Real-Time Traffic Prediction Using Wireless Positioning Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic analysis methods fail to accurately predict real-time traffic flows and user gatherings during public events, as they rely on stationary stochastic processes and require extensive empirical data, which is not feasible during non-standard conditions like public happenings.
Innovation Solution
A method and system that collect and analyze real-time positioning data from wireless communication networks to predict user gatherings by calculating differences in traffic flows between a target area and surrounding areas using reference days with similar statistical behavior, allowing for the forecasting of user arrivals and attendance times during public events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If stationary stochastic processes are used to predict user terminal movement, then the prediction method is simple and efficient, but it fails to accurately predict traffic flows during public happenings when user behavior becomes non-stationary
Solution Approach 1:
The patent transitions from static stochastic models to dynamic prediction approaches that adapt to changing conditions. The system continuously updates prediction models based on real-time data from public happenings, allowing the prediction mechanism to evolve from stationary to non-stationary conditions while maintaining accuracy.
Solution Approach 2:
The invention changes the fundamental parameters of the prediction model by incorporating time-varying factors specific to public happenings. Instead of using fixed stochastic parameters, the system adjusts prediction parameters dynamically based on event characteristics, location, and temporal patterns, enabling accurate prediction under non-stationary conditions.
2Quantity of substance
If Origin-Destination matrices are used for traffic analysis, then comprehensive traffic data can be obtained, but a large amount of empirical data collection is required which is not feasible during non-standard conditions
Solution Approach 1:
The patent performs preliminary actions by pre-processing and storing historical traffic data, event information, and spatial patterns before actual prediction is needed. This preparation allows the system to quickly generate predictions during public happenings without requiring extensive real-time data collection, reducing the time loss while maintaining data comprehensiveness.
Solution Approach 2:
The system creates simplified copies or representations of complex traffic patterns through aggregated statistical models. Instead of processing every individual data point in real-time, the system uses pre-computed statistical profiles and patterns that replicate traffic behavior, enabling comprehensive analysis with reduced data collection requirements and time investment.
3Measurement precision
If detailed zone partitioning is applied to improve analysis precision, then higher level of detail is achieved, but the complexity of data processing and computation increases
Solution Approach 1:
The patent applies segmentation by dividing the geographic area into multiple zones or regions for analysis. This segmentation allows detailed traffic analysis at different spatial resolutions while managing computational complexity through hierarchical processing - coarser zones for overview analysis and finer zones for detailed predictions, enabling precision without proportional increase in system complexity.
Solution Approach 2:
The invention implements local quality by applying different levels of analysis detail to different spatial locations based on their characteristics. High-detail analysis is applied to areas with public happenings or high traffic density, while simpler processing is used for low-activity areas. This selective approach maintains measurement precision where needed while reducing overall processing complexity.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
A method (200) is provided. The method is implemented by a system (100) coupled with a wireless communication network (105), comprising predicting the number of users of user terminals gathering in a target area (A(1)) within a region of interest (107) for attending a public happening occurring in a first day, said region of interest being under radio coverage by said wireless communication network, said predicting comprising performing the following operations during said first day: - collecting, from the wireless communication network, records relating to user terminals positioning data; - partitioning (230) a portion of the region of interest into an intermediate area (A(2)) surrounding the target area and into an external area (A(3)) which surrounds the intermediate area (A(2)); - for each observation time slot of a plurality of observation time slots of said first day, calculating (235) a first term indicative of the number of user terminals which are moving from the external area to the intermediate area during said observation time slot based on the collected records; - for each observation time slot, calculating (240) a second term indicative of the number of user terminals which, during a corresponding reference time slot of a reference day preceding in time said first day, moved from the external area to the intermediate area, said calculating the second term being based on records which were collected during said reference day, - for each observation time slot, calculating (250) the difference between said first term and said second term; -for each observation time slot, predicting (260) the number of new user terminals which will enter in the target area for attending the public happening in a forthcoming period subsequent to said observation time slot based on said calculated difference.