Motorola Mobility LLC

Nokia Solutions and Networks Oy

Motorola Corporation

Nokia Technologies Oy

Chandrasekar Sankaran

Fan Wang

Stanley J Benes

Amitava Ghosh

Jun Tan

Philip J Fleming

Frederick W Vook

Anand Bedekar

Suresh Kalyanasundaram

Timothy J Wilson

Eugene Visotsky

Rajeev Agrawal

Amitabha Ghosh

Alan P Rottinghaus

Nick W Whinnett

Igor Filipovich

Jamil Shihab

Hua Xu

Ian Dexter Garcia

Michael Honig

Lawrence A Downing

Fei Tong

Systems, methods, apparatuses, and computer program products for scheduling radio resources across a group of one or more user equipment (UEs) are provided. One method may include encoding every sequence of multi-user multiple-input multiple-output (MU MIMO) beam combinations into a unique numerical value, adding a Q value into the encoded numerical value to produce a Q encoded value (Q), providing each of the sequence of beam combinations with a unique designated bin, passing a matrix comprising a state representation for each of the beam combinations through a deep Q network (DQN), and outputting, by the deep Q network (DQN), one or more optimal beams to assign to a user equipment (UE).

One shot multi-user multiple-input multiple-output (MU-MIMO) resource pairing using reinforcement learning based deep Q network (DQN)

Systems, methods, apparatuses, and computer program products for multi-user (MU) multiple-input multiple-output (MIMO) user pairing selection are provided. One method may include selecting multi-user multiple input multiple output (MU MIMO) candidate beams using deep neural network(s) (DNNs), and selecting paired users based on the selected beams. The deep neural network(s) (DNNs) are trained to maximize multi-user priority metric (MU-PM) or a heuristic of the multi-user priority metric (MU-PM).

Methods and apparatuses of multi-user multiple-input multiple-output beam selection and user pairing using deep learning

A method of determining a sequence of actions includes training a first deep Q-network (DQN); providing a plurality of entries of a first multi-dimensional matrix as input to the DQN, the first matrix representing a first state, each entry of the first matrix representing an action that can be taken in the first state; determining, using the first DQN, a plurality of Q-values for the plurality of entries of the first matrix, respectively; executing a first action, the first action being the action represented by the entry, from among the plurality of entries, for which the first DQN determined the highest Q-value among the plurality of determined Q-values; accumulating a reward based on executing the first action; and transitioning from the first state to a next state in accordance with a first set of rules and the executed first action.

Method and apparatus for facilitating resource pairing using a deep Q-network

The specification and drawings present a new method, apparatus and software related product (e.g., a computer readable memory) for a power control (e.g., at least for DL power control) in a cell using a concept of a relative load in a network such as a wireless network. The relative load may be a ratio of cell's own load relative to an average load in a cluster of neighboring cells for scaling a nominal power value (or a function of the power value). The load can be measured, for example, either as a number of UEs in the cell, a RB utilization in the cell, a DL information traffic in the cell, or any other metric that's representative of the load in the cell. The resulting computed power for a DL signaling can be either on a per cell basis or on a per UE basis.

Downlink power control using relative load

A method, system and communication network for transmitting information signals via uplink (UL) collaborative SDMA, in a wireless communication system. Base station receiver estimates a channel gain associated with the transmission path(s) of each user and keeps a matrix of normalized covariance, between users. Based on the estimated channel gain and the normalized covariance, ULS utility is able to compute channel capacity. Based on capacity estimates of (1) the multiplexed user signals and (2) the individual user signals, signals are either multiplexed for UL SDMA or are transmitted individually. An optimal selection of multiplexed signals may be based upon: (1) a cross user interface measurement; and (2) a selection mechanism based on eigen-decomposition techniques. The ULS utility enables a UL scheduler to pair information signals with clear spatial distinction and minimal correlation, based on capacity evaluations.

Uplink spatial division multiple access (SDMA) user pairing and scheduling

An OFDMA communication system is provided that schedules partial usage of subchannels (PUSC) major groups by assigning a beam matrix of a plurality of beam matrices to each major group of multiple PUSC major groups, receiving an uplink signal from each user equipment of multiple users equipment, which uplink signal provides an indication of a channel response associated with the user equipment, determining, for each user equipment of the multiple users equipment and based on the indication of channel quality received from the user equipment, a beam matrix and major group to which to assign the user equipment, and based on the determination of a beam matrix and major group to which to assign each user equipment of the multiple users equipment, assigning two or more users equipment of the multiple users equipment to a same beam matrix and major group during a same time slot.

Method and apparatus for major group scheduling in a fixed beam communication system

An apparatus and method for fractional frequency reuse in a communication system. The method includes a first stepof partitioning a sub-frame into resource partitions, wherein at least one resource partition has frequency reuse of one and all other resource partitions have frequency reuse greater than one. A next stepincludes scheduling resources for a mobile station in a sector in each resource partition used by that sector.

Fractional frequency reuse in a communication system

An apparatus and method to provide uplink interference control in a WiMAX communication system includes a stepof receiving a downlink channel quality indication from an MS. A next stepselecting a modulation product ratio modification parameter in response to the channel quality indication. A next stepincludes modifying a modulation product ratio by the modification parameter. A next stepincludes mapping the modified modulation product ratio against a table of modulation coding schemes. A next stepincludes applying the mapped modulation coding scheme.

Uplink interference control in a wiMAX communication system

Various embodiments are described to provide for the transmission and reception of data in an improved manner. Data transmission is improved by including in a transmitter a null generator () to generate an output data symbol sequence that exhibits nulls in the frequency domain at particular frequencies that an input data symbol sequence does not. A pilot inserter () then adds a pilot symbol sequence to this output data symbol sequence to create a combined symbol sequence. Since the pilot symbol sequence exhibits pilot signals corresponding to the nulls of the output data symbol sequence in the frequency domain, the combined symbol sequence exhibits pilots that are orthogonal to the data in the frequency domain.

Method and apparatus for transmission and reception of data

A radio communications device () that has multiple receive antennas processes received data communications signals to select between space time coding and spatial multiplexing as a selected transmission technique from a base device () that has multiple transmit antennas. A channel throughput (--) for each transmission technique is estimated based on signal to interference and noise ratios (--) of signals being transmitted through a MIMO channel () as measured by a receiver (). The transmission technique with the higher estimated throughput is determined. If spatial multiplexing is determined to have the higher estimated throughput and the throughput of each layer of the spatially multiplexed signal is greater than a threshold, spatial multiplexing is selected. Otherwise, space time coding is selected.

Growing community of inventors

Arlington Heights, IL, United States of America

Chandrasekar Sankaran