1. NOAA’s National Weather Service
Results of Ambient RF Environment and Noise Floor Measurements Taken in The U.S. In 2004 and 2005
Presented By:
Robert Leck
NOAA’s National Weather Service
(QSS Group Inc.)
March 2006

2. Overview NOAA operates a wide variety of sensitive systems.
Increased noise floor levels can negatively impact the operation of these systems.
Base lining the current noise floor levels will provide a basis for future comparative measurements.
NOAA operates a wide variety of very sensitive ground based and space borne meteorological observing and remote sensing systems. Recent changes in U.S. rules as well as possible additional changes in the future generated the need for baseline data on the noise floor of frequency bands used by NOAA.
In order to quantify future deviations in current noise floor levels, (The “noise floor” in this context refers to the power spectral density of thermal noise plus the ambient signal background within a particular frequency band.), ambient noise floor testing, and characterization within frequency bands of interest to NOAA must take place prior to the widespread changes in the RF environment. Baseline data needed to be collected and archived for future comparative analysis and interference studies
This report details the results of a series of RF Ambient Environment measurements. These measurements were taken at various NOAA facilities across the United States. The measurements took place over a five-month period at a sampling of the following locations; NOAA NWS Weather Forecasting Offices, NOAA NWS Remote Weather Radar sites and NOAA NESDIS Earth Station sites. The sampling of sites was in urban, sub-urban, airport, and rural areas. The RF ambient Environment at these locations was measured, analyzed and tabulated. A summary of the results is presented here for use as a baseline

3. Ambient RF Environment Measurement System (ARFEMS) Objectives
Profile the ambient noise floor within frequency bands of interest to NOAA.
Establish an archived database.
Establish a basis for future comparative measurements.
Collect data that can be used for additional interference studies.
Principal areas of emphasis in this study include the measurement and profiling of the ambient noise floor in Urban, Sub-Urban, Rural, and Airport environments. The data will be used as a reference baseline for impact testing during the deployment of new potentially interfering radio technologies and to address future regulatory changes. The baseline data, when combined with future tests, can be used to help spectrum regulators make informed decisions regarding implementing new regulations and deployment of new radio technologies.
Specific objectives included:
·         Profile the ambient noise floor within frequency bands, which are used by NOAA systems.
·         Identify existing services within the NOAA Frequency Bands of Interest.
·         Establish an archived database.
·         Establish a basis for comparative measurements prior to, during and after the deployment of new radio technologies and U.S. and international regulatory changes.
·         Collect data to be used as the basis of additional controlled interference studies.
The resultant data will provide a baseline measurement set and will be used for future comparative studies. The data can also be used as a reference for additional interference measurements and can also be used by regulators to make informed decisions regarding the impact of increased noise floors on other system deployment initiatives within the baselined frequency ranges.

4. Frequency Bands of Interest
MHz
GHz
MHz
GHz
MHz
GHz
MHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz
GHz
6.4 GHz GHz
GHz
The measurements encompassed the profiling of the RF Ambient environment across specific meteorological and space sciences frequency bands of interest .

5. Site Locations
Site selection was based upon a number of factors. These factors include: practicality, access, environment and facilities. Testing at every NOAA site was not possible. As a result, testing was conducted at a representative set of sites in urban, suburban, and airport locations. The sites were selected from weather forecasting offices, remote radar sites, satellite earth stations and airport environments.
This graphic overlays these potential test sites on a map of the United States. (The green diamonds indicate sites at which measurements have been taken. The red diamonds indicate additional measurement sites where data may be taken in the future.)

6. Site Locations Wallops Island, VA San Diego, CA Albany, NY Phoenix, AZ
Photographs of several measurement locations.
Albany, NY
Phoenix, AZ

7. Remote Spectrum Analyzer Monitor
System Overview
LNA
Bandpass Filter
Coaxial
Switch
Temperature
Sensor
Harmonic Mixer
Direct To
Spectrum Analyzer
Above 26 GHz
Horn Antennas
Ethernet
RF Input
Frequency Range 2
Frequency Range N
Frequency Range X
Frequency Range Y
Frequency
Range N
Laptop w/Lab View
Remote Spectrum Analyzer Monitor
GPIB/Ethernet
Converter
Hub
RS-232
Variable
Azimuth Drive
Remote
Video Camera
Frequency Range 1
The measurement system was made up of a series of Signal Processing Modules (SPM’s). A high level diagram of the ARFEMS hardware can be found in Figure 9. Each SPM covers a specific frequency range and incorporates terminations for test system noise characterization testing. RF switching, gain, isolation, filtering, signal rejection capabilities and power and switch position status indicators were included in the SPM design.
The main component of the system was an advanced Spectrum Analyzer (Agilent E4440A PSA Series). The E4440A covers a frequency range of 3 MHz to 26.5 GHz. External mixers were used to facilitate measurements at frequencies above 26.5 GHz.
A laptop computer running a customized application in Lab View 7.1 controlled the test system. An additional laptop, as shown in Figure 9, was connected to a remote camera. (The remote camera provided visual feedback on the antenna rotation and polarization positioning.) The control computers were typically located at least 50 meters from the measurement system. A separate LCD monitor was connected to the Spectrum Analyzer via a VGA extender to provide a real-time display of measurement activity.
Laptop For Remote Video Camera

8. System Overview
Photographs of the assembled system and a typical Signal Processing Module (SPM).

9. Measurement Methodology
A spectrum analyzer was the primary measurement device.
Horizontal and vertical polarization profiles were captured over a 360-degree range of azimuths.
The ambient RF environment was profiled while characterizing the noise floor.
The overall measurement approach is based upon profiling the existing ambient RF environment while characterizing the noise floor[1] . A Spectrum Analyzer was used as the primary measurement device. Antennas covering the frequency ranges of interest (See Table 1) were interfaced through RF Signal Processing Modules to the Spectrum Analyzer. Horizontal and vertical polarization profiles were captured over a 360-degree range of azimuths resulting in a full characterization of the RF Ambient Environment. Given the range of frequencies over which we made our measurements it was more cost effective to purchase high gain directional antennas and vary the polarization in order to: 1) detect horizontally and vertically polarized signals within the given test bands and 2) assess their impact on the ambient noise floor.
A laptop computer controlled azimuth positioning, polarization positioning, and equipment configuration and data acquisition.
The ambient noise floor was characterized by using the Noise Marker functionality of the spectrum analyzer. This technique provided a 1 Hz resolution bandwidth measurement at a number of points across the given frequency range under investigation and did so independently of the Spectrum Analyzer real time measurement resolution bandwidth setting.
[1] In what follows, noise floor shall refer to the measured power spectral density of thermal noise plus the ambient signal background. Thermal noise is due to natural sources and the thermal agitation of electrons in the front end of the radio receiver. The thermal noise floor is given by kTO, where k is Boltzmann’s constant and TO is the temperature in degrees Kelvin, and the product kTO gives the power per unit bandwidth. If no man-made signals are present, then the measured noise floor will be approximately equal to kTO multiplied by the measurement bandwidth.

10. Data Analysis Methodology Data Reduction
The Ambient Noise Floor was derived from post processing of Spectrum Analyzer measurements.
Data was processed to compensate for system gains and losses.
Interfering signals were removed from the ambient noise floor measurement.
All measurements were referenced to a measurement resolution bandwidth of 1 Hz.
The data reduction methodology is one through which the acquired measurement data is processed. The result of this processing yields the actual RF Ambient Signal and Ambient Noise Level as seen at the receiving antenna. The resultant data will be used as a baseline against which future field measurements will be compared. The purpose, as outlined earlier, in making these comparisons will be to determine if any changes to the ambient noise floor have taken place as the result of the deployment of new communications technology or change in regulations.
In order to be able to accomplish this task, the trace and noise marker data was processed to compensate for system gains and losses and, where appropriate, converted to reflect a measurement resolution bandwidth of 1 Hz.[1]
An Excel Macro was written to operate on all the collected measurement data sets. Although some of the calculations were performed in mW (for addition and subtraction purposes), the final results were converted to and expressed in dBm.
[1] Our original intent was to characterize the noise contribution of the SPM and compensate for it in the data correction computations. This would have added an additional level of compensation. Unfortunately, this could not be implemented as the ambient noise levels were, at higher frequencies, buried in the measurement system noise and could not be mathematically extracted for data reduction or data compensation purposes.

11. Data Analysis Methodology Data Reduction
In some cases the field measurements were equivalent to the measurement systems noise floor.
In these cases the ambient noise floor was not detectable and was assumed to be below or closely equivalent to the system noise floor.
Our original intent was to characterize the noise contribution of the SPM and compensate for it in the data correction computations. This would have added an additional level of compensation. Unfortunately, this could not be implemented as the ambient noise levels were, in most cases, buried in the measurement system noise and could not be mathematically extracted for data reduction or data compensation purposes.
The noise floors in many of the upper frequency bands were below the noise floor of the measurement system [1]. As a result, noise floor levels within those bands were not easily detectable. This fact did not have a negative impact on the base lined data as the system noise floors were designed to be below the sensitivity of deployed NOAA NWS and NESDIS systems.
[1] The limiting factor from a system level noise floor perspective were the Signal Processing Modules noise floors. This fact did not have a negative impact on the collection of baseline data as the measured MDS (Minimum Detectable Signal) levels of the SPM’s were well below the levels at which NOAA NWS and NESDIS systems would be impacted negatively.

12. Measurement Results (Horizontal Polarization)
Ambient Noise Level
(Horizontal Polarization)
Urban
Suburban
Rural
Airport
136 MHz-138 MHz
-130
26.25 GHz GHz
162 MHz-174 MHz
-140
25.5 GHz GHz
400 MHz-420 MHz
-150
23.6 GHz-24.1 GHz
-160
440 MHz-460 MHZ
-170
18.5 GHz-19 GHz
-180
1.54 GHz MHz
-190
18 GHz-18.5 GHz
-200
1.670 GHz GHz
17.6GHz- 18 GHz
2 GHz-2.3 GHz
10.6 GHz-11.3 GHz
2.7 GHz-3.0 GHz
9.3 GHz-9.5 GHz
5.6 GHz GHz
8.025 GHz GHz
6.4 GHz-6.8 GHZ
7.45 GHz-7.83 GHz

13. Measurement Results (Horizontal Polarization)
Ambient Noise Level
(Horizontal Polarization)
Urban
Suburban
Rural
Airport
136 MHz-138 MHz
-130
26.25 GHz GHz
162 MHz-174 MHz
-140
25.5 GHz GHz
400 MHz-420 MHz
-150
23.6 GHz-24.1 GHz
-160
440 MHz-460 MHZ
-170
18.5 GHz-19 GHz
-180
1.54 GHz MHz
-190
18 GHz-18.5 GHz
-200
1.670 GHz GHz
17.6GHz- 18 GHz
2 GHz-2.3 GHz
10.6 GHz-11.3 GHz
2.7 GHz-3.0 GHz
9.3 GHz-9.5 GHz
5.6 GHz GHz
8.025 GHz GHz
6.4 GHz-6.8 GHZ
7.45 GHz-7.83 GHz

14. Measurement Results (Horizontal Polarization)
Ambient Noise Level
(Horizontal Polarization)
Urban
Suburban
Rural
Airport
136 MHz-138 MHz
-130
26.25 GHz GHz
162 MHz-174 MHz
-140
25.5 GHz GHz
400 MHz-420 MHz
-150
23.6 GHz-24.1 GHz
-160
440 MHz-460 MHZ
-170
18.5 GHz-19 GHz
-180
1.54 GHz MHz
-190
18 GHz-18.5 GHz
-200
1.670 GHz GHz
This graphic summarizes the results of Urban, Suburban, Rural and Airport Horizontally Polarized Ambient Noise Floor measurements.
At lower measurement frequencies (< 1.7GHz) variations on the order of 1 to 5 db were seen in the ambient noise floor measurements across Urban, Suburban, Rural and Remote environments
The ambient noise floor was higher in the frequency bands that are banded by the red lines.
At frequencies outside of the banded area, from 1.7 GHz 23.6 GHz, the ambient noise floor was below the sensitivity of the measuring equipment.
Changes on the order of 1 to 5 dB were seen across the Urban, Suburban, Rural and Airport environments with the1.36 MHz to 1.7 GHz and GHz to 27 GHz bands.
The noise floors outside of the banded areas were either equivalent to below the noise floor of the measurement system [1]. As a result, noise floor levels within those bands were not easily detectable. (This fact did not have a negative impact on the base lined data as the system noise floors were designed to be below the sensitivity of deployed NOAA NWS and NESDIS systems.)
Urban environments exhibited the highest level of ambient noise.
[1] The limiting factor from a system level noise floor perspective were the Signal Processing Modules noise floors. This fact did not have a negative impact on the collection of baseline data as the measured MDS (Minimum Detectable Signal) levels of the SPM’s were well below the levels at which NOAA NWS and NESDIS systems would be impacted negatively.
17.6GHz- 18 GHz
2 GHz-2.3 GHz
10.6 GHz-11.3 GHz
2.7 GHz-3.0 GHz
9.3 GHz-9.5 GHz
5.6 GHz GHz
8.025 GHz GHz
6.4 GHz-6.8 GHZ
7.45 GHz-7.83 GHz

15. Measurement Results (Vertical Polarization)
Ambient Noise Level
(Vertical Polarization)
Urban
Suburban
Rural
Airport
136 MHz-138 MHz
-130
26.25 GHz GHz
162 MHz-174 MHz
-140
25.5 GHz GHz
400 MHz-420 MHz
-150
23.6 GHz-24.1 GHz
-160
440 MHz-460 MHZ
-170
18.5 GHz-19 GHz
-180
1.54 GHz MHz
-190
18 GHz-18.5 GHz
-200
1.670 GHz GHz
17.6GHz- 18 GHz
2 GHz-2.3 GHz
10.6 GHz-11.3 GHz
2.7 GHz-3.0 GHz
9.3 GHz-9.5 GHz
5.6 GHz GHz
8.025 GHz GHz
6.4 GHz-6.8 GHZ
7.45 GHz-7.83 GHz

16. Measurement Results (Vertical Polarization)
Ambient Noise Level
(Vertical Polarization)
Urban
Suburban
Rural
Airport
136 MHz-138 MHz
-130
26.25 GHz GHz
162 MHz-174 MHz
-140
25.5 GHz GHz
400 MHz-420 MHz
-150
23.6 GHz-24.1 GHz
-160
440 MHz-460 MHZ
-170
18.5 GHz-19 GHz
-180
1.54 GHz MHz
-190
18 GHz-18.5 GHz
-200
1.670 GHz GHz
17.6GHz- 18 GHz
2 GHz-2.3 GHz
10.6 GHz-11.3 GHz
2.7 GHz-3.0 GHz
9.3 GHz-9.5 GHz
5.6 GHz GHz
8.025 GHz GHz
6.4 GHz-6.8 GHZ
7.45 GHz-7.83 GHz

17. Measurement Results (Vertical Polarization)
Ambient Noise Level
(Vertical Polarization)
Urban
Suburban
Rural
Airport
136 MHz-138 MHz
-130
26.25 GHz GHz
162 MHz-174 MHz
-140
25.5 GHz GHz
400 MHz-420 MHz
-150
23.6 GHz-24.1 GHz
-160
440 MHz-460 MHZ
-170
18.5 GHz-19 GHz
-180
1.54 GHz MHz
-190
18 GHz-18.5 GHz
-200
1.670 GHz GHz
This graphic summarizes the results of Urban, Suburban, Rural and Airport Vertically Polarized Ambient Noise Floor measurements. The results are similar to those that were discussed in the preceding slide.
17.6GHz- 18 GHz
2 GHz-2.3 GHz
10.6 GHz-11.3 GHz
2.7 GHz-3.0 GHz
9.3 GHz-9.5 GHz
5.6 GHz GHz
8.025 GHz GHz
6.4 GHz-6.8 GHZ
7.45 GHz-7.83 GHz

18. Conclusions
The ambient noise floor was higher in frequency bands below 400 MHz.
At measurement frequencies of less than 2GHz, variations on the order of 1 to 5 db were seen in the ambient noise floor measurements across urban, suburban, rural and airport environments.
The ambient noise floor was significantly higher in the lower frequency bands due to the presence of many systems and signals. As the frequency ranges of the measurement bands of interest increased, the ambient noise floor measurement declined and became noise limited by the measurement system’s noise floor. In other words, at frequencies above 3 GHz the ambient background noise levels were sufficiently low to not be detectable due to the inherent noise floor created by an SPM. However, as discussed in section 4.1, this does not impact the baseline results since the system MDS was sufficiently low to detect any change in ambient noise levels well before NOAA systems are impacted. The significant additional cost to gain only a slightly lower SPM noise floor was not justified.

19. Conclusions
At frequencies within the 2 GHz to 23.6 GHz bands, the ambient noise floor was below the sensitivity of the measuring equipment for horizontally polarized measurements.
At frequencies within the 2 GHz to 10.6 GHz bands, the ambient noise floor was below the sensitivity of the measuring equipment for vertically polarized measurements.

20. Conclusions
At frequencies within the 23.6 GHz to 28 GHz bands a variation of 1 to 5 dB was seen in the ambient noise floor measurements across urban, suburban, airport and remote environments for horizontally polarized measurements.

21. Conclusions
At frequencies within the 10.6 GHz to 28 GHz bands a variation of 1 to 5 dB was seen in the ambient noise floor measurements across urban, suburban, airport and remote environments for vertically polarized measurements.

22. Conclusions
Urban environments exhibited the highest level of ambient noise.
The data set provides a baseline for future comparative measurements.
Small variations in the ambient noise floor across urban, suburban, rural and remote sites were only seen at lower measurement frequencies. Urban environments exhibited the highest level of ambient noise. There was little to no change in the environmental measurements within any of the upper frequency bands. We found this result to be consistent with the ambient noise floor measurement being limited by the system noise floor.
The full data set provides an excellent baseline for future comparative measurements. The data can be used as the basis of future broadband interference and NOAA NWS and NESDIS systems protection studies. It can also be used by regulatory agencies as a reference in making decisions regarding the deployment of new services within the sections of the frequency spectrum in which measurement were made.

23. Next Steps Additional testing is planned for the 2006-2007 time frame.
Testing will take place at sites that have been tested in the past.
Comparative analysis of the and the data will be made and summarized in an upcoming report.