Home > Research and Market Data > Technology research > Research > Propagation Studies > Indoor-outdoor interface

For a range of wireless communication applications including mobile telephony, wireless broadband last mile, UWB services and mesh based networking, accurate prediction of signal propagation involves modelling of signals moving from indoors to outdoors and vice versa. In particular, we expect fixed and mobile communications to converge over coming years. This will likely require excellent indoor coverage. While this may be achieved with indoor transmitters, it will be important to understand leakage of signal outside the building which could cause interference to other users.

Broadly there are two approaches to modelling this situation: a simple, high level, empirical approach, and a complex, deterministic approach. The objective of this work is to obtain a model that is simple while being more reliable and accurate than existing approaches. Therefore this work has used the complex modelling approach, but is then drawing general rules from it to form a more accurate, simple empirical model. Further detail is given below.

Derivation of an accurate empirical model from a detailed, deterministic model

A number of simple empirical models are available based mainly upon predicting the signal losses for each penetration of a wall or a floor. These can generate coarse results for typical scenarios, but since they do not capture the true behaviour of radio signals, their accuracy is limited. For instance, radio energy can travel through stairwells and lift-shafts bypassing floors and ceilings. It is also known that transmission between floors can be via windows and reflection from a nearby building.

Conversely, much more complex deterministic models exist which perform intensive calculations on a detailed 3-dimensional model of a specific building environment. Whilst this type of model can produce accurate results for a specific case, it requires high-resolution data and has a large computational requirement.

A more accurate empirical model could be developed through making large numbers of measurements in many different types and configurations of buildings. Unfortunately, this approach would be hugely expensive. Instead, this work has utilised the complex and accurate 3D deterministic models available to derive a simpler and faster empirical model. Figure 1 shows an example of such a deterministic model, allowing signal strength prediction at large numbers of positions within and around selected buildings. The deterministic model is based upon a ray tracing approach, where large numbers of signal paths from a hypothetical transmitter are traced and their interaction with the surrounding environment analysed.

Figure 1a: Screen shot of ray-tracing software

Figure 1b: Ray pattern from an external source

Statistics are being extracted from these detailed models to produce a more accurate, but simple empirical model.

The research has identified a number of specific buildings for which CAD data is available and which are representative of the UK. This 3D data is then fed into the deterministic model. Figure 2 shows a perspective view of an example case; a modern high-rise tower with a central reinforced concrete core surrounded by open-plan floors.

Figure 2: Model of a high-rise office block. The 5th floor and above have been removed to show the internal layout.

• Predicting coverage and interference involving the indoor-outdoor interface [pdf]
  Final Report

Back to top