Back top previously visited page (this function is disabled)Back to homepage (P)Back to previous page (E)Jump to next page (V)Display terms (this function is disabled)List of terms (this function is disabled)Print this page (this function is disabled)Jump to sitemap (D)Search in current page (this function is disabled)Help (S)

Building Spatial Databases – Theory / The network model

Learners guide


In this unit the network model along with the navigation methods will be shown.


To make the self-test successfully

The network model

It was intended to define objects by a network of segments with network topology like transportation organization, shipping, logistical challenges, public utility networks and relations. All of them are modelled based on graphs where nodes are defined by business logics rules. Arcs and polylines are distinguished.

As you can see, it is recommended to handle polylines containing many nodes as arcs, especially in routing optimization questions, because these polylines do not have junction points. Many graph algorithms exists for routing questions, however due to the content-length limitations of this book, routing optimization will be left out of discussion.

Figure 27. Nodes, polylines and arcs compose networks [Rigaux et al]

Linear reference

Localization can be determined in various ways. The most common way is by giving x,y,z coordinates along with postal codes, which have been applied for quite a long time. (Let us assume that a friend gives us some coordinates when we ask for his address; we would be surprised, because the postal code based localization method is known and accepted.) Line-based features are like public utility and transportation networks (roads, railways). For describing any point on it, let us use the following method, which is one of the most commonly used models based on the network model. For example, let each kilometre stone at roads and riversides include its exact coordinates; however, in many cases, the localization depends on the location of these stones (e.g. 150 metres from the 123rd kilometre stone along the downstream). Usually, relative localization is used for giving information on road incidents, like an accident occurred on road 4 between the 52nd and 53rd kilometre stones. Kilometre stones on rivers and roads are like break points on map segments, and they can be considered as points.

Line segments representing edges in a network model do not always reside on break points (a 10-km long straight railroad segment, which has only one start and end point, no mid points). It is unnecessary to store nodes on a straight line segment. By storing nodes on each break point in map segments, a linestring can be generated from it. However, it is not correct, because we would introduce many redundant points in the geometry (if larger, multiple level junction points are stored on each path's crossing points, or if a river has been redirected during the construction of a power plant or a dam). Significant length increase may occur, which – if we keep the segment-based logic –, would result the modification of all kilometre-based segments. It is not possible to do so.

Eventually, it does not matter what method is being used for localization if they are congruent and leading to the same result. Methods may vary depending on application areas. A few of these localization methods are geographical coordinates; road segment number; relative distance to segment identifier; relative distance to town border; relative distance from kilometre stones; distance from junction points; distance from significant points. Each application area is building their information systems on different scale levels depending on business logic corresponding to their field of interest. By using different scales and satisfying business logic needs, segmentation is a crucial problem as shown in Figure 28.

Figure 28. Schematic representation of linear referencing applied on different scales and business logic.

Figure 29. A linestring represented by nodes (n1,n2,n3 and n4) matched with segment break-points.

In Figure 29, a Linestring is represented by n1,n2,n3 and n4 nodes coordinates and by linear referencing (segmentation) on the same nodes (e.g. distance from a given segment number, as shown in the following table).

Table 2. Georeference table of nodes n1,n2,n3 and n4.

Table 3. Linear reference table of nodes n1,n2,n3 and n4.

You can open the (larger) image in new window.Table#2.n1, n2, n3, n4 nodeos with georeferencestable_2_full.jpgTable#2.n1, n2, n3, n4 nodeos with georeferences
You can open the (larger) image in new window.Table#3.   n1, n2, n3, n4 nodeos with linear referencestable_3_full.jpgTable#3. n1, n2, n3, n4 nodeos with linear references

Back to table of contents (J)

Dynamic segmentation

Layer-based approach means the segmenting of graphical data, and applies business logic on segments that are part of the graphical database. Due to the popularity of object-oriented methodology, dynamic segmentation may be applied on any attribute of graphical data, resulting a more flexible way of building and designing systems (especially in linear referencing systems, where segments are implemented by linear referencing as shown in Table 4 and Figure 30.)

Table 4. Linear referencing has been applied that resulted in symbol K.

Figure 30. A Linestring with segment break points and K symbol linestring defined by linear referencing as defined in Table 4.

You can open the (larger) image in new window.Table#4   Segment K definition by linear referencetable_4_full.jpgTable#4 Segment K definition by linear reference

Back to table of contents (J)

Új Széchenyi terv
A projekt az Európai Unió támogatásával, az Európai Szociális Alap társfinanszirozásával valósul meg.

A Társadalominformatika: moduláris tananyagok, interdiszciplináris tartalom- és tudásmenedzsment rendszerek fejlesztése az Európai Unió támogatásával, az Európai Szociális Alap társfinanszírozásával, az ELTE TÁMOP 4.1.2.A/1-11/1-2011-0056 projekt keretében valósult meg.
Generated with ELTESCORM framework