# US 2024/0005069 A1 · Rru undervoltage risk prediction method, apparatus, and system, device, and medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0005069 A1
- Document ID: US20240005069A1
- Status: Published application
- Kind: A1
- Application number: 18/254,621
- Filed: 2021-10-09
- Published: 2024-01-04
- Earliest claimed: 2020-11-27
- Assignees: Zte Corporation
- Applicants: Zte Corporation
- Inventors: Yanliang ZHAO; Xiaolei YU
- Web page: https://patentgrep.com/patent/US20240005069A1

## CPC

- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06N 3/098: Distributed learning, e.g. federated learning
- H04W 24/04: Arrangements for maintaining operational condition
- G06Q 10/04: Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202)
- Y02D 30/70: in wireless communication networks
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

RRU undervoltage risk prediction method, apparatus, and system, and medium are disclosed. The method may include: acquiring a device parameter of a target RRU; inputting the device parameter of the target RRU into a prediction model, where the prediction model is acquired by federated learning through a common node and at least one local node, and the common node connects with each of the at least one local node, and each of the at least one local node comprises at least one RRU; and performing a prediction of the risk of under-voltage failure for the target RRU by means of the prediction model.

## Claims

### Claim 1

A method for predicting risk of under-voltage failure for a Remote Radio Unit (RRU), comprising,
acquiring a device parameter of a target RRU;
inputting the device parameter of the target RRU into a prediction model, wherein the prediction model is acquired by federated learning through a common node and at least one local node, and the common node connects with each of the at least one local node, and each of the at least one local node comprises at least one RRU; and
performing a prediction of the risk of under-voltage failure for the target RRU by means of the prediction model.

### Claim 2 (depends on 1)

The method according to claim 1, wherein initial models of the local node and the common node are the same, and the prediction model is acquired through operations of the federated learning comprising,
calculating, by each of the at least one local node, a model parameter of a model of the respective local node according to sample data of under-voltage failure of RRU in the respective local node, and reports the calculated model parameter to the common node;
integrating, by the common node, each reported model parameter to update a model parameter of a model of the common node, and sends the updated model parameter to each local node;
updating, by each of the at least one local node, the model of the respective local node according to the sent model parameter;
repeating the above operations until a loss function of the model of the respective local node converges; and
acquiring the prediction model according to the model of the common node or the model of the respective local node.

### Claim 3 (depends on 2)

The method according to claim 2, wherein the initial model is established based on a linear regression algorithm.

### Claim 4 (depends on 1)

The method of claim 1, wherein the device parameter of the target RRU comprises any one of, overall power consumption, overall input voltage, overall transmission power, RRU model, maximum configurable power of carrier, actual transmission power of carrier, rectifier module capacity, cable length, cable diameter, or a combination thereof.

### Claim 5 (depends on 1)

The method of claim 1, wherein,
the local node is a local node where a proportion of under-voltage failure alarms of RRUs reaches a preset threshold, and
the proportion of under-voltage failure alarms is the proportion of a quantity of RRUs with under-voltage failure alarms to a total quantity of RRUs in the local node.

### Claim 6 (depends on 1)

The method of claim 1, wherein, the common node is a Network Data Analytics Function (NWDAF) server.

### Claim 7

A device for predicting risk of under-voltage failure for a Remote Radio Unit (RRU), comprising,
an acquisition module, configured to acquire a device parameter of a target RRU;
an input module, configured to input the device parameter of the target RRU into a prediction model, wherein, the prediction model is acquired by federated learning through a common node, and at least one local node, and each of the at least one local node comprises at least one RRU; and
a prediction module, configured to perform a prediction of the risk of under-voltage failure for the target RRU by means of the prediction model.

### Claim 8

A system for predicting risk of under-voltage failure for a Remote Radio Unit (RRU), comprising a common node and at least one local node,
wherein initial models of the common node and each of the at least one local node are the same, and each of the at least one local node comprises at least one RRU; and
the common node is configured to:
before a loss function of a model of each of the at least one local node converges, receive each model parameter reported by each local node, integrate each reported model parameter and update the model parameter of the model of the common node, and
send the updated model parameter to each local node; and
each of the at least one local node is configured to:
before the loss function of the respective local node converges, calculate the model parameter of the model of the respective local node according to sample data for under-voltage failure of RRU in the respective local node,
report the calculated model parameter to the common node,
receive the model parameter sent by the common node, and
update the model of the respective local node according to the sent model parameter; and
each of the at least one local node is further configured to:
after the loss function of the respective local node converges, utilize the model of the respective local node as the prediction model to acquire a device parameter of the target RRU,
input the device parameter of the target RRU into the prediction model, and
perform a prediction of the risk of the under-voltage failure of the target RRU by means of the prediction model.

### Claim 9 (depends on 1)

An electronic apparatus, comprising,
at least one processor; and,
a memory in communication with the at least one processor; wherein,
the memory stores an instruction executable by the at least one processor which, when executed by the at least one processor, causes the at least one processor to carry out the method of claim 1.

### Claim 10 (depends on 1)

A non-transitory computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to carry out the method of claim 1.

### Claim 11 (depends on 2)

The method of claim 2, wherein the device parameter of the target RRU comprises any one of, overall power consumption, overall input voltage, overall transmission power, RRU model, maximum configurable power of carrier, actual transmission power of carrier, rectifier module capacity, cable length, cable diameter, or a combination thereof.

### Claim 12 (depends on 3)

The method of claim 3, wherein the device parameter of the target RRU comprises any one of, overall power consumption, overall input voltage, overall transmission power, RRU model, maximum configurable power of carrier, actual transmission power of carrier, rectifier module capacity, cable length, cable diameter, or a combination thereof.

### Claim 13 (depends on 2)

The method of claim 2, wherein, the local node is a local node where a proportion of under-voltage failure alarms of RRUs reaches a preset threshold, and
the proportion of under-voltage failure alarms is the proportion of a quantity of RRUs with under-voltage failure alarms to a total quantity of RRUs in the local node.

### Claim 14 (depends on 3)

The method of claim 3, wherein, the local node is a local node where a proportion of under-voltage failure alarms of RRUs reaches a preset threshold, and
the proportion of under-voltage failure alarms is the proportion of a quantity of RRUs with under-voltage failure alarms to a total quantity of RRUs in the local node.

### Claim 15 (depends on 4)

The method of claim 4, wherein, the local node is a local node where a proportion of under-voltage failure alarms of RRUs reaches a preset threshold, and
the proportion of under-voltage failure alarms is the proportion of a quantity of RRUs with under-voltage failure alarms to a total quantity of RRUs in the local node.

### Claim 16 (depends on 2)

The method of claim 2, wherein, the common node is a Network Data Analytics Function (NWDAF) server.

### Claim 17 (depends on 3)

The method of claim 3, wherein, the common node is a Network Data Analytics Function (NWDAF) server.

### Claim 18 (depends on 4)

The method of claim 4, wherein, the common node is a Network Data Analytics Function (NWDAF) server.

### Claim 19 (depends on 5)

The method of claim 5, wherein, the common node is a Network Data Analytics Function (NWDAF) server.
