| Font Size: |
  |
 |
MEDIA ROOM
White Papers
Experimental Evidence Showing that ATC will not cause Harmful Interference
As part of the ATC proceeding, the Federal Communications Commission (FCC) developed an analytical model for predicting uplink interference potential to an L-band Mobile Satellite System (MSS) from a deployment of Ancillary Terrestrial Components.1 In the ATC Order the FCC used its analytical model to establish limits on the US-wide ATC deployment such that the uplink interference potential to a co-channel Inmarsat satellite will not exceed a certain limit.2 In predicting uplink interference, the Commission's model takes into account several interference mitigation mechanisms, such as outdoor blockage, power control, vocoder factor, voice activity, and polarization isolation.3 The biggest interference mitigation contributor is power control to which the model attributes 20 dB of average interference suppression.4
Given the relevance of power control in reducing uplink interference potential, SkyTerra commissioned LCC International, Inc. to conduct measurements, using existing PCS network topologies, to determine the accuracy of the Commission's model. LCC recorded the output power of a cdma2000 mobile terminal as the terminal was driven around in dense-urban, urban, and suburban environments in the Washington D.C. area, while communicating with infrastructure base stations (and being handed-off at times from one base station to another). A convertible vehicle was used in gathering the data, and the cdma2000 terminal was strapped to the side of a simulated human head which was provided for the experiment by Qualcomm, Inc. The data recorded by LCC represents the power delivered to the antenna port of the terminal by the Power Amplifier (PA) of the terminal. This quantity, as recorded, includes the effect of power control and energy absorption by the simulated human head, but does not include the effect of voice activity (a continuous voice stream was applied to the terminal).
The findings of the study are impressive: Whereas for a cdma2000 mobile terminal operating outdoors and having a maximum output power capability of -7 dBW the Commission's model predicts an average emitted power of +3 dBm (-27 dBW), for the Washington D.C. dense-urban area, we found that, on average, the average PA output was -12.22 dBm; 15.22 dB less than the Commission's model predicts! For the Washington D.C. urban area that was tested, we found that, on average, the power output of the terminal PA was -6.46 dBm; 9.46 dB less than the Commission's model predicts. For the suburban area that was tested, we found that, on average, the power output of the terminal's PA was -1.2 dBm; 4.2 dB less than the Commission's model predicts.
As stated earlier, the measurements do not include the effect of voice activity which will further reduce interference. (A continuous voice stream was applied to the terminal during data collection). The measurements do include the effect of the human body. All measurements were taken with the terminal strapped to the side of a simulated human head.5 This biases the PA out, on average, to be approximately 3 dB higher than it would be otherwise.6
Since the measurements of mobile terminal output power were taken with a PCS terminal communicating with infrastructure at PCS frequencies, and since propagation in the L-band is at least one dB better compared to propagation at PCS frequencies, an additional one dB of interference protection will, in practice, be provided by the L-band ATC. It is also noteworthy that no attempt was made to use any special-purpose vocoder. The terminal was communicating with base stations using the Enhanced Variable Rate Codec (EVRC) which is standard in cdma2000 networks. Finally, we observe that, given the very conservative predictive nature of the Commission's model, an ATC that is authorized a frequency reuse that, according to the Commission's model may generate up to 6% ΔT/T interference potential, in real life will generate substantially lower interference. The Commission should take significant comfort in that.
Summary of Measured Data Compared to FCC's Model Predictions
| Type of Environment | Measured Average Power Provided to Terminal Antenna | Ratio of Measured Average Power to Average Power Predicted by Commission's Model |
| Dense-Urban | -12.22 dBm = -42.22 dBW | -15.22 dB |
| Urban | -6.46 dBm = -36.46 dBW | -9.46 dB |
| Suburban | -1.20 dBm = -31.20 dBW | -4.20 dB |
Thus far we have presented the average power levels provided by the terminal's PA to the terminal's antenna. However these levels do not represent the power levels that would be launched toward a satellite since (a) the simulated human head absorbs at least 3 dB of the radiated power, (b) the antenna of the terminal is not perfectly matched and thus does not radiate all of the power provided to it, and (c) the antenna gain in the direction of a satellite (averaged over azimuth) is less than 0 dBi. Measurements were conducted in an anechoic chamber at Qualcomm's facilities to characterize the power that would be launched in the direction of a satellite. The antenna gain pattern of the test terminal, inclusive of the effects of the simulated human head and mismatch losses, was measured and is as depicted in the 3-D image below. Following the 3-D image, the averaged (over azimuth) antenna gain as a function of elevation angle is presented. This pattern is derived from the same anechoic chamber measurements depicted in the 3-D radiation pattern and includes the effect of head absorption and antenna mismatches. It is seen that more than 7 dB of additional interference suppression is provided by the antenna pattern for any elevation angle that may represent the location of a satellite.
3-D Plot of Composite Antenna Gain
(including effect of simulated human head plus mismatch losses)
Composite Gain Pattern With Losses
Authored by:
Peter D. Karabinis, Ph.D.
Santanu Dutta, Ph.D.
William W. Chapman, Ph.D.
1 See ATC Order Appendix C2.
2 See ATC Order Appendix C2 Table 2.1.1.C at 206.
3 See ATC Order Appendix C2 Table 2.1.1.C at 206.
4 The remaining interference mitigation mechanisms (outdoor blockage, vocoder factor, voice activity, and polarization isolation) are assigned significantly lower average values: Outdoor blockage = 3.1 dB; vocoder factor = 3.5 dB; voice activity = 1 dB; polarization isolation = 1.4 dB.
5 The simulated head, as provided by Qualcomm, was filled with materials designed to exhibit RF absorption properties similar to a human head.
6 See "Effects on Portable Antennas of the Presence of a Person," IEEE Transactions on Antennas and Propagation, Vol. 41, No. 6, June 1993.