Mobile Signalling Trajectory Segmentation for Antenna Hopping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining trajectory data from mobile device signalling events are unreliable, costly, and lack contextual information, particularly failing to account for antenna hopping and varying inter-antenna distances, leading to inaccurate trajectory determination.
Innovation Solution
A device and method that preprocess, process, and post-process mobile communication device signalling events to identify long stationary states and trajectory segments, applying threshold durations and distances to label segments as stop or movement types, and refine these labels to improve accuracy, while managing antenna hopping and smoothing trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a high distance threshold is used to detect large movements, then wide vision is achieved, but small movements within stops are missed
Solution Approach 1:
The patent segments the trajectory analysis into two distinct parts: stop detection using a first (lower) distance threshold to identify stationary periods with small movements, and movement detection using a second (higher) distance threshold to identify transitions between stops. This segmentation allows each threshold to be optimized for its specific purpose without compromising the other.
Solution Approach 2:
The patent dynamically adjusts the distance threshold based on the current state (stop or movement detection phase). During stop detection, a lower threshold is applied to capture small movements within the stop area. During movement detection, a higher threshold is applied to identify significant displacements between stops, thereby adapting the threshold to the operational context.
2Measurement precision
If a low distance threshold is used to detect small movements, then precise vision is achieved, but noise from antenna hopping is generated
Solution Approach 1:
The patent performs preliminary stop detection using a lower distance threshold before conducting movement detection. By first identifying stop periods and their geographic boundaries, the system establishes a reference framework that allows subsequent movement detection to distinguish between legitimate small movements within a stop and false positives from antenna hopping, thereby filtering noise in advance.
Solution Approach 2:
The patent introduces stop detection as an intermediary step between raw signaling events and final trajectory determination. This intermediary process filters out antenna hopping noise by establishing stop boundaries based on lower threshold criteria, thereby mediating between the conflicting requirements of detecting small movements and avoiding false positives.
3Device complexity
If arbitrary distance thresholds are applied, then processing simplicity is maintained, but trajectory accuracy is substantially impacted
Solution Approach 1:
The patent changes the parameters (distance thresholds) based on the detection phase: a first distance threshold for stop detection and a second distance threshold for movement detection. This parameter adaptation allows the system to maintain processing simplicity while significantly improving trajectory accuracy, as each threshold is optimized for its specific detection purpose rather than using a single arbitrary value.
4Productivity
If time thresholds are used to classify stationary states, then processing speed is improved, but accuracy in distinguishing stops from antenna hopping deteriorates
Solution Approach 1:
The patent adds the spatial dimension (distance thresholds) to the temporal dimension (time thresholds) already used in prior art. By combining both dimensions, the system maintains the processing speed benefits of time-based filtering while adding spatial criteria to accurately distinguish between genuine stops and antenna hopping events, thereby resolving the accuracy-speed trade-off.
Data Source
AI summary
A device for determining a trajectory receives mobile or wireless network signalling events each comprising a mobile communication device identifier, a timestamp, and a signalling identifier associated with a location and a location radius, preprocesses the associated signalling events to identify one or more long stationary state groups, trajectory segments and long stationary state segments, smooths events at the boundaries of long stationary state segments, processes them to associate each segment with a label belonging to a group comprising a movement and a stop, returns the long stationary state segments and the trajectory segments as trajectory data.

