Edge Network Caching via Mobility Prediction Error

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributing content with low latency to mobile devices with complex mobility patterns poses challenges due to limited memory in network edge devices, as existing methods lack an intelligent decision-making mechanism for caching content based on mobility predictions.

Innovation Solution

A network device at the edge intelligently decides whether to cache content by considering the popularity of the content in its region and the prediction error of a mobile device's transition probability, using a utility function that balances these factors to optimize memory usage and latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If content is cached at network edge devices to reduce latency, then content delivery speed is improved, but memory resources are consumed

Engineering Contradiction:
Improvecontent delivery latencyVSAvoidmemory resource consumption
Core Design Contradiction:
SpeedVSQuantity of substance

Solution Approach 1:

The patent changes the decision parameter from static popularity metrics to dynamic mobility prediction accuracy. By using prediction error metrics that quantify the reliability of mobility forecasts, the system adapts caching decisions based on the actual accuracy of mobility predictions, allowing more intelligent allocation of limited memory resources at network edge devices.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system implements feedback by measuring prediction errors and using this information to adjust caching decisions. The prediction error metric provides feedback on the reliability of mobility forecasts, enabling the system to learn from past predictions and improve future caching decisions, thereby optimizing both latency reduction and memory utilization.

Inventive Principle:
Principle #23Feedback

2Productivity

If caching decisions are based on popularity lists, then frequently accessed content is cached, but content may be cached for devices with unpredictable mobility patterns, wasting memory

Engineering Contradiction:
Improvecontent access efficiencyVSAvoidmemory wastage
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent introduces a new parameter - prediction error metric - that complements or replaces traditional popularity-based metrics. This parameter change enables the system to distinguish between content that should be cached based on actual mobility predictability versus content that would waste memory resources on devices with unpredictable movement patterns.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system applies local quality by making caching decisions specific to each device's mobility characteristics rather than using uniform popularity-based rules. By evaluating prediction errors for individual devices, the system tailors caching strategies to local conditions, caching content only for devices with predictable mobility patterns while avoiding memory wastage on devices with erratic movement.

Inventive Principle:
Principle #3Local quality

3Loss of time

If content is pre-fetched based on mobility predictions, then latency is reduced, but predictions may be inaccurate leading to incorrect caching decisions

Engineering Contradiction:
Improvecontent delivery timeVSAvoidprediction accuracy
Core Design Contradiction:
Loss of timeVSReliability

Solution Approach 1:

The system performs preliminary actions by pre-fetching content based on mobility predictions, but crucially, it does so conditionally - only when prediction accuracy thresholds are met. This preliminary action is taken in advance for high-confidence predictions, reducing latency for those cases while avoiding premature caching decisions for low-confidence predictions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system provides beforehand cushioning by using prediction error metrics to buffer against inaccurate predictions. By measuring and accounting for prediction errors in advance, the system creates a safety mechanism that prevents incorrect caching decisions, ensuring that pre-fetching operations are only performed when the confidence level is sufficient to warrant the memory resource allocation.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentEP3827574B1In-network content caching exploiting variation in mobility-prediction accuracy
Publication Date: 2024.10.16 CISCO TECHNOLOGY INC
  • EP3827574B1 patent drawingFigure 1
  • EP3827574B1 patent drawingFigure 2
  • EP3827574B1 patent drawingFigure 3

AI summary

A network device in a network edge receives content directed to a mobile device attached to the network edge from an upstream network device and forwards the content toward the mobile device. The network device makes a decision whether to cache the content based at least in part on a popularity of the content in a region covered by the network device and a prediction error for an estimated probability that the mobile device will transition from the region to another region. The popularity is directly correlated with a first bias toward caching the content. The prediction error is inversely correlated with a second bias toward caching the content. The decision is implemented: the network device either caches the content or foregoes caching the content, in accordance with the decision.