Access Point Data Caching via Location Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In networks with multiple WLAN access points, the wireless connection between an access point and a mobile device often has greater bandwidth than the connection between the access point and the wired network (backhaul), causing a bottleneck effect that limits data transfer rates.
Innovation Solution
A method that uses an n-gram language model to predict which access points a mobile device will come into range of in the future, allowing data to be cached at those points for efficient transmission when the device is in range, thereby leveraging the higher bandwidth of the mobile device's connection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is transmitted in real-time from the core network to the mobile device via access points, then the data transfer rate is limited by the backhaul connection bandwidth, but the system complexity is reduced
Solution Approach 1:
The system predicts future access points the mobile device will encounter and pre-caches data at these access points before the device arrives. This preliminary action allows data to be ready for immediate transmission when the device connects, bypassing the slow backhaul connection during the actual data transfer and thereby increasing the effective data transfer rate without requiring fundamental system redesign
2Productivity
If data is cached at multiple access points to overcome backhaul limitations, then data transfer efficiency is improved, but the complexity of data management and prediction increases
Solution Approach 1:
The system introduces a prediction module that acts as an intermediary between the core network and access points. This module uses location data and n-gram language models to predict future access points, enabling intelligent data routing and caching decisions. The prediction module manages the complexity of determining where to cache data, while access points simply execute caching instructions, thereby improving data transfer efficiency without making individual access points overly complex
3Area of stationary object
If the mobile device moves through multiple access points quickly, then the available transmission time at each access point is reduced, but the coverage area is expanded
Solution Approach 1:
The system predicts the sequence of access points the mobile device will encounter based on location data and movement patterns. By pre-caching data at these predicted access points in advance, the system ensures data is ready for immediate transmission when the device arrives, compensating for the brief transmission window at each access point and enabling effective data delivery across expanded coverage areas
Solution Approach 2:
The system creates copies of data at multiple predicted access points along the device's trajectory. Instead of relying on a single transmission opportunity, the data is replicated at multiple locations, ensuring the device can receive data from the nearest available access point with cached data, thereby overcoming the limitation of brief transmission times while maintaining expanded coverage
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system for transmitting data via a core portion (50, 60, 70) of a computer network to a mobile device (10) is described. The network includes a plurality of access points (21, 22, 23, 24) wherein each access point includes a memory (121, 122, 123, 124) where data for onward transmission to the mobile device may be cached and is operable to communicate with the mobile device over a short-range wireless connection, when the mobile device is in range thereof, and with the core portion of the network over a backhaul connection on a substantially continuous basis. The system comprises: a location collection server (50) for receiving location data from the mobile device; an access point prediction engine (60) for generating predictions of one or more of the access points which the mobile device is considered likely to come in range of in the future based on recent location data received at the location collection server (50); and a content delivery server (70) for transmitting the data or a portion thereof to the one or more predicted access points via the backhaul connection for caching at the access points and for onward transmission to the mobile device in the event that the mobile device comes into range of the access point, or one or more of the access points where data is transmitted to more than one predicted access point.