Link Group Controller Admission Control for Wireless Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for admission control and path selection in communication networks with wireless links lack a generic solution that can handle diverse radio technologies and heterogeneous scenarios, as they fail to efficiently consider resource sharing, interference, and rate adaptation across different wireless link technologies.
Innovation Solution
A model-based approach is implemented where communication links are grouped, with each group having a Link Group Controller that computes technology-independent metrics for capacity and cost, allowing for efficient admission control and path selection across various wireless and wired technologies, including OFDMA and CSMA/CA, by modeling links as unidirectional point-to-point and using scheduling units for capacity measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If information on link state is pushed from LGCs to RMEs whenever link state changes, then the RME has current information for admission control and path selection, but signaling cost becomes excessively high
Solution Approach 1:
The patent implements periodic sampling of link state information at predetermined intervals instead of continuous pushing. The RME receives link state information at regular time intervals, which reduces signaling overhead while maintaining sufficiently current information for admission control and path selection decisions.
Solution Approach 2:
The patent uses a simplified link state model that captures only the essential parameters needed for admission control and path selection, rather than transmitting complete link state information. This partial information approach reduces signaling cost while providing sufficient data for resource management decisions.
2Loss of energy
If information on link state is pulled by RMEs from LGCs when admission control or path selection is needed, then signaling cost is reduced, but the RME's tasks are delayed due to querying time
Solution Approach 1:
The patent pre-fetches link state information at predetermined intervals before it is actually needed for admission control or path selection. The RME receives and stores link state information periodically, so when admission control or path selection needs to be performed, the information is already available, eliminating query delays.
Solution Approach 2:
The system uses periodic sampling to proactively collect link state information at regular intervals, transforming the reactive pull model into a proactive periodic update model. This reduces both signaling cost (compared to continuous pushing) and decision delay (compared to on-demand pulling).
3Loss of energy
If RME keeps track of available resources based on a model with infrequently updated parameters, then signaling overhead is reduced, but the solution cannot efficiently handle diverse radio technologies and heterogeneous scenarios
Solution Approach 1:
The patent develops a universal link state model and resource management methodology that can be applied across different radio technologies (WiFi, WiMAX, cellular) and network scenarios. The model uses technology-independent parameters that can capture the essential resource state regardless of the underlying radio technology, enabling the same admission control and path selection mechanisms to work heterogeneously.
Solution Approach 2:
The patent transforms technology-specific link state parameters into a standardized set of parameters suitable for generic resource management. By changing the parameter representation to be technology-agnostic while retaining the ability to model different radio technologies, the system achieves both low signaling overhead and broad adaptability.
4Measurement precision
If wireless link capacity is measured and observed at the link itself by LGC, then accurate current link state information is obtained, but the RME cannot access this information without high signaling cost
Solution Approach 1:
The LGC measures link state information locally with high precision and transmits this information to the RME at periodic intervals. This periodic transmission maintains measurement precision while significantly reducing signaling cost compared to continuous or on-demand information exchange.
Solution Approach 2:
The LGC acts as an intermediary that performs local measurements and maintains a local model of link state, then selectively shares this information with the RME. This intermediary approach allows precise local measurement while reducing the burden of continuous information exchange over the network.
Data Source
Figure 1a~1b
Figure 2
Figure 3
AI summary
A method for supporting admission control and/or path selection in a communication network, said network including a plurality of wireless communication links, possibly of different link technologies, established between a plurality of network nodes, characterized in the steps of partitioning said communication links of said network into groups of communication links - link groups -, assigning each link group an associated link group controller that is configured to observe and/or measure the performance and/or quality of said link- group, based on said observations and/or measurements, computing estimates of metrics for the capacity of the respective link group and the costs for using each of the communication links of the respective link group, and transmitting said estimates to at least one resource management entity being configured to take admission control and/or path selection decisions on the basis of said estimates. Furthermore, a communication network with admission control and/or path selection supporting functionality is described.