In-band Data Corrects Schedule Drift in Dynamic Ad Replacement

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Dynamic ad replacement systems face challenges in accurately detecting ads in media streams due to schedule drift, where the timing of ad appearance deviates significantly from scheduled times, leading to missed ad-replacement opportunities.

Innovation Solution

A computing system monitors for schedule drift by comparing scheduled and actual ad timing data, applying a time offset to align scheduled ad times with actual occurrences, ensuring timely detection and replacement of ads.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If schedule data is used to determine when to monitor for ads, then monitoring efficiency is improved, but ad detection accuracy deteriorates due to schedule drift

Engineering Contradiction:

Improvemonitoring efficiencyVSAvoidad detection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system uses in-band timing data from the media stream as feedback to detect schedule drift. By comparing scheduled ad times with actual ad occurrence times detected through fingerprint matching, the system identifies drift and applies corrective time offsets to future scheduled times, thereby maintaining accurate ad detection despite initial scheduling errors

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts the scheduled time parameter for ad monitoring by applying a time offset. When schedule drift is detected, the system modifies the scheduled time parameter to compensate for the drift, transforming the monitoring approach from static schedule-based to dynamic drift-corrected timing

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If monitoring is performed continuously, then ad detection accuracy is improved, but system complexity and resource consumption increase

Engineering Contradiction:

Improvead detection accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by generating and storing reference fingerprints of the media stream in advance. These reference fingerprints are created before ad replacement decisions are made, enabling efficient later comparison with ad database fingerprints without requiring continuous real-time analysis of the entire media stream

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system segments the ad detection process into distinct phases: generating reference fingerprints from the media stream, comparing these references against ad database fingerprints, and making replacement decisions based on matches. This segmentation allows each component to be optimized independently and reduces overall system complexity

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11695972B2Use of in-band data to correct schedule drift

Publication Date: 2023.07.04 ROKU INC

AI summary

A method and system to help control when to monitor for presence of replaceable advertisements in a media stream. An example method includes determining a time offset based at least on a difference between (i) a time of occurrence of a content event in a media stream as indicated by data carried in-band with the media stream and (ii) a time of occurrence of the content event in the media stream as indicated by schedule data that is not carried in-band with the media stream. Further, the method includes applying the determined time offset as a basis to adjust a scheduled time of an advertisement in the media stream, and using the adjusted scheduled time of the advertisement in the media stream as a basis to control when to monitor for presence of a replacement advertisement in the media stream.