Touch Input Coordinate Smoothing for Mobile Terminal Scrolling

The patent applies preliminary action by predicting future touch coordinates based on historical touch data before the actual touch occurs. The system pre-calculates expected touch positions using prediction algorithms (such as linear prediction or Kalman filtering) and prepares correction values in advance, allowing the display to update smoothly without waiting for actual touch feedback. This resolves the contradiction by improving display smoothness through proactive coordinate prediction while keeping the correction mechanism relatively simple.

The patent implements feedback mechanisms by continuously comparing predicted touch coordinates with actual touch coordinates and using the difference (error) to refine future predictions. The system monitors touch input deviations and adjusts prediction parameters dynamically, creating a closed-loop correction system. This feedback approach improves display smoothness by continuously optimizing coordinate accuracy while maintaining manageable system complexity through iterative refinement rather than complex upfront calculations.

The patent applies partial action by selectively smoothing only the coordinate data that requires it, rather than processing all data uniformly. The system identifies specific coordinate sequences that benefit from smoothing (such as drag operations) and applies smoothing algorithms only to those cases, while leaving other touch operations (such as simple taps) unprocessed. This approach improves scrolling smoothness where needed while maintaining high processing speed for operations that don't require smoothing, thus resolving the contradiction between smoothness and speed.

The patent implements parameter changes by dynamically adjusting the smoothing intensity based on the type of touch operation and context. The system modifies smoothing parameters (such as filter strength or prediction horizon) according to the detected touch pattern, allowing aggressive smoothing for scrolling operations while using minimal or no smoothing for rapid sequential touches. This adaptive parameter adjustment improves scrolling smoothness without significantly impacting overall processing speed, resolving the contradiction between these two parameters.

The patent applies segmentation by dividing the coordinate prediction and smoothing process into distinct stages: initial prediction phase, refinement phase, and validation phase. Each stage uses different levels of computational complexity appropriate to its purpose. The system segments touch operations into categories (such as taps, drags, pinches) and applies different prediction algorithms to each category, using simpler models for common operations and more complex models only when necessary. This segmented approach improves touch coordinate accuracy while minimizing overall computational energy consumption by avoiding unnecessary complex calculations.

The patent implements self-service by designing prediction algorithms that use the device's own historical touch data as the primary input source, eliminating the need for external reference data or complex calibration processes. The system learns from its own operational patterns and automatically adjusts prediction parameters based on accumulated touch history, reducing the need for energy-intensive external computations or manual calibration. This self-service approach improves coordinate accuracy using the device's existing resources while keeping computational energy consumption relatively low.