Methods, apparatus and machine-readable media relating to channel quality prediction in a wireless network
The disclosure provides a method for estimating channel quality in a wireless network. The method comprises: obtaining information relating to a channel quality for a first time period, as measured by a wireless node located in an environment comprising one or more machines having mechanical parts which are movable in a periodic pattern; and providing the information as an input to a predictive model, developed using a machine-learning algorithm, to obtain a predicted channel quality in the environment for a second, subsequent time period.
1 . A method for estimating channel quality in a wireless network, the method comprising:
obtaining information relating to a channel quality for a first time period, as measured by a wireless node located in an environment comprising one or more machines having mechanical parts which are movable in a periodic pattern; and
providing the information relating to the channel quality and information about the periodic pattern as an input to a predictive model, developed using a machine-learning algorithm, to obtain a predicted channel quality in the environment for a second, subsequent time period.
2 . The method of claim 1 , further comprising providing information relating to the periodic pattern as a second input to the predictive model.
3 . The method of claim 2 , wherein the information relating to the periodic pattern comprises: an indication of a frequency of the periodic pattern and/or an indication of a start time of each cycle of the periodic pattern.
4 . The method of claim 2 , wherein the obtaining the information relating to the periodic pattern comprises:
receiving an indication of the information relating to the periodic pattern directly or indirectly from the one or more machines or controllers thereof; or
analyzing the information relating to the channel quality to determine the information relating to the periodic pattern.
5 . The method of claim 2 , wherein the information relating to the periodic pattern for the first time period has a length equal to a periodicity of the periodic pattern.
6 . The method of claim 1 , wherein the information relating to the channel quality for the first time period comprises a channel quality indicator index.
7 . The method of claim 1 , wherein the information relating to the channel quality for the first time period comprises a channel spectral response.
8 . The method of claim 1 :
wherein the predictive model comprises a first part learned from data measured by a plurality of wireless nodes located in the environment, and a second part learned from data measured by the wireless node;
wherein the first part comprises one or more first layers of a neural network; and
wherein the second part comprises one or more second layers of the neural network, the second layers being higher than the first layers.
9 . The method of claim 1 , wherein the method is performed by the wireless node.
10 . The method of claim 9 , further comprising receiving the predictive model from a network node.
11 . The method of claim 9 , further comprising transmitting the predicted channel quality to a serving network node for the wireless node.
12 . The method of claim 1 , wherein the method is performed by a network node.
13 . The method of claim 12 , further comprising utilizing the predicted channel quality to select a modulation and coding scheme for transmissions to the wireless node.
14 . The method of claim 1 , wherein the wireless node is located on one of the machines having mechanical parts which are movable in a periodic pattern.
15 . A node for estimating channel quality in a wireless network, the node comprising:
processing circuitry;
memory containing instructions executable by the processing circuitry whereby the node is operative to:
obtain information relating to a channel quality for a first time period, as measured by a wireless node located in an environment comprising one or more machines having mechanical parts which are movable in a periodic pattern; and
provide the information relating to the channel quality and information about the periodic pattern as an input to a predictive model, developed using a machine-learning algorithm, to obtain a predicted channel quality in the environment for a second, subsequent time period.
16 . The node of claim 15 , wherein the instructions are such that the node is operative to provide information relating to the periodic pattern as a second input to the predictive model.
17 . The node of claim 16 , wherein the instructions are such that the node is operative to obtain the information relating to the periodic pattern by:
receiving an indication of the information relating to the periodic pattern directly or indirectly from the one or more machines or controllers thereof; or
analyzing the information relating to the channel quality to determine the information relating to the periodic pattern.
18 . The node of claim 16 , wherein the information relating to the periodic pattern for the first time period has a length equal to a periodicity of the periodic pattern.
19 . The node of claim 15 :
wherein the predictive model comprises a first part learned from data measured by a plurality of wireless nodes located in the environment, and a second part learned from data measured by the wireless node;
wherein the first part comprises one or more first layers of a neural network; and
wherein the second part comprises one or more second layers of the neural network, the second layers being higher than the first layers.
20 . The node of claim 15 , wherein the node is the wireless node.