Vehicle Crowd Sensing Resampling for Communication Cost Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vehicle crowd sensing systems face inefficiencies in data usage and accuracy due to redundant detection reports for confirmed events, leading to increased cellular communication costs without ensuring accurate event detection.
Innovation Solution
A method involving the development of an inherent error model and action model to determine resampling instructions for detection reports, which decides whether vehicles should upload or not upload reports based on the reliability of event detection, using a hybrid criteria model and PID policy evaluation to minimize unnecessary data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all vehicles upload detection reports for every event, then event detection accuracy is improved through corroboration, but cellular communication costs and data usage increase significantly
Solution Approach 1:
The system changes the parameter of data upload probability based on the confidence level of the inferred event. When confidence is high (ground-truth likely), the upload probability is reduced or eliminated. When confidence is low, upload probability increases. This dynamic parameter adjustment resolves the contradiction by adapting communication behavior to event reliability.
Solution Approach 2:
Instead of requiring all vehicles to upload reports (excessive action), the system applies partial action by selecting only a resampled subset of vehicles to upload reports. The resampling ratio is adjusted based on event confidence, applying full upload requirement only when necessary and reducing it when ground-truth is likely, thus minimizing communication costs while maintaining detection accuracy.
2Reliability
If multiple detection reports are collected for corroboration, then event reliability is improved, but data transmission redundancy increases
Solution Approach 1:
The system dynamically changes the required number of corroborating reports (resampling ratio) based on the confidence level of the inferred event. When confidence is high, the required number of additional reports is reduced to zero or minimal. When confidence is low, more reports are required. This resolves the contradiction by adapting data collection requirements to the actual reliability needs of each event.
3Loss of energy
If resampling ratio is increased to reduce data usage, then communication costs are reduced, but event detection accuracy may be compromised
Solution Approach 1:
The system uses confidence levels as a parameter to dynamically adjust the resampling ratio. High confidence events receive low or zero resampling ratios (minimal additional reports required), while low confidence events receive higher resampling ratios (more reports required). This resolves the contradiction by allocating communication resources based on actual detection needs rather than applying a uniform ratio.
Data Source
AI summary
A vehicle crowd sensing system and method of selective sensing for the vehicle crowd sensing system. The method, in one implementation, involves receiving a plurality of detection reports from a first set of vehicles, each detection report including an event, a position qualifier of the event, and a severity qualifier of the event; developing an inherent error model for the event that includes a compilation of the position qualifiers of the event and the severity qualifiers of the event; and determining a resampling instruction for the event. The resampling instruction is based on an action model and the action model is at least partly a factor of the inherent error model.