Adaptive Pursuit Learning for Directional Small-Cell Interference

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing wireless network technologies face challenges in mitigating interference in dense small-cell deployments, particularly due to the lack of robust antenna beam selection techniques that can adapt to dynamic environmental changes and the difficulty of integrating directional antennas into existing wireless PHY and MAC stacks.

Innovation Solution

The implementation of a machine learning-based system, LinkPursuit, which uses adaptive pursuit algorithms for real-time antenna state selection in directional small-cell networks, leveraging reconfigurable antennas and a synchronous Time-Division Multiple Access (TDMA) MAC to achieve simultaneous directional transmission and reception, and employs a hybrid synchronization mechanism for network-wide synchronization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If directional antennas are deployed in dense small-cell networks, then spatial reuse and network capacity are improved, but interference management complexity and system reliability deteriorate due to cross-link interference

Engineering Contradiction:

Improvenetwork capacityVSAvoidinterference mitigation

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent implements dynamic antenna state selection where transmitters and receivers continuously adapt their beam directions based on real-time channel conditions and interference levels. The system transitions from static omnidirectional or fixed directional patterns to dynamic beamforming that adjusts orientation and width to maximize signal quality while minimizing interference to other links.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent employs feedback mechanisms where receivers measure the quality of received signals and transmit this information back to transmitters. This feedback loop enables the system to learn from past transmissions and adjust future beam directions to avoid interfering with other links, improving both reliability and spatial reuse over time.

Inventive Principle:

Principle #23Feedback

2Adaptability or versatility

If machine learning-based antenna state selection is implemented, then adaptability to environmental changes is improved, but device complexity and computational requirements increase

Engineering Contradiction:

Improveenvironmental adaptationVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the complex antenna state selection problem into smaller sub-problems handled by individual network nodes. Each transmitter-receiver pair independently learns and adapts their own beam patterns based on local observations, rather than requiring centralized control of the entire network. This distributed approach reduces overall system complexity while maintaining adaptability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system implements self-service through autonomous learning where each node independently adapts its antenna states based on observed interference patterns and channel conditions. The machine learning algorithms enable nodes to automatically adjust their behavior without external intervention, reducing the need for complex centralized management while improving environmental adaptability.

Inventive Principle:

Principle #25Self-service

3Productivity

If synchronous directional transmission is implemented across the network, then spatial reuse is improved, but coordination overhead and synchronization requirements increase

Engineering Contradiction:

Improvespatial reuseVSAvoidsynchronization complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent implements preliminary synchronization actions where transmitters and receivers establish time-aligned communication slots before data transmission begins. By pre-coordinating transmission timing and beam directions, the system enables multiple links to operate simultaneously without interference, maximizing spatial reuse while managing coordination overhead through structured time division.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10694526B2Adaptive pursuit learning method to mitigate small-cell interference through directionality

Publication Date: 2020.06.23 UNIV OF OULU

AI summary

A learning protocol for distributed antenna state selection in directional cognitive small-cell networks is described. Antenna state selection is formulated as a nonstationary multi-armed bandit problem and an effective solution is provided based on the adaptive pursuit method from reinforcement learning. A cognitive small cell testbed, called WARP-TDMAC, provides a useful software-defined radio package to explore the usefulness of compact, electronically reconfigurable antennas in dense small-cell configurations. A practical implementation of the adaptive pursuit method provides a robust distributed antenna state selection protocol for cognitive small-cell networks. Test results confirm that directionality provides significant advantages over omnidirectional transmission which suffers high throughput reduction and complete link outages at above-average jamming or cross-link interference power.