Fabric-Aware Neighbor List Pruning for Wi-Fi Roaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current WLAN deployments face inefficiencies in client roaming due to ignorance of backend infrastructure and multi-tenancy Wi-Fi services, leading to varying roam times, often taking hundreds of milliseconds, as they do not account for fabric node assignments and service level agreements (SLAs) when determining neighbor lists for client devices.
Innovation Solution
A wireless network controller prunes neighbor lists based on predefined assignments corresponding to Access Points (APs) and client devices, focusing on APs within the same fabric node and service provider, to steer clients to optimized APs for faster roaming, using information from 802.11k and 802.11v BSS transition requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If complete neighbor lists are provided to client devices, then clients can roam to any available AP, but roaming time increases to hundreds of milliseconds due to backend infrastructure variations
Solution Approach 1:
The neighbor list is segmented into multiple categories based on fabric node assignments and service provider affiliations. Instead of providing a single comprehensive list, the system divides the neighbor APs into groups (e.g., same fabric node, different fabric node, different service provider) and provides prioritized subsets to client devices, enabling faster roaming decisions while maintaining access to diverse AP options.
Solution Approach 2:
The system performs preliminary filtering and categorization of neighbor APs before providing the list to client devices. By pre-processing the neighbor list to identify and prioritize APs within the same fabric node and service provider, the system prepares optimized roaming options in advance, allowing clients to make rapid roaming decisions without performing extensive evaluations during the roaming event itself.
2Adaptability or versatility
If all neighboring APs are included in the neighbor list, then clients have maximum roaming options, but connection denials increase due to SLA mismatches and multi-tenancy restrictions
Solution Approach 1:
The neighbor list is customized with local quality characteristics based on the specific client device's service provider affiliation and the APs' fabric node assignments. Each client receives a tailored neighbor list that emphasizes APs compatible with their service provider and fabric node, rather than a uniform list for all clients. This local optimization ensures that clients are steered toward APs where they are most likely to successfully connect based on their specific service agreements.
3Device complexity
If roaming optimizations ignore backend infrastructure and SLAs, then implementation is simpler, but roaming performance varies widely due to fabric node assignments
Solution Approach 1:
The controller acts as an intermediary between the complex backend infrastructure (fabric nodes, SLAs, multi-tenancy arrangements) and the client devices. It absorbs the complexity of interpreting fabric node assignments and service provider agreements, translating this information into simplified, prioritized neighbor lists that clients can use for fast roaming. This intermediary approach shields clients from infrastructure complexity while maintaining high roaming efficiency.
Data Source
AI summary
Embodiments herein describe pruning a neighbor list sent to a client device using a predefined assignment corresponding to an AP or the client device which can reduce the time required for the client device to roam between APs in a WLAN deployment. The pruned neighbor list serves as a Wi-Fi client steering mechanism to affect which AP the client device chooses when roaming. In one embodiment, a controller uses information regarding fabric nodes in the backend infrastructure to prune the neighbor list to include only the APs that are assigned to the same fabric node as the current AP. In another embodiment, the controller considers a service provider corresponding to the client device when pruning the neighbor lists. The list may include only the neighboring APs assigned to provide Wi-Fi access for the same service provider.