Set-Top Box Return Path Detection for Media Device On/Off State
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for accurately determining the on/off state of media devices connected to set-top boxes (STBs) are inadequate, leading to inaccuracies in audience measurement due to STBs remaining powered on without media devices being active.
Innovation Solution
A machine learning algorithm trained using common homes data to predict the on/off state of media devices based on return path data (RPD) from STBs, leveraging features extracted from RPD to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If STBs remain powered on continuously, then availability for media delivery is maintained, but accuracy of audience measurement deteriorates due to inability to detect when media devices are actually off
Solution Approach 1:
The patent introduces return path data as an intermediary indicator to infer the state of the media device. Instead of directly monitoring the media device, the system uses RPD from the STB as a proxy to determine whether the media device is on or off, resolving the contradiction between STB availability and measurement accuracy.
Solution Approach 2:
The patent replaces direct physical monitoring of media device power states with a data-based inference system. Machine learning algorithms analyze return path data patterns to substitute for direct mechanical detection of device on/off states, enabling accurate measurement without requiring STBs to be turned off.
2Device complexity
If traditional on/off detection methods are used, then system complexity is minimized, but measurement accuracy deteriorates due to inability to distinguish STB power state from media device power state
Solution Approach 1:
The patent changes the detection parameter from direct power state monitoring to analysis of return path data characteristics. By analyzing patterns in RPD such as tuning data, viewing data, and communication timing, the system infers media device states without directly monitoring power consumption or physical state.
Solution Approach 2:
The patent replaces simple binary detection with machine learning-based pattern recognition. Algorithms analyze multiple parameters in return path data to substitute for direct device state monitoring, achieving higher accuracy while maintaining relatively simple system architecture.
3Measurement precision
If RPD is collected and analyzed using machine learning, then measurement accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent performs preliminary actions by collecting and storing return path data over extended periods before analysis is needed. Training data is accumulated in advance, allowing machine learning models to be trained offline, which reduces real-time processing complexity while maintaining high measurement accuracy.
Solution Approach 2:
The patent collects excessive return path data beyond what is minimally required for basic measurement. By gathering comprehensive RPD including tuning data, viewing data, and communication metadata, the system provides rich training material for machine learning models, improving accuracy while managing complexity through selective feature extraction.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Example methods disclosed herein include accessing common homes data for a group of common homes, the common homes data including return path data and panel meter data. Disclosed example methods also include accessing common homes data for a group of common homes, the common homes data including first return path data and corresponding panel meter data associated with respective ones of the common homes, grouping the common homes data into view segments, classifying the view segments based on whether the return path data in respective ones of the view segments has matching panel meter data to determine labeled view segments, generating features from the labeled view segments, training a machine learning algorithm based on the features, and applying second return path data to the trained machine learning algorithm to determine whether a media device associated with the second return path data is on or off.