University of California

King Abdullah University of Science and Technology

Newport Media, Inc.

Sasken Communication Technologies Limited

Unist-Academy Industry Research Corporation

Ahmed Mohamed Eltawil

Fadi Joseph Kurdahi

Elsayed A Ahmed

Abdulkadir Çelik

Mohammed Fouda

Muhammad Sayed Khairy Abdelghaffar

Jun Ma

Khaled Nabil Salama

Bedri A Cetiner

Babak Daneshrad

Mohammed Affan Zidan

Xiaoyu Fu

Mohammad Mansour

Zhiying Wang

Marwen Zorgui

Amr M A Hussein

Gi Ju Jung

Su Gil Lee

Amin Khajeh

Jong Eun Lee

Asmaa Abdallah

Hasan Erdem Yantir

A data center system having plural components including core switches and racks R, each rack Rhaving plural servers; a global controller configured to control a traffic flow to each rack of the plural racks R; plural reconfigurable intelligent surface, RIS, modules, a RIS module being configured to receive a first electromagnetic signal from a first component and emit a second electromagnetic signal toward a second component, the second electromagnetic signal carrying a same information as the first electromagnetic signal; and a local controller configured to adjust an emitting direction of the second electromagnetic signal by changing a current flow through the RIS module.

Apparatus, systems, and methods for metasurface assisted wireless data centers

A machine learning based method for channel estimation for a multiple-input multiple-output, MIMO, system, the method including receiving a measured signal y[k] at a receiver of the system; finding subcarriers k of the measured signal y[k]; estimating, with a convolutional neural network, CNN, channel amplitudes ĝ[k] of the measured signal y[k]; reconstructing a channel Ĥ[k], between the receiver and a transmitter of the system, based on the channel amplitudes ĝ[k] and a low resolution whiten measurement matrix; and adjusting a parameter of the system based on the reconstructed channel Ĥ[k]. The channel amplitudes ĝ[k] are simultaneously estimated by the CNN.

Machine learning based channel estimation method for frequency-selective MIMO system

A physiological data acquisition system includes an array electrode sensor () having plural electrodes and configured to acquire physiological data; a single electrode sensor having a single electrode and configured to acquire additional physiological data; and a hub that is configured to receive the physiological data from the array electrode sensor and the additional physiological data from the single electrode sensor only along body communication channels. At least one of the array electrode sensor and the single electrode sensor is configured to send an energy request signal to the hub, along the body communication channels. The hub, in response to the received energy request signal, emits radio frequency signals, which are used by the at least one of the array electrode sensor and the single electrode sensor to harvest energy. Alternatively, the hub may be configured to send a wake up signal to at least one of the array electrode sensor and the single electrode sensor before emitting the radio frequency signals.

Combined wireless and body channel communication system for physiological data collection

A stuck-at fault mitigation method for resistive random access memory (ReRAM)-based deep learning accelerators, includes: confirming a distorted output value (Y0) due to a stuck-at fault (SAF) by using a correction data set in a pre-trained deep learning network, by means of ReRAM-based deep learning accelerator hardware; updating an average (&#x3bc;) and a standard deviation (&#x3c3;) of a batch normalization (BN) layer by using the distorted output value (Y0), by means of the ReRAM-based deep learning accelerator hardware; folding the batch normalization (BN) layer in which the average (&#x3bc;) and the standard deviation (&#x3c3;) are updated into a convolution layer or a fully-connected layer, by means of the ReRAM-based deep learning accelerator hardware; and deriving a normal output value (Y1) by using the deep learning network in which the batch normalization (BN) layer is folded, by means of the ReRAM-based deep learning accelerator hardware.

Stuck-at fault mitigation method for ReRAM-based deep learning accelerators

Disclosed are various approaches for a controller that can generate and use non-stationary polar codes for encoding and decoding information. In one example, a method includes performing, by an encoder of the controller, a linear operation on at least one vector of information to be stored in a memory. The linear operation includes generating a polar encoded representation from the at least one vector of information. The linear operation also includes generating an output using at least one permutation that is based on a statistical characterization analysis of channels of the memory and a channel dependent permutation that is applied to the polar encoded representation. In some aspects, the statistical characterization analysis includes a respective reliability level of each one of the plurality of channels, and the channel dependent permutation includes an ordered permutation that orders the channels according to their respective reliability level.

Non-stationary polar codes for resistive memories

Various examples are provided examples related to resistive content addressable memory (RCAM) based in-memory computation architectures. In one example, a system includes a content addressable memory (CAM) including an array of cells having a memristor based crossbar and an interconnection switch matrix having a gateless memristor array, which is coupled to an output of the CAM. In another example, a method, includes comparing activated bit values stored a key register with corresponding bit values in a row of a CAM, setting a tag bit value to indicate that the activated bit values match the corresponding bit values, and writing masked key bit values to corresponding bit locations in the row of the CAM based on the tag bit value.

Resistive content addressable memory based in-memory computation architecture

Embodiments of transceivers with one or more reconfigurable antennas are described. In one embodiment, a reconfigurable antenna transceiver includes a transmit chain, a receive chain, and a reconfigurable antenna having a plurality of reconfigurable modes. The transceiver may also include an antenna controller configured to set a mode of the reconfigurable antenna. According to other aspects, the transceiver may also include a signal processor configured to transmit a set of training symbols during a training interval. The antenna controller may be further configured to select a respective mode of the reconfigurable antenna for each training symbol in the set of training symbols. Additionally, the antenna controller may be configured to calculate a received Signal-of-Interest to Interferer Ratio (SIR) for each training symbol of the set of training symbols. In this context, a system utilizing a reconfigurable antenna may achieve significant rate improvement compared to half-duplex systems.

Full duplex reconfigurable antenna self-interference cancellation systems

Embodiments of full-duplex self-interference cancellation systems are described. In one embodiment, a full-duplex transceiver includes a digital signal processor that processes digital signals, a transmit chain that receives a first digital baseband signal from the digital signal processor and converts it to a first RF signal, a receive chain that receives a second RF signal and converts the second RF signal to a second digital baseband signal, and an auxiliary receive chain that receives a portion of the first RF signal and converts it to an auxiliary digital baseband signal. The transceiver may further include a self-interference canceller that applies a channel transfer function to the auxiliary digital baseband signal to generate a cancellation signal and subtracts the cancellation signal from the second digital baseband signal to cancel self-interference at the transceiver. The suppression of self-interference in the transceiver assists the transceiver to achieve higher efficiency in full-duplex communications.

Full-duplex self-interference cancellation systems

A power management technique utilizing a method for accurately and rapidly estimating the change in the statistical distribution of data at each block in a communication system leading to or originating from a memory that is experiencing voltage scaling induced errors is disclosed. An appropriate memory supply voltage that maximizes power savings is found by exploiting the available SNR slack while keeping system performance within a required margin.

Cognitive power management for communication systems

Embodiments of full-duplex systems with reconfigurable antennas are described. In one embodiment, a full-duplex reconfigurable antenna transceiver includes a transmit chain, a receive chain, and a reconfigurable antenna having a plurality of reconfigurable modes. The transceiver may also include an antenna controller configured to set a mode of the reconfigurable antenna. According to other aspects, the transceiver may also include a signal processor configured to transmit a set of training symbols during a training interval. The antenna controller may be further configured to select a respective mode of the reconfigurable antenna for each training symbol in the set of training symbols. Additionally, the antenna controller may be configured to calculate a received Signal-of-Interest to Interferer Ratio (SIR) for each training symbol of the set of training symbols. In this context, a full-duplex system utilizing a reconfigurable antenna may achieve significant rate improvement compared to half-duplex systems.

A method of processing an in-phase signal component and a quadrature signal component of a low intermediate frequency (LIF) signal includes estimating and correcting an amplitude imbalance between a digitized in-phase signal component and a digitized quadrature signal component at a first point in time, and estimating and correcting a phase imbalance between the digitized in-phase signal component and the digitized quadrature signal component at a second point in time in response to the correcting process. The digitized in-phase signal component corresponds to the in-phase signal component at the first point in time and the digitized quadrature signal component corresponds to the quadrature signal component at the first point in time. The second point in the time is subsequent to the first point in time.

Digital approach for image rejection correction in low intermediate frequency conversion tuner

A universal rake receiver architecture includes modular independent processing units that can be flexibly programmed to support different modes of operation. The processing units are capable of performing the basic correlation calculations of DS-CDMA and each unit has an internal local memory and controller that controls its mode of operation. Each unit performs the required synchronization and demodulation operations for a multipath of a signal in the digital domain using all-digital frequency and timing correction techniques. Frequency feedback need not be supplied to the analog section of the receiver. Interpolation most preferably is used to find the optimum sampling position of each incoming chip. This independence allows the receiver to be used with one to several antennas without design modifications.

Growing community of inventors

Oakland, CA, United States of America

Ahmed Mohamed Eltawil