has a plurality of processor cores, each with a plurality of peripheral devices, wherein each peripheral device has an output. Furthermore, a housing with a plurality of assignable external pins and a protected pin ownership logic for each assignable external pin is provided and configured to be programmable to assign an output function of an associated assignable external pin to only one of the plurality of processor cores.