Uplink Power Control via Machine Learning Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling uplink power levels in wireless communication networks face challenges in balancing signal quality, interference, and power consumption, particularly in managing dynamically changing noise and interference between neighboring cells.
Innovation Solution
The method involves grouping neighboring cells into clusters, identifying clusters with varying signal quality, and creating a model using machine learning to map nominal uplink power levels to estimated signal quality values. This model is then used to adjust the nominal uplink power levels in cells with lower signal quality to match the signal quality of cells with higher signal quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the nominal uplink power level (poNominal) is increased to improve uplink signal strength and SINR, then the uplink bitrate per PRB improves, but the number of PRBs that can be transmitted per TTI decreases and interference in neighboring cells increases
Solution Approach 1:
The patent applies local quality by differentiating power control strategies for different cell types. Cells are categorized into first-type cells (with adequate signal quality) and second-type cells (with poor signal quality). The model derives different optimal poNominal values for each cell type based on their specific characteristics, rather than applying a uniform power level across all cells. This allows each cell to operate at its optimal power level locally, resolving the contradiction between signal strength and PRB transmission capacity.
Solution Approach 2:
The patent changes the power control parameter poNominal dynamically based on cell type and network conditions. The machine learning model determines optimal poNominal values by analyzing relationships between power levels and signal quality metrics (SINR, RSRP) for different cell types. This parameter adjustment resolves the contradiction by finding the optimal power level that balances signal strength requirements with PRB transmission capacity for each specific cell context.
2Reliability
If the nominal uplink power level is increased to improve uplink SINR, then effective uplink bitrate per PRB increases, but interference in neighboring cells increases reducing cluster throughput
Solution Approach 1:
The patent applies local quality by implementing cell-specific power control based on individual cell characteristics and their roles in the network. The model identifies first-type cells (with adequate signal quality) and second-type cells (with poor signal quality) and assigns different optimal poNominal values. This localized approach ensures that power increases are applied only where necessary (second-type cells) rather than uniformly across the cluster, thereby improving SINR locally without causing excessive interference to neighboring cells.
Solution Approach 2:
The patent converts the potential harm of increased interference into a benefit by using the machine learning model to predict and optimize power levels that improve signal quality without causing excessive interference. The model learns from historical data to identify power settings that achieve adequate SINR while minimizing negative impact on neighboring cells, effectively converting the trade-off between signal strength and interference into a win-win scenario through data-driven optimization.
3Reliability
If existing uplink power control methods are used to maintain adequate signal quality, then uplink coverage is maintained, but power consumption increases and the system cannot adapt to dynamically changing noise and interference
Solution Approach 1:
The patent applies dynamics by implementing adaptive power control that responds to changing network conditions. The machine learning model continuously learns from historical data and updates optimal poNominal values based on dynamically changing noise and interference patterns. This allows the system to adjust power levels adaptively rather than using static power control, maintaining adequate coverage while reducing power consumption by avoiding unnecessary power increases when conditions permit.
Solution Approach 2:
The patent implements feedback through the machine learning model that uses historical signal quality measurements and power level data to continuously optimize power control parameters. The model analyzes the relationship between poNominal settings and actual signal quality outcomes, using this feedback to refine future power control decisions. This feedback mechanism enables the system to maintain coverage reliability while optimizing power consumption by learning from past performance and adapting to changing conditions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of controlling uplink power levels of radio cells in a wireless communications network is disclosed. The method comprises grouping sets of neighboring cells into a plurality of cell clusters, identifying at least one first cell cluster comprising at least one cell being subjected to interference and having a signal quality exceeding a signal quality threshold value, identifying at least one second cell cluster comprising at least one cell being subjected to interference and having a signal quality below the signal quality threshold value; creating a model based on the identified first cell cluster to map a nominal uplink power level to an estimated signal quality value thereof, and adjusting a nominal uplink power level of the at least one cell of the second cell cluster by applying the nominal uplink power level derived from the created model to attain a corresponding estimated signal quality value therein.