Abstract: Methods, systems, and apparatus, for handling applications in an ambient computing system. One of the methods includes determining, by a low-power processing component, that particular sensor signals have a particular property. In response, a machine learning engine performs an inference pass over a machine learning model using the sensor signals to generate a model output. If the model output of the machine learning engine matches an application-specific condition, one or more of the other processing components are activated to execute an particular application corresponding to the application-specific condition.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a prefetch processing to prepare an ambient computing device to operate in a low-power state without waking a memory device. One of the methods includes performing, by an ambient computing device, a prefetch process that populates a cache with prefetched instructions and data required for the ambient computing device to process inputs to the system while in the low-power state, and entering the low-power state, and processing, by the ambient computing device in the low-power state, inputs to the system using the prefetched instructions and data stored in the cache.

Abstract: Methods, systems, and apparatus, for performing low-power vision sensing. One computing device includes a vision sensor configured to generate vision sensor data and an ambient computing system configured to repeatedly process the vision sensor data generated by the vision sensor according to a low-power detection process. If a detection is indicated by the low-power detection process, the ambient computing system wakes one or more other components of the computing device to perform a high-power detection process using the vision sensor data.

Abstract: Methods, systems, and apparatus, for handling applications in an ambient computing system with a privacy processor. One of the methods includes to remain in a monitoring power state until a controller receives an interrupt indicating that one or more sensor signals are present. The one or more sensor signals are provided as input to a machine learning engine. An inference pass is performed by the machine learning engine to generate an output representing a particular context that is specific to a particular user. It is determined that one or more components of an ambient computing system should be disabled based on the on the particular context for the particular user. In response, the one or more components of the ambient computing system are disabled.

Abstract: Methods, systems, and apparatus, for handling applications in an ambient computing system. One of the methods includes determining, by a low-power processing component, that particular sensor signals have a particular property. In response, a machine learning engine performs an inference pass over a machine learning model using the sensor signals to generate a model output. If the model output of the machine learning engine matches an application-specific condition, one or more of the other processing components are activated to execute an particular application corresponding to the application-specific condition.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for virtualizing external memory as local to a machine learning accelerator. One ambient computing system comprises: an ambient machine learning engine; a low-power CPU; and an SRAM that is shared among at least the ambient machine learning engine and the low-power CPU; wherein the ambient machine learning engine comprises virtual address logic to translate from virtual addresses generated by the ambient machine learning engine to physical addresses within the SRAM.

Abstract: Methods, systems, and apparatus, for handling applications in an ambient computing system. One of the methods includes determining, by a low-power processing component, that particular sensor signals have a particular property. In response, a machine learning engine performs an inference pass over a machine learning model using the sensor signals to generate a model output. If the model output of the machine learning engine matches an application-specific condition, one or more of the other processing components are activated to execute an particular application corresponding to the application-specific condition.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for virtualizing external memory as local to a machine learning accelerator. One ambient computing system comprises: an ambient machine learning engine; a low-power CPU; and an SRAM that is shared among at least the ambient machine learning engine and the low-power CPU; wherein the ambient machine learning engine comprises virtual address logic to translate from virtual addresses generated by the ambient machine learning engine to physical addresses within the SRAM.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a prefetch processing to prepare an ambient computing device to operate in a low-power state without waking a memory device. One of the methods includes performing, by an ambient computing device, a prefetch process that populates a cache with prefetched instructions and data required for the ambient computing device to process inputs to the system while in the low-power state, and entering the low-power state, and processing, by the ambient computing device in the low-power state, inputs to the system using the prefetched instructions and data stored in the cache.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a prefetch processing to prepare an ambient computing device to operate in a low-power state without waking a memory device. One of the methods includes performing, by an ambient computing device, a prefetch process that populates a cache with prefetched instructions and data required for the ambient computing device to process inputs to the system while in the low-power state, and entering the low-power state, and processing, by the ambient computing device in the low-power state, inputs to the system using the prefetched instructions and data stored in the cache.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for virtualizing external memory as local to a machine learning accelerator. One ambient computing system comprises: an ambient machine learning engine; a low-power CPU; and an SRAM that is shared among at least the ambient machine learning engine and the low-power CPU; wherein the ambient machine learning engine comprises virtual address logic to translate from virtual addresses generated by the ambient machine learning engine to physical addresses within the SRAM.

Abstract: Methods, systems, and apparatus, for handling applications in an ambient computing system. One of the methods includes determining, by a low-power processing component, that particular sensor signals have a particular property. In response, a machine learning engine performs an inference pass over a machine learning model using the sensor signals to generate a model output. If the model output of the machine learning engine matches an application-specific condition, one or more of the other processing components are activated to execute an particular application corresponding to the application-specific condition.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a prefetch processing to prepare an ambient computing device to operate in a low-power state without waking a memory device. One of the methods includes performing, by an ambient computing device, a prefetch process that populates a cache with prefetched instructions and data required for the ambient computing device to process inputs to the system while in the low-power state, and entering the low-power state, and processing, by the ambient computing device in the low-power state, inputs to the system using the prefetched instructions and data stored in the cache.

Abstract: A technique to provide a hardware security module that provides a secure boundary for retention of a secure key within the secure boundary and prevention of unauthorized accesses from external sources outside of the secure boundary to obtain the secure key. The hardware security module includes a security processor to unwrap and authenticate a secure key within the secure boundary to decrypt or encrypt data and to provide data through a single interface that communicates with external sources, so that all data transfers between the secure boundary, formed by the hardware security module, and external sources are transferred only through the interface. The hardware security module ensures no unwrapped key leaves the secure boundary established by the hardware security module.

Abstract: Provided is a method for transferring data from one clock domain within a synchronizer to another domain within the synchronizer. The method includes determining system clock parameters within the synchronizer and analyzing a first domain clock signal based upon the system clock parameters. Next, a second domain clock signal is analyzed based upon the first domain clock signal and the system clock parameters. A determination is made as to when to transfer data from a first clock domain to a second clock domain in accordance with the analysis of the first and second domain clock signals, and an enable signal is provided to affect the data transfer from the first domain to the second clock domain.

Abstract: A modular integrated circuit includes a hub module that is coupled to a plurality of spoke modules via a plurality of hub interfaces. The plurality of hub interfaces provide a plurality of signal interfaces between the hub module and each of the plurality of spoke modules, wherein each of the plurality of signal interfaces is isolated from each of the other signal interfaces of the plurality of signals interface, and wherein each of the plurality of signal interfaces operates in accordance with a common signaling format.

Abstract: A technique to provide a hardware security module that provides a secure boundary for retention of a secure key within the secure boundary and prevention of unauthorized accesses from external sources outside of the secure boundary to obtain the secure key. The hardware security module includes a security processor to unwrap and authenticate a secure key within the secure boundary to decrypt or encrypt data and to provide data through a single interface that communicates with external sources, so that all data transfers between the secure boundary, formed by the hardware security module, and external sources are transferred only through the interface. The hardware security module ensures no unwrapped key leaves the secure boundary established by the hardware security module.

Abstract: A technique to provide a hardware security module that provides a secure boundary for retention of a secure key within the secure boundary and prevention of unauthorized accesses from external sources outside of the secure boundary to obtain the secure key. The hardware security module includes a security processor to unwrap and authenticate a secure key within the secure boundary to decrypt or encrypt data and to provide data through a single interface that communicates with external sources, so that all data transfers between the secure boundary, formed by the hardware security module, and external sources are transferred only through the interface. The hardware security module ensures no unwrapped key leaves the secure boundary established by the hardware security module.

Abstract: A data processing apparatus includes a volatile memory, a random number generator adapted for generating random numbers from which one or more keys are generated, and a memory encryption unit (MEU). The MEU is configured to receive an N-bit block of data and to divide the N-bit block of data into two more sub-blocks of data, where each sub-block contains fewer than N-bits. The MEU is further configured to encrypt each sub-block of data using the one more keys, to combine the encrypted sub-blocks into an N-bit block of encrypted data, and to write the encrypted N-bit block of data to the volatile memory.

Abstract: A modular integrated circuit includes a hub module that is coupled to a plurality of spoke modules via a plurality of hub interfaces. A memory module stores hub software and hub data and configuration data. The hub software operates in accordance with a memory map that includes a plurality of first reserved blocks corresponding to memory reserved for the plurality of spoke modules, and at least one second reserved block corresponding to memory reserved for at least one optional spoke module. The plurality of first reserved blocks are activated based on the configuration data and the at least one second reserved block is deactivated based on the configuration data.