Home Articles The extraction of topological features from LIDAR DEM data with the application...

The extraction of topological features from LIDAR DEM data with the application of wireless systems design


Antony Winston Roullier-Callaghan, Prof. M. Al-Nuaimi, Dr. D. Kidner

I. Introduction
One effective solution to wireless communications systems design in urban areas is to utilise Digital Elevation Model (DEM) datasets along with Geographical Information Systems (GIS) techniques within radiowave propagation models. The aim of this project is to design and implement a ‘Radio Wave Coverage and Planning System (RAWCAPS) which can predict the signal field strength from a pre-positioned transmitter in a given topographical region. The emphasis of the extracted features is to identify and classify feature outlines such as building footprints and regions of vegetation as these particular features would have the maximum effect on radiowave propagation modelling.

II. LIDAR Dataset
A certain constraint for the systems being developed is the sole use of LIDAR data as a data source.

This is mainly due to the cost effectiveness of LIDAR as opposed to existing topological feature vector data. The LIDAR is used in the form of Digital Elevation Model (DEM) data. The DEMs used in this study where spaced at 1m intervals, with the extent of each site being 500m x 500m. Six sites of the Cardiff UK, LIDAR DEMs were designated for their terrain specific properties. One attribute of the LIDAR DEM, that causes concern is the accuracy of it. As can be seen in Figure 1, ground and roof levels are not separated by thin edges constituting realistic wall widths, thus the need for various processing techniques such as thinning.

Figure 1. LIDAR DEM data

III. Topological feature extraction (TOFEX) system
The ultimate goal is to extract terrain features solely from the LIDAR dataset using GIS software. The features would be represented within the software applications as vector polygons which can then be exported as a vector output file. The vector information can be exported to an external data file for various purposes. Various GIS software applications were used in the implementation of the features. The method includes various GIS and image processing techniques including filtering, smoothing, edge detection, vectorisation, and feature enhancing techniques. In Figure 2, features represented in brown, can be seen, superimposed onto a 2D image of the DEM.

Figure 2. Extracted Polygons with height data from LIDAR DEM data
The yellow outlines depict the idealised feature outlines from the Ordnance Survey LandLine.Plus vector data. Replication of these vector outlines is the ultimate goal of the TOFEX system. The features can be classified using pattern recognition or mathematical morphology.

IV. Wireless communication as an application of the GIS ‘TOFEX’ system

Various applications of the extracted features exist. One such application is in the wireless communication systems design. The extracted features can be used in existing microcell ray tracing models and propagation modelling. Work in this project has involved the implementation of a visibility analysis of the DEM data, implementation of LOS and single knife-edge diffraction propagation techniques as depicted in Figure 3.

Figure 3. Signal Strength Minus LOS and diffraction losses