Distributed Cluster Configuration for Mobile Network QoS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing decentralized wireless networks face challenges in selecting an optimal configuration for radio resources that meet quality of service (QoS) constraints, as current methods either fail to find optimal solutions due to high computational complexity or do not jointly manage frequency and logical channel configurations.
Innovation Solution
A distributed method where cluster leaders collect radio resource parameter values from nodes, evaluate communication quality indices, and select configurations that satisfy the greatest number of nodes while minimizing resource consumption, updating channel configurations as needed based on alarm thresholds and predefined subsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If optimal resource allocation is pursued by including QoS characteristics in the objective function, then quality of service is improved, but computational complexity increases significantly
Solution Approach 1:
The patent segments the resource allocation problem into two phases: a configuration selection phase where predefined configurations are evaluated against QoS constraints, and a resource allocation phase that distributes resources based on selected configurations. This segmentation reduces computational complexity by avoiding the need to optimize all resource parameters simultaneously while still achieving QoS goals.
Solution Approach 2:
The patent employs preliminary action by pre-defining multiple candidate configurations before the actual resource allocation process. These predefined configurations are evaluated in advance against QoS constraints, and the most suitable ones are selected for subsequent resource distribution. This preliminary selection reduces the computational burden during the main allocation process.
2Device complexity
If resource allocation is broken down into multiple steps (link allocation, bandwidth allocation, subcarrier allocation, MCS allocation), then computational complexity is reduced, but optimality of the solution is compromised
Solution Approach 1:
The patent introduces dynamics by enabling flexible adjustment of resource allocation across different steps based on actual network conditions and QoS requirements. The system can dynamically reconfigure link allocation, bandwidth distribution, subcarrier assignment, and MCS parameters in response to changing conditions, allowing the system to achieve optimality even when following a multi-step approach.
Solution Approach 2:
The patent utilizes parameter changes by allowing dynamic modification of allocation parameters at each step based on feedback from network conditions and QoS performance. The system adjusts link allocation, bandwidth, subcarriers, and modulation parameters in response to measured performance, enabling the system to converge toward optimal solutions while maintaining manageable computational complexity.
3Device complexity
If cluster leaders allocate resources independently without centralized coordination, then device complexity is reduced, but overall network efficiency deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where cluster leaders receive QoS performance information from their members and adjust their resource allocation decisions accordingly. This feedback loop enables distributed decision-making while maintaining network efficiency, as each cluster leader learns from actual performance data and optimizes local allocations to contribute to overall network goals.
Solution Approach 2:
The patent applies universality by designing a framework where cluster leaders perform multiple functions: they allocate resources, monitor QoS performance, select configurations, and coordinate with other clusters. This multi-functionality at the distributed level replaces the need for a centralized controller, maintaining network efficiency while reducing device complexity through autonomous operation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for determining the configuration of a set of Ni nodes and/or links of a cluster (10i) having a cluster leader, CHk, characterized in that it comprises at least the following steps: a) triggering an information request (202) by querying one or more Ni nodes or links and requesting said nodes or links to evaluate, for each communication channel, a communication quality index II[c], then selecting a configuration (204) when the cluster leader has received the communication quality indices, b) verifying whether the received quality indices conform to a given constraint, retaining the indices that satisfy the largest number of Ni nodes or links, and triggering a configuration selection procedure (204), c) if the retained configuration satisfies all the queried Ni nodes or links, then updating (206) the list c of channels and communicating (207) this configuration to the nodes or links of the cluster, d) if not,Update (211) the list of communication channels, or remain in the initial configuration of all nodes or links.