ARCHIVED—Evaluation of Electromagnetic Field Intensity in the City of Toronto

Spectrum Management
Ontario Regional Engineering

Prepared by:
Ben Nguyen, P. Eng.
Sumesh Mohabeer, P. Eng.
Vicky Lai
Elisabeth Lander

June 2002

Acknowledgement

This project could not have been accomplished without the enormous support and encouragement of many people. An extra special thanks to Carl Olsen, Director of Spectrum Management for the Ontario Region, who has been a dedicated supporter from the beginning, allowing us the flexibility and time to undertake such an overwhelming study.

The data necessary for analysis and proving theories could not have been possible without the help of Dave Slingerland, Joe Doria and Watson Reed of the Ontario Region. Many thanks are owed to Dr. Arto Chubukjian and Andrew Lam from DGSE, and Brian Kasper of the DEB Lab in Ottawa for their advice and assistance in the selection of measurement equipment.

A special thank you to Dr. Martial Dufour of the CRC in Ottawa for the use of the Spectrum Explorer and the necessary software, not to mention his expert advice on its theory of operation. Also to Don Paskovich of DGSE, for organizing the equipment loan with CRC, without which none of this would have been possible. Additional thanks to all those who have encouraged and supported the team.

Note to reader: All acronyms have been defined in Appendix F.

Public Summary

Industry Canada is responsible for authorizing radiocommunication stations in Canada to ensure they conform to our policies, regulations and standards. One of the requirements is that all stations must comply with Safety Code 6 (SC 6) guidelines - "Limits of Human Exposure to Radiofrequency Electromagnetic Fields in the Frequency Range from 3kHz to 300 GHz". Within the city of Toronto, the Department performed an audit involving the measurement of radio frequency levels at 61 locations. The selected sites were representative of residential, industrial and commercial areas, parks, schools and airports. The measurements were conducted at pedestrian head level (1.75 metres above ground level).

The audit confirmed that all of the measured sites were substantially less than the recommended limits specified in the Safety Code 6 guidelines. The site with the highest radio frequency level, 16 times less than the Safety Code 6 limit, is located between Metro Hall and Roy Thompson Hall. Further analysis of the data determined that the ten sites classified as residential, are on average 7194 times less than the Safety Code 6 limit. The 16 surveyed sites classified as schools and parks are on average 5154 times less, the eight industrial area sites are on average 4464 times, the 26 sites in commercial areas are on average 324 times less, and the site near Pearson International Airport is 125,000 times less than Safety Code 6 limits.

Executive Summary

Industry Canada has always taken an interest in the area of radio-frequency (RF) electromagnetic fields because of our licensing obligation to ensure compliance with Health Canada's Safety Code 6 guidelines "Limits of Human Exposure to Radiofrequency Electromagnetic Fields in the Frequency Range from 3 kHz to 300 GHz". With the increase in the number of antenna installations across Canada, our commitment and involvement has become even more essential. The proliferation of antennas has resulted in the public becoming more aware and concerned regarding RF levels. To fulfill our responsibilities and address the public's growing concern, the Ontario regional office has conducted a study in the City of Toronto, one of the most radio-frequency congested areas in Canada, to determine the current RF ambient levels.

Field measurements were conducted at 61 sites citywide to provide the Department with current data. The selected sites were representative of residential, industrial, and commercial areas, parks, schools and airports. Signals at each site were recorded for a minimum of six minutes per frequency band to capture the variations associated with the local environment, and to comply with the Safety Code 6 guidelines. The measurements were taken 1.75 meters above the ground, at pedestrian head level, for a total of three hours. Because most of the RF energy is concentrated at the lower end of the radio spectrum and due to the frequency limitations of the equipment, measurements were conducted between 150 kHz and 3 GHz.

The study concluded that the maximum measured ambient RF level of the sites surveyed was less than 6% of Health Canada's Safety Code 6 exposure limit for persons not classed as RF and microwave exposed workers (including the general public). This level, 6%, can equally be indicated as about 16 times less than the limit. Measurements were taken at 1.75 m above ground level. The maximum level was measured at a single location within the area between Metro Hall and Roy Thompson Hall. The average of the maximums for the residential areas surveyed was 0.0139% or 7194 times less than Safety Code 6 limits, based on the ten residential sites surveyed; 0.0194% or 5154 times less, for schools and parks (16 surveyed sites); 0.0008% or 125000 times less, near Pearson International Airport (one site surveyed); 0.0224% or 4464 times less, for industrial areas (8 sites surveyed); and 0.3086% or 324 times less, for commercial areas (26 surveyed sites). For data analysis and comparison purposes, if both the Metro Hall and Spadina Parkette measurements were to be removed from the calculation, because of their proximity to the clustered broadcast facilities in the core of Toronto, the average Safety Code 6 percentage for commercial areas would drop dramatically to 0.0556% or 1798 times less than the Safety Code 6 limit. Both the average and maximum levels are well below the current limits recommended by Health Canada.

The analysis of the data also indicated that broadcasting services were the main contributors of the measured signal level at the surveyed sites and ranged from 44% to 71% of this level. Land-mobile services (traditional two-way systems, paging, trunking, and specialized mobile services) contributed 10% to 26% of the measured level. Wireless telephone services (cellular and personal communication systems) ranged from 9% to 24% of the measured level. The total accuracy of these figures is approximately ±3 dB, due to measurement method and the associated equipment, cable and antenna accuracies.

Table of Contents

• Introduction
• Measurement Activities
  • Site Selection
  • Equipment Selection
  • Measurement Method and Setup
• Data Analysis
• Results
• Conclusions
• Technical Recommendations
• Appendix A - Site Locations and SC6 Values
• Appendix B - Example of Site Information and Maps
• Appendix C - Antenna and Cable Calibration Charts
• Appendix D - SurveyEquipment
• Appendix E - Photographs of antennas
• Appendix F - List of Acronyms
• Appendix G - Sites Suggested by City of Toronto officials
• Appendix H - ContactList
• Appendix I - References

---

Evaluation of Electromagnetic Field Intensity in the City of Toronto
(PDF, 1.23 MB, 36 pages) – PDF Readers

---

Introduction

Over the past decade, the Canadian wireless industry has experienced substantial growth, resulting in increases in the number of antenna installations located on rooftops and towers. Spectrum congestion, rising sensitivity toward existing and proposed antenna towers, and the public's health concerns regarding radio frequency (RF) levels are becoming more apparent across Canada. The Golden Horseshoe Area (GHA)^{Footnote 1} of Ontario, but more specifically the City of Toronto, has one of the most congested radio environments in Canada.

Industry Canada, namely the Ontario Region of Spectrum Management, in its mandate to improve the way it manages its governance of the radio spectrum, has undertaken the task of determining the current RF ambient levels throughout the City of Toronto, through a series of measurements and data analysis. The intent of this study was to audit numerous locations throughout the city to verify that Safety Code 6 (SC 6) exposure limits for persons not classed as RF and microwave workers (including the general public), were met for publically accessible ground level areas. A second objective was to identify whether any one specific category licensee (such as broadcast, cellular, personal communication systems (PCS), local multipoint communication systems (LMCS), paging, aeronautical and land mobile) was a primary contributor to the RF level within the 150 kHz to 3 GHz radio spectrum. The electromagnetic field intensity in the industrial, scientific and medical (ISM) frequency band was also examined to assist manufacturers in designing radio equipment to operate in radio congested environments. In addition, the Department was interested in the relationship between these RF levels and Health Canada's recommended exposure limits^{Footnote 2}.

With this knowledge, departmental officials will be able to more efficiently authorize stations, as well as address an ever growing concern from the public regarding RF energy in their neighbourhoods. In addition, the data gathered will allow the Department to review its application processing procedures with respect to SC 6 levels associated with simple applications, as well as formally respond to City of Toronto officials regarding the issue of field intensity levels at various locations in the city. The Department has agreed to share the results of the survey with the City, the telecommunication's industry, and any other interested party.

The field measurements were conducted at 61 locations within the City of Toronto, 12 of which, listed in Appendix G, were suggested by City of Toronto officials. The locations are representative of residential, industrial, and commercial areas, as well as schools, parks, and airports. The measurement process involved the use of digital analyzers (Spectrum Explorer and Vector Signal Analyzer) to perform band-power measurements on signals received within the 150 kHz to 3 GHz portion of the radio spectrum. The data collected was analyzed, normalized and compared to the limits established by Health Canada to obtain a total Safety Code 6 figure for each location.

Prior to performing the measurements, preliminary surveys of the selected locations were conducted with a Narda meter and associated probes, to verify that the recommended SC 6 levels were not exceeded. Due to the probes' limited sensitivity and inaccuracy in registering valid readings for low level signals, additional equipment was used for precise measurements.

The report consists of three main sections pertaining to each step of the study, from measurement activities, data analysis, to conclusions. The measurement activities portion focuses on the site selection criteria, the theory behind the equipment's operation, measurement procedure and methods, data collection, and equipment accuracy. The section dealing with data analysis discusses the various methods of recording the data, and the formulas used in the data analysis to calculate the SC 6 levels at each site, the different methods of determining a maximum versus likely scenario, and a summary of the contributions of the various radio-communication services. The conclusion and technical recommendations section discusses the RF levels at the sites measured in relation to the SC 6 guidelines, proper measurement techniques and comments on the strengths and limitations of the equipment. Attached at the end of the report are several appendices listing the maximum and average recorded levels of each measured site, and calibration curves. Any parties interested in the raw data or the complete program for analysis, should contact the authors. A contact list has been provided in Appendix H.

Measurement Activities

Site Selection

Throughout the City of Toronto, 61 locations were surveyed. The selection process was based on several criteria, firstly sites representing various geographical areas namely residential, industrial, and commercial areas, schools, parks, and airports, and secondly the inclusion of several sites selected by City of Toronto officials, as listed in Appendix G. Roof top locations were not included as part of this study due to equipment transportation difficulties and the complexity of various rooftop antenna configurations.

A preliminary survey was conducted to establish convenient and acceptable measuring locations ensuring minimal reflection and diffraction possibilities at each site. The Narda meter and probe were used as a preliminary check to verify that the locations did not exceed SC 6 levels. However, given the relatively low RF levels at each of the sites, and the sensitivity of the Narda probes, the meter could not accurately register the site's RF energy. During this survey, the exact geographical coordinates were identified with a GPS. Basic details of the site parameters were noted; visible antenna installations, building and tree blockages, and line of sight to the CN Tower at each site. Photographs, maps and observations for one of the surveyed sites is available in Appendix B.

Equipment Selection

The selection of digital analyzers was based on a prior survey conducted by a team of departmental experts from the DEB Lab, DGSE, regional and Toronto district offices, and Acton RSSC, that reviewed available equipment, instrumentation technologies and measurement procedures. The team conducted a small experiment whereby FM and TV broadcast emissions from the CN Tower and First Canadian Place were measured. To compare the different measuring methods, three devices were employed: a Spectrum Explorer using digital signal processor (DSP) technology, developed by the CRC using off-the-shelf hardware from Agilent Technologies and National Instruments, an HP8594E Spectrum Analyzer (analog), and a Narda 8718 Meter with an 8760N E-Field Probe. The team concluded that the digital analyzer was more aptly suited for determining field intensity levels in the survey environment and as a result, it was selected as the primary measuring device.

Historically, to determine RF levels, the carrier power was measured using a swept-tune spectrum analyzer. However, with the advancement of technology, analyzers are now capable of measuring the band-power of a signal within a specified bandwidth. This process involves a newly developed technique of band-power measurement, whereby the area under the signal signature is measured. The received signal is digitized using an analog-to-digital converter (ADC), then analyzed by a digital signal processor (DSP). For the purposes of this study, the HP 89441A Vector Signal Analyzer and Spectrum Explorer, both digital analyzers, were selected. The instruments perform a Fast Fourier Transform (FFT) calculation, essentially measuring the output power of hundreds of narrowband filters with center frequencies distributed across the span of the specified bandwidth and then summing them together.

Ideally, one omnidirectional antenna covering the entire 150 kHz to 3 GHz radio spectrum range would have been preferred, however due to a lack of commercial antenna availability and time constraints, this was not possible. Therefore, the spectrum was divided into four frequency ranges and antennas were selected to cover each range. The omnidirectional Singer 92198-3 antenna was used to receive signals in the 150 kHz to 30 MHz range, the home-built omnidirectional whip antenna for 30 MHz to 80 MHz, the omnidirectional Rhode and Schwarz HK014 for 80 MHz to1300 MHz and the directional Tecom 201031 log-periodic antenna for the final portion, 1300 MHz to 3000 MHz. Photographs of the Tecom 201031 log-periodic and the Rhode and Schwarz HK014 antennas are listed in Appendix E.

The custom-built whip, Rhode and Schwarz and Tecom antennas were all calibrated internally at the departmental engineering laboratory in Ottawa, while the antenna manufacture calibration chart was used for the Singer antenna. The calibrated curves for each antenna are shown in Appendix C.

Measurement Method and Setup

All field measurements were conducted 1.75 meters above the ground, as recommended in the Safety Code 6 guidelines. Both the whip and Singer antennas were positioned so their feed points were 1.75 meters above the ground, the Rhode and Schwarz HK014 had its phase center 1.75 meters above the ground, and the Tecom 201031 log-periodic was positioned so its feed point and center were 1.75 meters above the ground. Precautions were taken to position the antennas in a manner that would minimize reflections, diffractions, and interactions with people and vehicles, as well as to ensure that within the spatial averaging area, as mentioned in the Safety Code 6 guidelines, the variation of the received signal was less than 0.5 dB.

Prior to the measurements, frequency band plans, assignable and assigned channels (with their respective bandwidths), within the 150 kHz to 3 GHz frequency range, were entered into the Spectrum Explorer software program, as scanning lists. These lists contained all the channels to be measured in this study.

The Spectrum Explorer settings were optimized for speed and accuracy, and are shown in Figure No. 02.

At the outset of each measurement, both the Vector Signal Analyzer and Spectrum Explorer were monitored for overload. If overload was observed on the Vector Analyzer, attenuation was added. If overload was observed on the Spectrum Explorer, adjustments were made by inserting additional attenuation either externally, internally or both. By adding attenuation, the sensitivity of the analyzer is reduced, consequently eliminating low level signals. Because these low level signals are no longer recorded, the measurements will report band-power results that are a little lower than those actually present. However, the eliminated signals are typically 60 dB below, or one million times less than, the measured signal level. Consequently, these adjustments reduced the total energy entering the front end of the analyzer.

The signals at each location were recorded for a minimum of six minutes per frequency band to capture the variations due to the local environment, and to comply with the Safety Code 6 guidelines. All antennas, except for the Singer 92198-3 and the custom-built whip, were rotated 360° to obtain the maximum signal level. The sites were normally monitored for three hours resulting in the collection of approximately one gigabyte of data.

The Vector Spectrum Analyzer logged the signal strength of those received frequencies in the 150 kHz to 30 MHz range. To ensure that the receiver unit was stable before the digitization process took place, a settling time of five milliseconds was selected. The Spectrum Explorer measured those signals within the 30 MHz to 3 GHz frequency spectrum. The Explorer took several milliseconds to complete a sub-band, since it is only capable of digitizing 4.51 MHz of spectrum at once. The sweep time is also dependant on the number of bins (equivalent to the resolution bandwidth of an analog analyzer) selected for each channel. Both analyzers were software controlled by a personal computer that dated and time stamped the logged data.

The Vector Analyzer's frequency span was sub-divided into eight frequency ranges, accommodating the eight bands of the omnidirectional Singer 92198-3 antenna. The list of frequency sub-bands for the Vector Analyzer is as listed in Table No. 01.

Table No. 01 - List of Frequency Sub-Bands for the Vector Analyzer

Frequency Sub-Band	Major Types of Services	Antenna used
150 - 300 kHz	Maritime & Aeronautical	Singer 92198-3
300 - 590 kHz	Aeronautical & AM Radio	Singer 92198-3
590 - 1100 kHz	AM Radio	Singer 92198-3
1100 - 2100 kHz	AM Radio & Mixed	Singer 92198-3
2100 - 4100 kHz	Mixed	Singer 92198-3
4100 - 8000 kHz	Mixed	Singer 92198-3
8000 - 15,000 kHz	Mixed	Singer 92198-3
15,000 - 30,000 kHz	Mixed	Singer 92198-3

To simplify the settings for the Spectrum Explorer, the monitored frequency band (30 MHz to 3 GHz) was divided into smaller sub-bands to accommodate the various receive antennas, types of services and expected receiving signal levels and is listed in Table No. 02.

Table No. 02 - List of Frequency Sub-Bands for the Spectrum Explorer

Frequency Sub-Band	Major Types of Services	Antenna used
30 - 80 MHz	Land Mobile & Television	Whip
80 - 88 MHz	Television Broadcast	HK 014
88 - 108 MHz	FM Radio	HK 014
108 - 136 MHz	Aeronautical	HK 014
136 - 174 MHz	Land Mobile	HK 014
174 - 406 MHz	Television Broadcast	HK 014
406 - 470 MHz	Land Mobile	HK 014
470 - 806 MHz	Television Broadcast	HK 014
806 - 849 MHz	Mobile	HK 014
849 - 902 MHz	Cellular, Trunking, Paging	HK 014
902 - 960 MHz	Trunking, Paging, Fixed links	HK 014
960 - 1300 MHz	Aeronautical	HK 014
1300 - 3000 MHz	PCS, Digital Radio Broadcast, Wireless Cable TV	Tecom 201031

Each of frequency sub-band was further divided into frequency channels with fixed bandwidths of 10 kHz, 50 kHz, 100 kHz, etc. A sample of the frequency channels is listed in Table No. 03.

Table No. 03 - Samples of Monitored Channels

Channel Number	Center Frequency (MHz)	Channel Bandwidth (Hz)	Filter Resolution (Hz)	Minimum Number of Filters
1	88.005	10000	312.5	9
2	88.015	10000	312.5	9
3	88.025	10000	312.5	9
---	---	---	---	---
299	90.985	10000	312.5	9
300	90.995	10000	312.5	9
301	91.005	10000	312.5	9

The maximum received signals of the frequency channels from the monitored frequency bands for several services are shown in Figure No. 03 below.

For the purpose of this analysis, the maximum received signal level recorded within the monitoring period, for each frequency channel, was identified. These levels were converted to an equivalent electric field intensity and normalized with respect to the appropriate Safety Code 6 limit. For each site, the squared normalized electric field intensity values of all the frequency channels were summed to obtain an overall SC 6 figure.

It was assumed that all active frequencies were transmitting simultaneously, which in probability is extremely low. This results in an over-estimation of the actual RF energy present in the surveyed area. However, it allows the Department to reasonably assure the public that at the 61 locations surveyed, the RF levels are below the calculated figures, at the time of measurement.

Several assumptions were made throughout the course of the measurements and the project itself. One fundamental premise was that the length of time necessary to complete one pass of the frequency band was equal to, or shorter than, the transmitting time of the signals. This meant that all transmitted signals were logged by the system, and none were missed due to short duty cycles. Another crucial assumption was that at the time of measurement, all transmitters were fully operational and at usual power. Lastly, if the addition of attenuation was necessary to counteract overload, the lower level signals that were no longer recorded would not significantly impact the overall results. This is valid, because the eliminated signals are typically 60 dB below, or one million times less than, the measured signal level.

Calibration of the antenna gain, antenna factor and cable losses were all conducted in the DEB Lab in Ottawa. The total measurement and equipment accuracy for this study is ± 3dB. This figure includes a receiver accuracy of ±1.0dB, an antenna gain variation of ±0.5dB, cable and connectors uncertainties of ±0.2 dB, data truncation of ±0.1dB and other factors relating to measurement procedures contributed an additional ±1.0 dB. The uncertainty due to equipment of ±2.8 dB is arrived at by adding these individual uncertainties results. This is a conservative figure.