Wireless InSite^® Indoor Example

Wireless InSite is capable of making site-specific predictions for a variety of indoor geometries. InSite's floor plan editor allows the user to create a custom indoor environment by specifying wall locations, wall heights, ceilings, floors, windows, and doorways. The material properties of each of these features can be changed to accurately reflect those in real life. Simulation of signal propagation within indoor environments is carried out using InSite's Full 3-D model which includes effects of diffraction, reflections, and transmissions through walls.

The following example utilizes InSite's indoor modeling abilities to reproduce signal strength measurements published in “A Novel and Efficient Hybrid Model of Radio Multipath-Fading Channels in Indoor Environments,” by J. H. Tarng, W. Liu, Y. Huang, and J. Huang, IEEE Trans. Antennas Propagat., March 2003. Using a stationary transmitter located in the hallway, the paper provides received power measurements for both line of site and non-line of sight locations within an engineering building at the National Chiao-Tung University in Hsin-Chu. The example illustrates InSite's ability create a floor plan, use a range of different receiver types, and to view propagation paths between a transmitter and receiver.

The layout given in Figure 1 was reproduced within Wireless InSite using the Floor Plan Editor. Wall thickness information was estimated since it was not explicitly given in the paper. The model contains gypsum board and concrete walls, glass windows, wooden doors, and metallic objects representing bookshelves, cabinets, and other features. Permittivity and conductivity values for the different materials are summarized in Table 1.

Table 1. Material properties

As detailed in the paper, a vertically polarized, half-wavelength dipole was used as the transmitting antenna. The single transmitter operates at a frequency of 2.44 GHz. The same antenna was also used at the receiving points. Both the receiving and transmitting antennas were located 1.6 m above the floor. The locations of the transmitter and receivers within the floor plan were estimated from Figure 1 since they were not explicitly given in the paper. Figure 3 shows the Wireless InSite floor plan complete with transmitter (green) and receivers (red).

All measurements presented in the Tarng et al. paper [1] were made with a radiated power of 13 dBm. Due to the calibration of the measurement system, all received power measurements are given by the quantity. where PR is the received power, P_TR(1) is the received power at 1 m from the transmitter, and all powers are in watts. Before any measurements were taken, the transmitter was calibrated in an anechoic chamber by measuring (P_T/P_R(1)). To simulate this calibration in InSite, a receiver was placed 1m away from the transmitter, with the radiating power of the transmitter set to 0 dBm, and the received power at 1m was calculated without considering any transmissions, reflections or diffractions to simulate the effects of an anechoic chamber. The received power was found to be –36 dBm, and the ratio (P_T/P_R(1)) is then found to be 36 dB. This ratio can also be determined by considering thatwhere λ is the wavelength and G_T and G_R are, respectively, the gains of the transmitting and receiving antennas. Assuming ideal half-wave dipoles with a gain of 1.64 (2.15 dBi) gives (P_T/P_R(1)) = 35.92 dB. Because Wireless InSite reports all received power predictions in dBm, it is not possible to compare the predictions directly to the measured Ω quantity. Instead of computing the actual measured received power from the measured Ω, we have chosen to consider Ω to be the received power in watts, and to make the necessary adjustments to the InSite received power predictions by using a transmitted power of 13 dBm + 36 dB = 49 dBm.

Because several of the receiver locations in the paper are collinear, receiver routes were used to simplify plotting and to capture the behavior of the received power vs. distance. Receiver points were used for locations 24 through 26 (see Figure 1) since they are more spread out and separated by walls. The receiver routes and points are identified in Figure 4 and Figure 5.

Received power predictions were calculated using Wireless InSite's Full 3-D propagation model with a maximum of 3 reflections, 4 transmissions, and 1 diffraction per ray.

Route 1 provides an example of InSite's ability to predict power for receivers within line of sight of the transmitter. InSite's results show good agreement with the five measurements taken along the hallway.

Routes 2 through 4 consist of receivers with no direct line of sight to the transmitter. Accurate power predictions require proper modeling of signal transmission through the gypsum board walls and effects of the two metallic objects that lie between the transmitter and routes 2 and 3.

Comparison of calculations and measurements for route 2 are given in Figure 7. For the first receiver location along this route, the InSite prediction is about 7 dBm lower than the measurement. Examining InSite's predictions near this region shows the model is behaving as expected. The large metal object that lies between the transmitter and the receiving point creates a shadowed region directly behind it. Although diffracted rays and reflected rays do reach this region, they are not powerful enough to increase the received power to that measured. The metal objects in the paper are described as bookshelves, cabinets, an air conditioning unit, but the paper does not provide any information as to where each type of object is located. Without a more accurate description of the nature of the metallic object and its dimensions, it is difficult to improve the prediction at this point. Away from shadowed region, however, Wireless InSite results are closer to the measured values.

Route 3 shares some characteristics with route 2. The first measurement point is partially shielded by a metallic object. Comparison between Wireless InSite's results and the two measurement points are given in Figure 8.

Route 4 lies on the other side of a gypsum board wall and in between a thick concrete wall and a metallic object. Results for route 4 are given in Figure 9. The calculated power tends to be slightly above the measurements, but generally shows good agreement with the measurements.

In addition to placing receivers along a route, Wireless InSite is also capable of placing a grid of receivers within floor plans. The grid of receivers allows the user to view received power over a large area and to quickly recognize any fast-fading effects that may be present in that region. Figure 10 shows receiver grids that were created in the vicinity of the receiving routes. The colors in the region correspond to the power received at that particular point. In Figure 10, it is easy to see the shielded region created by the metallic object that lies directly between the transmitter and route 2. It is also possible to see the locations of nulls created by interference between rays of comparable amplitude reaching the same point by different paths.

Calculated received power values for points 24 through 26 are compared to the measurements in Table 2. InSite's predictions for these points are in good agreement with the measurements. The received power in the vicinity of the points is displayed in Figure 11. Again, the receiver grid allows easy visualization of the received power near the points and the location of regions of interference.

Table 2. Comparison of calculated and measured power received for points 24, 25, and 26.

For any given receiver, the ray paths between the transmitter and receiver can be viewed to gain a better understanding of signal propagation between the points. Figure 12 shows the 10 most powerful rays that reach receiving point 25.

This example demonstrates Wireless InSite's ability to recreate a site-specific floor plan and used it to model signal propagation within an indoor environment. The results could likely be improved if more accurate information on the floor plan and transmitter/receiver locations was available. However, even without exact wall thickness values or exact transmitter/receiver locations, the results show very good agreement with measurements.

References

1. J. H. Tarng, W. Liu, Y. Huang, and J. Huang, “A Novel and Efficient Hybrid Model of Radio Multipath-Fading Channels in Indoor Environments,” IEEE Trans. Antennas Propagat., vol 51, no. 3, pp 585-594, March 2003.