Abstract: An AI accelerator apparatus using in-memory compute chiplet devices. The apparatus includes a first semiconductor substrate having a plurality of chiplets, each of which includes a plurality of tiles. Each tile includes a plurality of slices, a central processing unit (CPU), and a hardware dispatch device. Each slice can include a digital in-memory compute (DIMC) device configured to perform high throughput computations. In particular, the DIMC device can be configured to accelerate the computations of attention functions for transformer-based models (a.k.a. transformers) applied to machine learning applications. The chiplets are in a full mesh connectivity configuration such that at least one of the die-to-die (D2D) interconnects of each chiplet is coupled to one of the D2D interconnects of each other chiplet using a non-diagonal link. The chiplets can also include other interfaces to facilitate communication between the chiplets, memory and a server or host system.

Abstract: Systems and methods are provided for pre-caching content to customer premises equipment (CPE) such as set-top boxes, fixed wireless access devices, etc. Video streaming services are increasingly popular and can put a strain on network resources during peak hours. Wireless network peak hour capacity demand can be reduced by utilizing off-peak hour content/video caching. A cache time during off-peak hours can be determined through the interaction of wireless network and content provider servers.

Abstract: A method of monitoring a fluid level in a power unit relies on a neural network model. The model is used to predict lateral acceleration during air ingestion events. These predictions are compared to measured values of lateral acceleration. When the measured values and the predicted values are not well correlated, a controller takes corrective action.

Abstract: Methods and systems for operating a voltage regulator are described. A integrated circuit can be configured to adjust at least one of a deadtime parameter and a drive strength parameter of a power stage based on at least one of an input voltage being provided to a power stage, a switch node voltage of the power stage, and an output current of the power stage. A controller of the power stage can be further configured to adjust a deadtime of the power stage based on adjustment of the deadtime parameter. The controller can be further configured to adjust a drive strength of the first driver and the second driver based on adjustment of the drive strength parameter.

Abstract: An alignment fixture for a reactor system may comprise a fixture body comprising an inner perimeter at least partially defining a shape which comprises an inner space of the fixture body, wherein the inner space is configured to receive a susceptor of a reactor system; and/or a measuring protrusion coupled to the fixture body at a first position and protruding from the fixture body toward the inner space. The measuring protrusion may comprise an indicator between the fixture body and a measuring protrusion end of the measuring protrusion.

Abstract: The present disclosure relates to a user equipment, a base station and respective methods. The user equipment (UE) receives a single uplink grant for a plurality of PUSCH transmissions. The single uplink grant comprises an antenna port field with an index value that is to be used for the plurality of PUSCH transmissions. The UE determines time-domain resources based on the received uplink grant. The determined time-domain resources prescribe a number of PUSCH transmissions and a length of each of the number of PUSCH transmissions. The UE transmits the number of PUSCH transmissions using the determined time-domain resources. Each of the number of PUSCH transmissions comprises at least one front-loaded demodulation reference signal, DMRS.

Abstract: Apparatuses, methods, and systems are disclosed for configuring a sensing reference signal. One method includes receiving, at a first device, configuration information from a second device. The configuration includes: a set of sensing reference signal sequence generation parameters; a set of sensing reference signal resource pattern parameters, wherein a pattern corresponding to the set of sensing reference signal resource pattern parameters includes time domain locations of symbols within a sensing reference signal, frequency domain locations of resource elements within the sensing reference signal, or a combination thereof; and information indicating to map a generated sequence based on the pattern on at least one antenna to create a sensing reference signal pattern. The method includes generating a sensing reference signal. The method includes transmitting and/or receiving the sensing reference signal according to the configuration information.

Abstract: A user equipment, a base station and respective methods are provided. The user equipment receives a physical uplink shared channel (PUSCH) config information element (IE) in form of radio resource control (RRC) signaling, the PUSCH config IE being applicable to a particular bandwidth part. The user equipment configures a table which is defined by a PUSCH time domain resource allocation list IE carried in the received PUSCH config IE, the table comprising rows, at least one row comprising a first set of values related to allocated time-domain resources for a plurality of PUSCH transmissions. The user equipment receives downlink control information (DCI) signaling carrying a time-domain resource assignment filed with value m, wherein the value m provides a row index m+1 to the RRC configured table. The user equipment determines allocated time-domain resources for the plurality of PUSCH transmissions based on the index of the slot carrying the received DCI.

Abstract: A method, apparatus, non-transitory computer readable medium, apparatus, and system for image generation include obtaining a sketch input depicting an object, processing the sketch input to obtain sketch guidance, and generating a synthesized image based on the sketch guidance using an image generation model, where the synthesized image depicts the object from the sketch input.

Abstract: Systems and methods related to rotor embedded impellers for axial flux machine cooling are disclosed herein. An axial flux electric machine may include a stator, a rotor spaced apart from the stator in an axial direction of the axial flux electric machine to form an air gap, and at least one impeller on the rotor, where the at least one impeller includes a rotor gap through the rotor and an airfoil element. Specific embodiments of the present application can efficiently cool an axial flux electric machine to allow for higher torque and power generation for the same base axial flux electric motor design (e.g., the same cost and volume). These embodiments provide a simple and effective method to extract heat from an axial flux electric machine by creating pressure drops and fluid flow patterns within the motor cavity.

Abstract: Systems and methods for discovering caller paths in a micro-service graph are disclosed. The system can receive a first request to interact with a first node, wherein the first node is associated with a software application. The system can generate a first bit encoding indicative of the first node. The system can transmit, from the first node, a second request to a second node associated with the software application, wherein the second request is associated with the first request. The system can generate, based on the first bit encoding, a second bit encoding indicative of the first node and the second node. The system can determine, based on the second bit encoding, a sequence of requests.

Abstract: Apparatuses, methods, and systems are disclosed for semi-static channel access with directional FFP. One method (800) includes receiving (805) a configuration for a plurality of FFPs, where each FFP is associated with a separate transmit beam for transmission within that FFP. The method (800) includes identifying (810) an initiated FFP and performing (815) communication activity during the initiated FFP using a beam corresponding to the initiated FFP, where the communication activity includes a transmission, a reception, or a combination thereof.

Abstract: A control device may be configured to transmit messages using an adaptive transmit power. The control device may determine a transmit power for transmitting the message based on a command in the message. The control device may determine a transmit power based on a change in lighting intensity caused by the command. The transmit power may be greater when the change in lighting intensity is above a threshold than when the change in light intensity is below the threshold. The control device may determine whether the message was successfully received based on the receipt of an acknowledgement message. The control device may increase the transmit power when the message fails to be received and retransmit the message including the command at an increased transmit power. The control device may store (e.g., learn) the increased transmit power for later use.

Abstract: A computing device can receive a query in a first query language that identifies a set of data to be processed and determine that at least a portion of the set of data resides in an external data system that uses a different query language. The query system can translate the query in the first query language in to a second query language for the external data system. To retrieve results of the translated query, the computing device may determine a quantity of results of generated by the translated query and generate one or more results readers to read the results in parallel. The computing device may further process the results and provide them to a user.

Abstract: A system includes a communications interface configured to receive a plurality of data series captured by a plurality of sensors. A first database includes at least a first data series of the plurality of data series. A second database includes metadata related to the first data series. One or more processors are configured to receive an instruction including an indication of a computation to perform and metadata filter criteria, query the second database to determine at least a portion of the first data series that satisfies the metadata filter criteria by reference to at least the metadata related to the first data series, access, from the first database and based on results of the query of the second database, the portion of the first data series, and perform the computation on the portion of first data series accessed from the first database.

Abstract: Apparatuses, methods, and systems are disclosed for enhanced UL repetition indication and repetition procedure provides one or more signaling mechanisms to support dynamic PUCCH repetition factor indication. A User Equipment apparatus includes a transceiver and a processor that receives from a network, a Radio Resource Control (“RRC”) configuration for one or more physical uplink control channel (“PUCCH”) resources sets and corresponding one or more PUCCH resource configurations for each of the configured PUCCH resource sets, where one or more of the corresponding PUCCH resource configurations indicate a number of slots over which PUCCH can be repeated and receives from the network a downlink control information (“DCI”), where the DCI includes a PUCCH resource indicator (“PRI”) to indicate one of a RRC configured PUCCH resources within a PUCCH resource set and a corresponding number of slots for PUCCH repetition.

Abstract: Aspects of the disclosure relate to NFT exchanges. A computing platform may generate a plurality of F-NFTs including F-NFT identifiers that configure the F-NFTs for end-to-end tracking. The computing platform may extract, from a voice communication, information of a request to transfer an F-NFT. The computing platform may generate, based on the information of the request, a text file. The computing platform may input, into a GPT-4 algorithm, the text file to produce a F-NFT contract identifying the F-NFT, the transferor of the F-NFT, and a transferee of the F-NFT. The computing platform may send, to an NFT exchange platform, the F-NFT contract, which may cause the NFT exchange platform to transfer custody of the F-NFT from an account of the transferor at a first institution to an account of the transferee at a second institution.

Abstract: The subject technology is related to autonomous vehicles (AV) and, in particular, to calibration of an sensors of an AV on open roads after calibrating the sensors in an undefined training area. An example method includes calibrating a plurality of sensors of the AV based on performing a plurality of training steps of increasing complexity within the undefined environment, wherein the plurality of sensors of the AV are uncalibrated with respect to each other, determining that the AV satisfies an open road requirement to navigate an open road based on completing the plurality of training steps in the undefined environment, and operating the AV on open roads subject to at least one constraint. The method of claim 1, wherein the at least one constraint comprises a list of prohibited maneuvers that limit operation of the AV on open roads. The constraints are configured to dynamically change.

Abstract: Apparatuses and methods for determining a route are provided. In some examples, the method comprises: receiving measurements of one or more motion sensors of a mobile device disposed in a vehicle; selecting, based on a start time and an end time of a first route, a subset of the measurements; determining, based on the subset of the measurements, turns or stops made by the vehicle as part of the first route and their directions; determining, based on the subset of the measurements, distances of path segments traveled by the vehicle between the turns or stops; constructing an estimated route including a sequence of the turns or stops and path segments; selecting, based on comparing the estimated route and a plurality of candidates routes from a map, the first route; and displaying the map including the first route.

Abstract: A method includes: receiving a first set of data associated with an element during a first stage of a plurality of stages; applying a first stage machine learning model to the first set of data to generate a prediction value, wherein the first stage machine learning model is trained with a first set of feature data that is available during the first stage; updating the prediction value by: receiving a second set of data associated with the element during a second stage of the plurality of stages; and applying a second stage machine learning model to the second set of data, wherein the second stage machine learning model is trained with (i) the first set of feature data, and (ii) a second set of feature data that is available during the second stage; and initiating performance of a mitigation action based on the updated prediction value.