Coded-Caching Serving Node Predicts Sub-Data Files
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Coded-caching in wireless communication networks increases backhaul traffic during low-traffic periods and reduces hit-rate due to outdated cache knowledge, leading to inefficient data transmission and energy consumption, as well as interference with neighbor nodes.
Innovation Solution
A serving node predicts data files to be requested by user terminals and divides them into sub-files, transmitting unique sets to cache nodes during low-traffic periods and providing complementary sub-files during high-traffic periods to enable efficient re-creation of requested files, reducing backhaul load and improving hit-rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If coded-caching is implemented to reduce peak backhaul traffic during high-traffic periods, then backhaul load is reduced during delivery phase, but data traffic increases significantly during low-traffic periods
Solution Approach 1:
The system performs preliminary data placement during low-traffic periods by predicting future requests and pre-loading corresponding data files into cache nodes. This advance preparation enables the system to handle peak demand during high-traffic periods without increasing overall data traffic, as the cached data is retrieved locally rather than transmitted through the backhaul.
Solution Approach 2:
The patent implements dynamic adjustment of placement and delivery phases based on traffic conditions. The serving node predicts future data requests and adjusts the timing of data placement accordingly, allowing the system to flexibly switch between placement-mode and delivery-mode operations. This dynamic approach optimizes backhaul usage by performing placement during low-traffic periods when backhaul capacity is underutilized.
2Loss of time
If coded-caching is implemented to reduce backhaul traffic, then transmission delay is reduced, but hit-rate decreases due to outdated cache knowledge
Solution Approach 1:
The system incorporates feedback mechanisms where cache nodes provide information about their cached data and request patterns back to the serving node. This feedback enables the serving node to update its predictions about future requests and adjust the placement strategy accordingly. The feedback loop ensures that cache knowledge remains current, improving hit-rate while maintaining low transmission delay through efficient coded-caching operations.
Data Source
AI summary
The present disclosure relates to a serving wireless communication node adapted to predict data files (A, B, C) to be requested by at least two served user terminals (2, 3) and to form predicted sub-data files (A1, A2; B1, B2; C1, C2). In a placement phase, the serving node is adapted to transmit predicted sub-data files (A1, B1; A2, B2) to cache nodes (APC1, APC2), each cache node (APC1, APC2) having a unique set of predicted different sub-data files of different predicted data files, and to receive requests (RA, RB) for data files from the served user terminals (2, 3). In a delivery phase, the serving node is adapted to transmit an initial complementary predicted sub-data file (Formula I) to the cache nodes (APC1, APC2), comprising a reversible combination of the remaining predicted sub-data files (A2, B1) for the files requested. If a cache node (APC1, APC2) requests re-transmission (RA2; RB1) of a predicted sub-data file (A2, B1), the serving node (AP) is adapted to transmit a further complementary predicted sub-datafile (Formula II) to the cache nodes (APC1, APC2), the file comprising a reversible combination of at least one re-transmitted predicted sub-data file (A2, B1) and at least one new predicted sub-data file (C1).