DTED maps consist of a square grid structure of cells covering the surface of Earth, with terrain elevation values sampled at the intersection points.

There are three distinct detail levels: Level 0, Level 1, and Level 2. DTED Level 0 contains the least amount of detail while requiring the minimum amount of storage space.

Higher levels require several gigabytes to represent the whole world, while at most a few hundred MBs will suffice for DTED Level 0 storage, depending on the data types used in the datafiles.

• A 3D Mobile Navigation application has been developed for civil and defense uses.
• Mobile devices inherently possess much less computation and storage capacity.
• The Digital Terrain Map used in this application features a 240-meter resolution and spans a mountainous area of 14,400 km².
• Capacity limitations lead directly to the adoption of DTED Level 0 types of digital maps in mobile handheld applications.

DTED Level 0 cells store the absolute maximum and minimum elevation markers of their corresponding Level 1 parent cells. Explicitly defining coordinates for these extremums increases the precision of local terrain mesh generation, pinning down structural variances relative to the broader surface average.

To calculate the spatial coordinates of a maximum or minimum feature point inside the grid cell boundaries, intermediate segmentation markers are generated based on the direct elevation deltas between corner points and the destination peak or valley target value.

As an example execution pathway, target boundaries p5-I and p5-II are localized inside the cell plane and are then bisected an additional time to pinpoint p5 (the designated maximum peak). Similarly, node p6 serves as the computed index representing the localized minimum valley point.

Mathematically, the smaller the structural elevation variance a specific corner boundary point maintains relative to the peak or valley threshold, the closer its coordinate position will sit next to that respective maximum or minimum node index.

If diagonally opposing matrix corners hold identical altitude values (e.g., the NW point matches the SE point elevation, and the NE point mirrors the SW point elevation), a mathematical conflict occurs: both the singular maximum and minimum calculations resolve onto the exact same coordinate junction—the exact geometric center of the cell space.

To resolve this collision scenario, a circular tracking distribution algorithm deploying multiple secondary maximum and minimum node offsets is applied across the active region.

If the delta between the average and the minimum value (AVG – MIN) tracks smaller than the delta between the maximum and the average value (MAX – AVG), it indicates that the vast majority of the spatial landmass area sits lower than the mean elevation, clustering nearer to the valley floor.

Positioning the singular maximum anchor point directly in the center of the cell grid while radiating four matching minimum anchors outward onto an external perimeter circle balances the resulting mesh surface geometry to fit this specific skew profile.

Even though only a marginal fraction of the grid elements within our target study region mathematically encounter the geometric criteria forcing a circular distribution, the technique is systematically applied as a stabilization override whenever maximum and minimum anchors map too closely together.

The total volume of explicit control nodes required to resolve circular geometry maps out to exactly three times the density profile of the standard singular approach. Consequently, this method outputs superior fidelity and smoother contour surface maps.

However, because the associated programmatic overhead involving matrix inversion and linear transformation steps scales heavily on low-power mobile architectures, the application threshold determining when to tag cells for circular processing must be carefully managed to remain within sensible hardware bounds.