Qualcomm Incorporated

Broadcom Corporation

King Abdullah University of Science and Technology

Centre National De La Recherche Scientifique

Université´ D'aix-Marseille

American University of Beirut

Mohammad Mansour

Louay A Jalloul

Sam Padinjaremannil Alex

Ahmed Mohamed Eltawil

Abdulkadir Çelik

Ali Chehab

Asmaa Abdallah

Hassan Aziza

Basma Hajri

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

Systems and methods are provided for efficient tree-based detection of multi-carrier modulated signals, such as Orthogonal Frequency Division Multiplexing (OFDM) symbols. In an embodiment, a plurality of signals occupying respective tones are received and processed to determine an order, based on a tone quality metric, for the plurality of signals. The plurality of signals are then dispatched to a pool of tree detectors in accordance with the order. In an embodiment, the order ensures that signals occupying higher quality tones, and requiring shorter detection times, are dispatched first to the pool of tree detectors. In another embodiment, a maximum runtime of busy tree detectors of the pool is updated based signal on actual detection times to exploit the time slack of early terminating detectors.

Efficient tree-based MIMO OFDM detection

Techniques for efficiently generating addresses for pruned interleavers and pruned de-interleavers are described. In an aspect, a linear address may be mapped to an interleaved address for a pruned interleaver by determining the total number of invalid mappings corresponding to the linear address. The linear address may be summed with the total number of invalid mappings to obtain an intermediate address. The interleaved address for the pruned interleaver may then be determined based on a non-pruned interleaver function of the intermediate address. The pruned interleaver may be a pruned bit-reversal interleaver, a pruned Turbo interleaver composed of a bit-reversal function and a linear congruential sequence function, or some other type of interleaver. The total number of invalid mappings may be determined iteratively, and each iteration may be performed in different manners for different types of pruned interleaver.

Efficient address generation for pruned interleavers and de-interleavers

Growing community of inventors

Beirut, Lebanon

Mohammad Mansour