Multiple user multiple input multiple output scheduler for fixed wireless access
Methods and systems for implementing MU-MIMO scheduling in a network are provided. The method begins with determining at least one signal condition metric for a plurality of signals used for communication between a base station and a plurality of devices. A machine learning model is trained to determine whether a device is a mobile device or a fixed wireless device using training data. The training data comprises at least one signal condition metric for the plurality of signals. Users may connect to the network by a mobile connection or through a fixed connection. An output of the machine learning model is then used to predict a type of connection to the network for each device of the plurality of devices. Then, based on the type of connection, MU-MIMO pairings are assigned to at least a portion of the plurality of devices.
1 . A method of multiple user, multiple input, multiple output (MU-MIMO) scheduling in a network, the method comprising:
determining at least one signal condition metric for a plurality of signals used for communication between a base station and a plurality of devices;
training a machine learning model to determine whether each device of the plurality of devices is a mobile device or a fixed wireless device using training data, the training data comprising the at least one signal condition metric for the plurality of signals;
using an output of the machine learning model to predict a type of connection to the network for the each device of the plurality of devices; and
based on the type of connection, assigning MU-MIMO pairings to at least a portion of the plurality of devices.
2 . The method of claim 1 , wherein a first device of the plurality of devices is determined to be a fixed wireless device.
3 . The method of claim 2 , wherein the fixed wireless device uses a fixed wireless access point.
4 . The method of claim 2 , wherein the fixed wireless device uses fixed customer premises equipment.
5 . The method of claim 1 , wherein a first device of the plurality of devices is determined to be a mobile device.
6 . The method of claim 1 , wherein the machine learning module determines a type of the plurality of devices based on a fluctuation of the at least one signal condition metric for each device of the plurality of devices.
7 . The method of claim 1 , wherein the at least one signal condition metric comprises at least one of a reference signal received power (RSRP), a reference signal received quality (RSRQ), or an angle of arrival.
8 . The method of claim 6 , wherein the machine learning module learns the fluctuation of the at least one signal condition metric relative to a mean signal metric of the at least one signal condition metric for the plurality of signals.
9 . The method of claim 6 , wherein the machine learning module learns the fluctuation of the at least one signal condition metric relative to a standard deviation of the at least one signal condition metric for the plurality of signals.
10 . The method of claim 1 , wherein each signal of the plurality of signals uses a same frequency and is sent at a same time to the plurality of devices.
11 . A system for multiple user, multiple input, multiple output (MU-MIMO) scheduling, in a network, comprising:
a base station having one or more antennas for receiving a plurality of signals used for communication between the base station and a plurality of devices;
a machine learning module configured to communicate with the base station, wherein the machine learning module is trained to determine whether a device is a mobile device or a fixed wireless device using training data, the training data comprising at least one signal condition metric for the plurality of signals; and
a scheduler configured to communicate with the machine learning module and the base station, the scheduler using an output of the machine learning module to predict a type of connection to the network for the plurality of devices, wherein the scheduler assigns MU-MIMO pairings to at least a portion of the plurality of devices.
12 . The system of claim 11 , further comprising an artificial intelligence (AI) module configured to communicate with the machine learning module and the scheduler.
13 . The system of claim 11 , wherein an AI module instructs the scheduler to revise a schedule of a downlink signal to the plurality of devices.
14 . The system of claim 13 , wherein the revision is based on a change in the at least one signal condition metric.
15 . The system of claim 14 , wherein the change in the at least one signal condition metric is based on tracking the at least one signal condition metric by the AI module.
16 . A non-transitory computer storage media storing computer-usable instructions that, when used by one or more processors, cause the processor to:
determine at least one signal condition metric for a plurality of signals used for communication between a base station and a plurality of devices;
train a machine learning model to determine whether each device of the plurality of devices is a mobile device or a fixed wireless device using training data, the training data comprising the at least one signal condition metric for the plurality of signals;
use an output of the machine learning model to predict a type of connection to the network for each device of the plurality of devices; and
assign MU-MIMO pairings to at least a portion of the plurality of devices based on the type of connection.
17 . The non-transitory computer storage media of claim 16 , wherein a first device of the plurality of devices is determined to be a fixed wireless device.
18 . The non-transitory computer storage media of claim 17 , wherein the fixed wireless device uses fixed customer premises equipment.
19 . The non-transitory computer storage media of claim 16 , wherein a first device of the plurality of devices is determined to be a mobile device.
20 . The non-transitory computer storage media of claim 16 , wherein the machine learning module determines a type of the plurality of devices based on a fluctuation of the at least one signal condition metric for each device of the plurality of devices.