Digital Audio Broadcasting

Supplementary Analog SCA Compatibility Tests

Test Plan and Procedures

Document No. 02-15

April 2002

Advanced Television Technology Center 1330 Braddock, Suite 200 Alexandria, VA 22314-1650 (703) 739-3850 (703) 739-3230 (Fax) www.attc.org

Table of Contents

1 INTRODUCTION 1

1.1 BACKGROUND 1 1.2 DOCUMENT SCOPE 1 1.3 DOCUMENT STRUCTURE 1 1.4 RELATED DOCUMENTS 1

2 TEST OVERVIEW 3

2.1 OBJECTIVES 3 2.2 METHODOLOGIES 3 2.3 TEST CONDITIONS 3 2.4 EVALUATION METHODS 3

3 SIGNAL DESCRIPTIONS 4

3.1 RF SIGNALS 4 3.1.1 DESIRED ANALOG – TYPE I 4 3.1.2 DESIRED ANALOG – TYPE II 4 3.1.3 DESIRED ANALOG – TYPE III 5 3.1.4 DESIRED HYBRID – TYPE I 5 3.1.5 DESIRED HYBRID – TYPE II 5 3.1.6 DESIRED HYBRID – TYPE III 6 3.1.7 UNDESIRED ANALOG 6 3.1.8 UNDESIRED HYBRID 6 3.1.9 ADDITIVE WHITE GAUSSIAN NOISE 7 3.2 STEREO CLIPPED USASI NOISE 7 3.2.1 INTRODUCTION 7 3.2.2 USASI NOISE SOURCE CHARACTERIZATION 7 3.2.3 RECORDING PROCESS AND CHARACTERIZATION 10 3.2.4 SURVEY OF “REAL-WORLD” BROADCAST STATIONS 11 3.2.5 MODULATED SPECTRUM CHARACTERIZATION 14

4 STANDARD METHODOLOGIES 17

4.1 FM B AND AND RF M EASUREMENTS 17 4.1.1 FM ANALOG POWER 17 4.1.2 IBOC HYBRID MODE POWER 18 4.1.3 POWER IN THE PRESENCE OF MULTIPATH 19 4.1.4 RATIO OF ANALOG TO DIGITAL POWER (IBOC HYBRID MODE) 20 4.1.5 ADDITIVE WHITE GAUSSIAN NOISE POWER 21 4.1.6 DEVIATION AND CHANNEL CONFIGURATION 24 4.2 BASEBAND AUDIO MEASUREMENTS 26 4.2.1 SIGNAL-TO-NOISE RATIO (SNR) 26

5 TEST PROCEDURES 28

5.1 SCA RECEIVER PERFORMANCE IN THE PRESENCE OF ANALOG AND IBOC ADJACENT CHANNELS 28 5.2 SCA RECEIVER PERFORMANCE IN THE PRESENCE OF AN ANALOG OR IBOC HOST 30 5.3 SCA RECEIVER PERFORMANCE IN THE PRESENCE OF AN ANALOG OR IBOC HOST IN MULTIPATH CONDITIONS 32

6 - APPENDIX A - LIST OF RECEIVERS UNDER TEST 34

7 - APPENDIX B - MULTIPATH PROFILES 35

7.1 CHARACTERIZATION OF STATIC MULTIPATH PROFILES 35 7.2 MULTIPATH PROFILES - STATICMP1 40 7.3 MULTIPATH PROFILES – STATICMP2 43 7.4 MULTIPATH PROFILES – STATICMP3 45

1 Introduction

1.1 Background The Advanced Television Technology Center (ATTC) has conducted an extensive DAB test program over the course of the past two years. This program has primarily focused on fulfilling the laboratory test requirements of the National Radio System Committee’s (NRSC) DAB Subcommittee in evaluating iBiquity Digital’s IBOC system. As such, many of the test program objectives, methodologies, and test conditions have been well defined by the various NRSC Working Groups.

However, other parties have recently expressed an interest in conducting IBOC DAB tests that are outside the scope of the NRSC process, in order to examine certain aspects of IBOC compatibility in more detail than originally required by the NRSC. In particular, National Public Radio (NPR) and the International Association of Audio Information Services (IAAIS) are interested in further evaluating the compatibility of FM IBOC with existing analog SCA services (e.g. “Reading For The Blind” SCA services).

This test plan is a result of a collaborative effort between iBiquity, NPR, IAAIS, and ATTC to develop test procedures that are appropriate for examining IBOC compatibility with 67kHz and 92kHz analog audio SCA services. In addition to greatly expanding the test conditions of the original NRSC tests, the field of SCA receivers under test has been expanded to encompass a wider variety of models and manufacturers. Multipath test conditions have also been added to the program to gauge the effect of multipath on IBOC compatibility with host SCA services. 1.2 Document Scope This document contains the laboratory plan and procedures to be utilized for the Supplementary Analog SCA Compatibility Tests of the iBiquity Digital In-Band On- Channel (IBOC) Digital Audio Broadcasting (DAB) System. 1.3 Document Structure This document is written in such a way that signal types and measurement methodologies used repeatedly throughout testing are defined once. Prior to defining the test procedures there are sections dedicated to the defining signals and measurement methodologies (Sections 3 and 4 respectively). When the test procedures call for a specific signal or measurement, the reader should refer to the appropriate section for the complete definition. 1.4 Related Documents For a detailed description of the physical test platform that will be used during supplementary analog SCA compatibility testing refer to the following ATTC documents:

ATTC Document No. 01-20, Digital Audio Broadcasting, Test Bed Proof of Performance Plan, Revision 2.0, November 2001 ATTC Document No. 01-01, Digital Audio Broadcasting, Test Bed Proof of Performance Record, Revision 2.0, November 2001

ATTC Document No. 01-16, Digital Audio Broadcasting, Test Bed Daily Calibration Procedure, Revision 4.0, October 2001 Digital Audio Broadcasting, Test Bed Daily Calibration Record of Test Results”(open ATTC Document)

The previous (NRSC specified) analog SCA compatibility test procedures and results are detailed in the following ATTC documents:

ATTC Document No. 01-03, Digital Audio Broadcasting, IBOC Laboratory Test Procedures – FM Band, Revision 4.2, August 2001 ATTC Document No. 01-16B, Digital Audio Broadcasting, SCA Compatibility of the IBOC System in the FM Band, Summary of Test Results, October 2001

2 Test Overview

2.1 Objectives The test procedures described in this document have one main objective, to quantify the impact of IBOC on existing SCA services. This is known as Compatibility testing. In compatibility testing various adjacent channel and multipath conditions are simulated in the laboratory. For each channel condition, tests will be conducted to determine the impact of IBOC, if any, on SCA services. 2.2 Methodologies To meet the objective described in the previous section, a series of controlled laboratory tests will be conducted. Each of these tests will include some channel impairment that is commonly observed in the FM broadcast band. These interference scenarios will be created in the laboratory, and the performance of certain commercially available SCA receivers 1 will be evaluated. The receivers will be tested under these conditions in the presence and absence of an IBOC signal on either the “host” or adjacent channel station. 2.3 Test Conditions Below is a list of the test conditions to be utilized during the testing process:

1. First adjacent channel interference 2. Second adjacent channel interference 3. “Host” channel interference 4. “Host” channel interference in the presence of multipath

NPR, the IAAIS, iBiquity Digital, and the ATTC developed the test scenarios outlined in this document. 2.4 Evaluation Methods The performance of the SCA receivers will be determined objectively. Objective evaluation of an SCA receiver requires the use of standard audio test equipment to measure and objectively quantify audio quality in an accurate and repeatable manner. Audio quality will be measured in terms of the audio signal-to-noise ratio (SNR).

1 NPR and the IAAIS have chosen the SCA receivers to be tested. For a list of these receivers refer to section 6 of this document.

3 Signal Descriptions The purpose of this section is to define all of the signals (RF and audio) to be used throughout testing. Later sections of this test plan may refer to a “Desired Hybrid Type I”, for example, the definition of which can be found in this section. 3.1 RF Signals

3.1.1 Desired Analog – Type I A desired analog Type I signal is designed for use in tests of analog host compatibility with multipath. This type of signal shall be used in the tests of both 67kHz and 92kHz subcarriers. The injection of both subcarriers for this type of signal is 10% regardless of which SCA is being tested. However, the audio only modulates the SCA being evaluated. When the 67kHz SCA is being tested, for example, there will be no audio present on the 92kHz SCA (resulting in 0kHz Deviation). This signal type will yield a total channel modulation of 110% when configured as shown in Table 3-1.

Table 3-1 - Desired Analog – Type I Signal Characteristics

Main Channel Main Channel Power 67kHz Subcarrier 92kHz Subcarrier Audio 1) 97.9MHz Clipped USASI Moderate 1) 10% injection 1) 10% injection with peaks equal (-62dBm) 2) Stereo Transmission 80% modulation 2) 150ms pre- 2) 150ms pre- (60kHz Peak emphasis emphasis 3) 10% Pilot Injection Deviation) 3) 6kHz Peak 3) 7kHz Peak 4) 75ms Pre-emphasis Deviation with a Deviation with a 400Hz Tone – 0kHz 400Hz Tone – 5) Total Modulation = with Silence 0kHz with Silence 110% 4) Test Dependent 4) Test Dependent Audio – 400Hz Audio – 400Hz Tone or Silence Tone or Silence 3.1.2 Desired Analog – Type II A desired analog Type II signal is designed for use in first adjacent interference tests, second adjacent interference tests, and analog host compatibility with no multipath. When using this type of signal, keep in mind that the injection of the SCA being evaluated is to be 10% while the other SCA should be turned off (0% injection). This configuration will yield a total channel modulation of 105% when configured as shown in Table 3-2.

Table 3-2 - Desired Analog – Type II Signal Characteristics

Main Channel Main Channel Power 67kHz Subcarrier 92kHz Subcarrier Audio 1) 97.9MHz Clipped USASI Moderate 1) 10% injection or Off 1) 10% injection or Off with peaks equal (-62dBm) 2) Stereo Transmission 85% modulation 2) 150ms pre- 2) 150ms pre- (63.75kHz Peak emphasis emphasis 3) 10% Pilot Injection Deviation)

3) 7kHz Peak 4) 75ms Pre-emphasis 3) 6kHz Peak Deviation with a Deviation with a 400Hz Tone 5) Total Modulation = 400Hz Tone 105% 4) Test Dependent 4) Test Dependent Audio – 400Hz Audio – 400Hz Tone or Silence Tone or Silence 3.1.3 Desired Analog – Type III The Desired Analog Type III signal is the same as a Desired Analog Type I signal, except that a Type III signal calls for no modulation on the main (host) carrier. This signal shall also be used in tests of analog host compatibility with multipath. The injection of both subcarriers for this type of signal is 10% regardless of which SCA is being tested. However, the audio only modulates the SCA being evaluated. When the 67kHz SCA is being tested, for example, there will be no audio present on the 92kHz SCA (resulting in 0kHz Deviation). This type of signal yields a total channel modulation of 30%. A Type III signal will have the characteristics listed in Table 3-3.

Table 3-3 - Desired Analog – Type III Signal Characteristics

Main Channel Main Channel Power 67kHz Subcarrier 92kHz Subcarrier Audio 1) 97.9MHz None Moderate 1) 10% injection 1) 10% injection (-62dBm) 2) Stereo Transmission 2) 150ms pre- 2) 150ms pre- emphasis emphasis 3) 10% Pilot Injection 3) 6kHz Peak 3) 7kHz Peak 4) 75ms Pre-emphasis Deviation with a Deviation with a 400Hz Tone – 0kHz 400Hz Tone – 0kHz 5) Total Modulation = with Silence with Silence 30% 4) Test Dependent 4) Test Dependent Audio – 400Hz Audio – 400Hz Tone or Silence Tone or Silence

3.1.4 Desired Hybrid – Type I This signal shall be defined as the spectral sum of an analog desired signal Type I (as defined in section 3.1.1) and the digital carriers as generated by an iBiquity Digital IBOC exciter in hybrid mode. The digital carriers utilize OFDM modulation. The power sum of all digital carriers in the hybrid signal has an average power that is 20dB less than the average power in the analog carrier. 3.1.5 Desired Hybrid – Type II A desired hybrid Type II signal shall be defined as the spectral sum of an analog desired signal Type II (section 3.1.2) and the digital carriers as generated by an iBiquity Digital IBOC exciter in hybrid mode. The digital carriers utilize OFDM modulation. The power sum of all digital carriers in the hybrid signal has an average power that is 20dB less than the average power in the analog carrier.

3.1.6 Desired Hybrid – Type III A Type III desired hybrid signal is defined as the spectral sum of an analog desired signal Type III (section 3.1.3) and the digital carriers as generated by an iBiquity Digital IBOC exciter in hybrid mode. The digital carriers utilize OFDM modulation. The power sum of all digital carriers in the hybrid signal has an average power that is 20dB less than the average power in the analog carrier. 3.1.7 Undesired Analog An undesired analog interferer signal is designed for adjacent channel interference tests that employ objective evaluation. Such a signal shall have the following characteristics:

Table 3-4 - Undesired Analog Signal Characteristics

Main Channel Main Channel Power 67kHz Subcarrier 92kHz Subcarrier Audio 1) Frequency: Clipped USASI Variable Off Off a) Upper First: Noise with peaks 98.1MHz equal 90% b) Lower First: modulation 97.7MHz (67.5kHz Peak c) Upper Second: Deviation) 98.3MHz d) Lower second: 97.5MHz 2) Stereo Transmission 3) 10% Pilot Injection 4) 75ms Pre-emphasis 5) Total Modulation = 100% 3.1.8 Undesired Hybrid An undesired hybrid interferer is designed to be used in adjacent channel interference tests that employ objective evaluation.

This signal shall be defined as the spectral sum of an analog undesired signal and the digital carriers as generated by an iBiquity IBOC exciter in hybrid mode. The digital carriers utilize OFDM modulation; further details are proprietary to iBiquity.

The analog portion of the signal shall be defined as in section 3.1.7.

The sum of all digital carriers in the hybrid signal shall have an average power that is 20dB below the average analog power. The measurement method for setting and verifying this power ratio is outlined in section 4.1.4.

The digital carriers may or may not be modulated by audio. Due to the randomization process of the audio, it is not expected that modulation of the digital carriers will affect test results in any manner.

3.1.9 Additive White Gaussian Noise Where AWGN noise is specified, a broadband noise signal shall be added to the spectrum at the specified power level. This noise shall have flat spectral characteristics between 88MHz and 108MHz. Beyond these frequency limits, the noise shall be sharply attenuated with a bandpass filter. The peak amplitude excursions of the noise signal shall have a Gaussian probability distribution.

Section 4.1.5 describes the procedure used to determine and set the power level of this additive white Gaussian noise. 3.2 Stereo Clipped USASI Noise

3.2.1 Introduction In order to simulate real world conditions in the laboratory, the FM signals (undesired as well as desired) must be modulated with a signal that simulates typical broadcasting program material. This need for a standard audio signal lead to the generation of a Clipped USASI Interferer.

USASI noise, developed by the United State of America Standards Institute, is designed to simulate the spectra of typical audio programming over a long period of time. USASI noise is created by passing the output of a white noise source through a network of filters. This network consists of a 100Hz high pass filter and a 320Hz low pass filter, both with a 6dB per octave roll off.

For these tests, a simulation of the frequency spectrum of a typical FM rock station is required. Rock stations in the FM band often broadcast highly processed audio. This high degree of processing can result in clipping of the audio signal. To accurately simulate these conditions for laboratory tests, the USASI noise interferer also needs to be clipped.

Also, for these tests to be repeatable, it must be possible to recreate the modulating audio signal. To ensure that the exact audio signal could be recreated, a recording of this signal was made.

A number of steps were involved in the creation of this interferer. Initially, the output of the noise generator had to be verified as USASI noise. Next, a recording of the noise was made. Verification that the recording process did not have a significant impact on the noise signal was performed. Finally the signal was clipped, using Cool Edit Pro software, to simulate processed rock. These steps are detailed below. 3.2.2 USASI Noise Source Characterization The USASI noise signal to be used for tests was generated by a Delta Electronics SNG-1 Stereo Noise Generator. The Delta noise generator noise mode called “NRSC” was used because in this mode, the unit generates stereo USASI noise. To verify that the noise generator was truly generating USASI noise a series of characterization tests were conducted. The Delta noise generator was configured as shown in Table 3-5 and the output was connected either to a Tektronix Oscilloscope or an Audio Precision AP-1. The results of this characterization are illustrated in the following sections.

Table 3-5 – Delta SNG - 1 Setup Parameter Description Noise Select USASI Noise Mode CW Output Select NRSC Output Level +4dBu RMS

Probability Distribution of Noise Amplitudes The plots included in this section illustrate the statistical data and average values of the voltage levels as seen at the output of the Delta noise generator. One of the specifications of USASI noise is that the voltage level distribution over time is Gaussian. The figure below illustrates a perfectly Gaussian probability distribution.

68%

95%

99%

m - 3s m - 2s m - 1s m m + 1s m + 2s m + 3s Figure 3-1 - Gaussian Probability Distribution By comparing the ideal curve and values of Figure 3-1 to the values on the right and the distribution curve on the left side of Figure 3-2 it can be seen that the distribution of voltage levels at the output of the Delta noise generator is very nearly Gaussian.

Percentage of Total Power within the Range Indicated

Probability Distribution

Maximum and Minimum Voltage Levels

Figure 3-2 – Statistical plot of the Delta Electronics Noise Generator output

Frequency Response The plot below is an averaged fast Fourier transform of the output of the Delta noise generator. This plot shows that the Delta noise generator output does roll off as dictated by the definition of USASI noise.

-5 -10 -15 -20 d B -25 u -30 -35 -40 -45 0 2.5k 5k 7.5k 10k 12.5k 15k 17.5k 20k Hz Figure 3-3 – Averaged FFT of Delta noise generator output

3.2.3 Recording Process and Characterization As stated in section 3.2.1, there is a need to record the USASI noise signal in order to ensure the repeatability of these tests. However, this recording process must not significantly affect the noise signal. A digital recording of the Delta noise generator was produced using the setup shown in Figure 3-4.

L AES Delta SNG-1 NVision Audio Digital Audio NV 1035 Work Station R A to D

Figure 3-4 – Device setup used to make digital recording of Delta noise generator

The Delta was configured as in Table 3-5. The A-to-D converter was used to convert the Delta outputs (left and right analog) to an AES audio signal. This was then recorded at a 44.1kHz sampling rate, with 16 bits of resolution. The following subsections detail the characterization of this recording.

Probability Distribution of Noise Amplitudes The plots included in this section illustrate the statistical data and average values of the voltage levels seen when the recording of the output of the delta noise generator is played back. Percentage of Total Power within the Range Indicated

Probability Distribution

Maximum and Minimum Voltage Levels

Figure 3-5 - Statistical plot of the recording of the Delta Electronics Noise Generator output

Frequency Response Figure 3-6 shows FFT’s of the Delta Noise Generator output and the recording of the output overlaid. -5 -10 -15 -20 d B -25 u -30 -35 -40 -45 0 2.5k 5k 7.5k 10k 12.5k 15k 17.5k 20k Hz Figure 3-6 - FFT of outputs of Delta noise generator overlaid with FFT of Left and Right channels of the USASI noise recording Clipping The next step in the development of the USASI interferer was to clip the signal. The signal was clipped by 3dB using Cool Edit Pro software. To do this the following steps were performed:

1. Select the entire waveform. 2. Under the Transform menu select Amplify. 3. In the Normalization box enter 0dB in the Peak Level field and press Calculated Now. This calculates the amplification required to normalize the signal to 0dBFS and places the value in the Amplification field. 4. Press the OK button. This will normalize the peaks to 0dBFS. 5. Under the Transform men select Hard Limiting. 6. Set the Max Amplitude field to 0dB, the Boost field to 3dB, the Look Ahead Time to 0, and the Release time to 100. 7. Press the OK button.

Bit Error Rate Validation For information regarding the Bit Error Rate validation please see the Proof of Performance Test Plan and sections 3.7 and 3.8 of the Proof of Performance Results as referenced in section 1.4 of this document. 3.2.4 Survey of “Real-World” Broadcast Stations To verify that the interferer developed actually did simulate real world conditions; a survey of some local (Washington D.C. metro area) radio stations was performed. Examination of

Off-Air radio stations operating in the FM broadcast band revealed a wide variety of stations utilizing many different degrees of audio processing.

Examples of some of the different types of radio stations are shown in the plots of this section. The first two plots below (Figure 3-7 and Figure 3-8) show the RF spectrum and demodulated baseband of a local station operating at 107.3MHz; this is an example of a station that is broadcasting heavily processed program material. The next two plots (Figure 3-9 and Figure 3-10) are of a public station operating at 90.9MHz. This is an example of a station broadcasting lightly processed material. The last two plots (Figure 3-11 and Figure 3-12) of this section are of a good example of a station broadcasting moderately processed program material, a local rock station operating at 101.1MHz.

Figure 3-7 - Max Hold and Averaged Traces of 107.3MHz local radio station

Figure 3-8 – Averaged Trace of 107.3MHz local radio station baseband

Figure 3-9 - Max Hold and Averaged Traces of 90.9MHz local radio station

Figure 3-10 – Averaged Trace of 90.9MHz local radio station baseband

Figure 3-11 – Max Hold and Averaged Traces of 101.1MHz local radio station

Figure 3-12 – Averaged Trace of 101.1MHz local radio station baseband

3.2.5 Modulated Spectrum Characterization This section contains a comparison of the Standard interferer develop and an off-air radio station. It was decided by the IAAIS and NPR that the “middle of the road” amount of processing would be sufficient for these tests. Therefore, it was decided that the standard interferer would be modeled after the local Washington DC station operating at 101.1MHz. The plots of this section illustrate this.

Figure 3-13 – Averaged Traces of 101.1MHz Off-Air overlaid with the desired channel Modulated with 3dB Clipped USASI

Figure 3-14 – Max Hold Traces of 101.1MHz Off-Air overlaid with the desired channel Modulated with 3dB Clipped USASI

Figure 3-15 - Averaged Trace desired channel modulated with 3dB Clipped USASI baseband

4 Standard Methodologies The purpose of this section is to describe the methods by which all measurements will be obtained. The methodologies described here will be used repeatedly throughout the testing process. Each procedure described in the following subsections is standardized for all tests. 4.1 FM Band and RF Measurements This section details the methodologies that will be used to make RF measurements in the FM band. 4.1.1 FM Analog Power Methodology This procedure specifies the method for measuring analog FM power. FM signals have the rather unique characteristic of constant power regardless of the content or amplitude of the modulating signal. Therefore, power can be measured with or without a modulating signal present.

If a modulating signal is present, then the power across the entire channel is integrated in order to determine overall power. Traditionally, there are three common laboratory methods of performing this integration:

1) Numerically, by utilizing a DSP based Vector Signal Analyzer (VSA) or through the use of band power markers on a spectrum analyzer 2) Physically, by detecting heat in a thermal sensor 3) Electrically, by using a diode to rectify the signal and take advantage of a diode’s “square-law” region of operation

For our testing purposes, we shall use method (3) in conjunction with an average power meter and RMS responding diode detector.

Setup The measuring instrument shall be an HP437B average power meter with an 8481D diode detection sensor, and will be configured as shown in Table 4-1. Table 4-1 - HP 437B Setup – Analog FM Power Parameter Description Sensor Type HP 8481D (diode detector) Limit Checking On Low Limit -70dBm High Limit -20dBm Cal. Factor 98.5% Note: “Preset” first, and then set above parameters

peak-to-average ratio of OFDM. Therefore, measurements with an IBOC signal must maintain 10dB of “headroom” below the sensor’s peak range.

In addition, note that measurements must be made under conditions with no interferers present since any out of band signal will artificially increase readings on the power meter.

Procedure 1) Configure the instrument according to the table found in the “Setup” section above. 2) The analog carrier can be either modulated or unmodulated. There is no procedural change for either case. 3) The power level shall be observed and the instantaneous reading recorded.

Presentation of Data The resulting measurement shall be expressed in dB units with a precision of 0.01dB 4.1.2 IBOC Hybrid Mode Power Methodology This procedure specifies the method for measuring the power of an IBOC signal in hybrid mode. A hybrid signal is the spectral sum of a traditional analog FM signal and OFDM digital carriers at the channel edges.

The true average power of this signal may be determined by integrating over the entire channel, using an average power meter or vector signal analyzer, as discussed in 4.1.1 above. However, the resulting number would not be easily related to traditional analog power measurements. For comparison purposes, we would like to be able to say that a -30dBm hybrid signal has the same amount of analog energy as a traditional –30dBm FM signal.

For this reason, all references to the power of a hybrid signal will actually be specifying power in the signal that results from a traditional analog FM signal. In other words, a -30dBm hybrid signal will actually have –30dBm of analog FM energy plus the energy resulting from the digital carriers.

Consequently, in order to exclude the digital energy from our measurements, it is not possible to directly measure hybrid mode power using a traditional average power meter (at least not without an impracticably steep bandpass filter). Therefore, there are three practical measurement methods that can be employed:

1) Remove digital sidebands and measure remaining analog power with a traditional average power meter or Vector Signal Analyzer 2) Use a Vector Signal Analyzer to numerically integrate over analog bandwidth 3) Use spectrum analyzer to measure power of the unmodulated analog carrier.

For our purposes, we will utilize the first method using an HP437B average power meter. This is facilitated by the fact that the test bed has electromechanical RF switches, which can readily switch the digital carriers in and out of the spectrum. In addition, the removal of the digital sidebands increases measurement accuracy.

Setup The measuring instrument shall be an HP437B average power meter with an RMS responding sensor, and will be configured as shown in Table 4-2. Table 4-2 - HP 437B Setup – Hybrid Mode Power Parameter Description Sensor Type HP 8481D (diode detector) Limit Checking On Low Limit -70dBm High Limit -20dBm Cal. Factor 98.5% Note: “Preset” first, and then set above parameters

Usage of the HP437B must take into consideration the dynamic range of the diode detector. Under no circumstances shall a measurement be taken outside the sensor’s measurement range. In addition, note that measurements must be made under conditions with no interferers present since any out of band signal will artificially increase readings on the power meter.

Procedure 1) Configure the instrument according to the tables found in the “Setup” section above. 2) The analog carrier can be either modulated or unmodulated. There is no procedural change for either case. 3) The digital carriers shall be removed by opening the appropriate electromechanical RF switch. The remaining energy will be purely analog FM. 4) The power level shall be observed and the instantaneous reading recorded.

Presentation of Data The resulting measurement shall be expressed in dBm, and rounded to the nearest tenth of a decimal place. It should be made clear that this power refers to the analog energy and does not represent the average power level of the entire hybrid signal (as discussed above).

4.1.3 Power in the Presence of Multipath Methodology This procedure defines the method that shall be used for determining the power of analog and hybrid signals in the presence of multipath. For a detailed discussion of the multipath scenarios to be used in testing refer to section 7 of this document.

Each multipath profile that will be used consists of a main path and one echo with a range of phase offsets. As the phase offset of the echo component is varied, the total envelope power of the signal will also change due to the constructive or destructive vector addition of the main path and the echo path.

If one wished to maintain a constant envelope power level at the receiver input terminals, the transmitted power level would have to be adjusted differently for each multipath echo phase offset. If one were to adopt this approach, the phase offset of the echo would no

longer be the only variable in each test. In addition, it would no longer be possible to correlate received signal power with geographical distance from a theoretical transmitter.

Since each test is designed to investigate the impact of only one variable (multipath), and the test results will ultimately be used in conjunction with geographical station coverage maps, the power of the desired signal will not be adjusted to compensate for a destructive echo phase. The power of a multipath-impaired signal will be expressed in terms of the power in the unimpaired main path only.

Setup The “bypass” multipath profile is used during this measurement. This consists of a single path with no loss.

Procedure 1) Load the bypass profile into the multipath simulator 2) Start the simulation 3) Disable any AGC functionality, or set AGC to “hold”. 4) For an analog signal, measure the power according to section 4.1.1. For a hybrid signal, measure the power according to section 4.1.2. 5) Load the multipath profile for the desired test (such as StaticMP1), but make no additional adjustments (such as AGC settings) to the simulator.

Presentation of Data The resulting power measurement shall comply with the format outlined in sections 4.1.1 or 4.1.2 of this document.

4.1.4 Ratio of Analog to Digital Power (IBOC Hybrid Mode) Methodology This procedure specifies the method used to determine the ratio of analog to digital power in a hybrid mode signal. In practice, the ratio of analog to digital power does not vary, and is not a user-defined parameter. It is a fixed parameter that is defined in the specifications and describes any hybrid mode signal. This measurement will therefore be performed periodically (during daily calibration) in order to verify that the test system is setup according to the definition of an iBiquity hybrid mode signal. iBiquity Digital Corporation has specified that the average digital power shall be 20dB below the average analog power. It should be emphasized that this is the average power of the digital carriers, not the peak power.

The procedure is very similar to that discussed in 4.1.2. Electromechanical RF switches are used to separate the digital subcarriers from the analog energy. The digital energy is switched out, and the analog power is measured according to section 4.1.1. Next, the digital energy is switched in and the analog switched out. The average digital energy is measured using an average power meter. The difference between these two measurements is verified to be 20dB.

Setup The measuring instrument shall be an HP437B average power meter with an RMS responding sensor, and will be configured as shown in Table 4-3. The HP437B provides a higher level of accuracy than the VSA, and as such is the preferred instrument for analog- to-digital ratio measurements. Table 4-3 - HP 437B Setup – Analog-to-Digital Ratio Parameter Description Sensor Type HP 8481D (diode detector) Limit Checking On Low Limit -70dBm High Limit -20dBm Cal. Factor 98.5% Note: “Preset” first, and then set above parameters

Usage of the HP437B must take into consideration the dynamic range of the diode detector. Under no circumstances shall a measurement be taken outside the sensor’s measurement range. In addition, extra care must be taken during IBOC measurements due to the high peak-to-average ratio of OFDM. Therefore, measurements with an IBOC signal must maintain 10dB of “headroom” below the sensor’s peak range.

Procedure 1) This measurement shall be made under conditions where no interferers are present since any out-of-band signal will artificially increase readings on the power meter. 2) The analog carrier can be either modulated or unmodulated. There is no procedural change for either case. 3) Remove the digital carriers by opening the appropriate electromechanical RF switch. The remaining energy will be purely analog FM. 4) Put the HP437B into “Relative” mode. 5) Remove the analog energy and re-insert the digital carriers via the electromechanical RF switches. 6) Record the “relative reading” on the HP437B. This is the ratio of analog-to-digital power.

Presentation of Data The measurement resulting from step six (6) shall be expressed in dB units with a precision of 0.1dB.

4.1.5 Additive White Gaussian Noise Power Methodology This subsection describes the methods which shall be used to measure the average power of an Additive White Gaussian Noise (AWGN) signal.

There are several ways to express the power level of a noise signal such as Additive White Gaussian Noise (AWGN). The first method is bandwidth independent, and has units equal to degrees Kelvin. If an AWGN signal has a power that is said to be 30,000K, then it has

the same amount of noise power as a theoretical resistor would at a physical temperature of 30,000 Kelvin. The second method is bandwidth dependent, and has units equal to Power per Unit Bandwidth. For our purposes, we need to measure noise power in a given bandwidth. Therefore, we shall use the latter method, and express our results in dBm/Hz.

Unfortunately, the measurement of noise power in dBm/Hz units requires some rather specialized procedures and/or equipment. There are two well-known methods used to measure the power level of AWGN in bandwidth dependent units.

The first method is informally referred to as the “ENB Method”. ENB is an abbreviation for Equivalent Noise Bandwidth. In reality, ENB is a property of a bandpass filter. Any bandpass filter has a passband with a finite bandwidth, which may be measured and characterized using network analyzers or similar instrumentation. This characterization may also involve the calculation of an ENB value for the filter (The details of ENB calculation are beyond the scope of this test procedure). This ENB calculation may then be used as a conversion factor to convert absolute power units (dBm) into power per unit bandwidth units (dBm/Hz). Using this methodology, one can band-limit AWGN noise using an ENB calibrated filter, and measure the resultant power in absolute units (using standard instrumentation such as a thermal power meter). These dBm units may then be converted to dBm/Hz units using the ENB conversion factor as follows:

dBm/Hz = dBm - 10log(ENB) where ENB is the bandpass filter’s Equivalent Noise Bandwidth, given in Hz units

The setup and procedures for performing a measurement using the ENB method are given in the sections below.

The second method of measuring noise power in dBm/Hz units employs an HP Vector Signal Analyzer (VSA). The VSA is a very sophisticated and rather unique device, because it is fundamentally a time domain device that performs rapid FFT’s for spectrum analysis. This architecture allows one to easily measure Power Spectral Density (PSD) without the normal spectrum analyzer concerns of calibrated resolution bandwidth filters. Since the PSD measurement provides results in dBm/Hz units, noise power density can also be directly measured in dBm/Hz units. The setup and procedures for this measurement methodology are also given below.

Unfortunately, both of these measurement methodologies (ENB and VSA) are fundamentally limited by the low level sensitivity of the instrumentation (either the power meter or the VSA). For example, a 30,000K noise signal has a power level which is rapidly approaching the inherent noise floor of the VSA, and is also significantly below the limits of any thermal power sensor. In order to overcome this limitation, noise power is measured at a much higher power level. Then, in order to achieve lower noise power levels, a known/calibrated attenuator is inserted in-line with the high level noise source. (Insertion of this inline attenuation is facilitated by our noise generation instrument, which contains precision attenuators installed internally).

Setup (ENB Method) The power measuring instrument shall be an HP437B average power meter with an 8481D diode detection sensor, and will be configured as shown in Table 4-4.

Table 4-4 - HP 437B Setup – Analog FM Power Parameter Description Sensor Type HP 8481D (diode detector) Limit Checking On Low Limit -70dBm Filter/Averaging Auto Note: “Preset” first, and then set above parameters

Because the AWGN noise will have a relatively high peak-to-average ratio, the power sensor must operate with at least 10dB of “headroom”. Therefore, this sensor may not be used to measure any AWGN noise power exceeding –30dBm.

Procedure (ENB Method) 1) Set the noise generator’s output attenuator to 0.0dB, and enable the calibrated internal FM Bandpass Filter (FM BPF). 2) Remove all other bandpass filters from the noise signal path, and replace with a “barrel” connector (Refer to the test bed schematics for locations of additional filters). 3) Measure the absolute power of the noise, using the instrumentation described above, and observing the appropriate power limits. 4) Measure the insertion loss (at 97.9MHz) of any filters which may have been removed in Step 2. Subtract the insertion loss of these filters from the measurement of Step 3. 5) Convert the result of Step 4 from dBm to dBm/Hz units using the equation described in the methodology section. Note that the ENB of the noise generator’s filter is available on the unit’s calibration certificates. 6) Replace any bandpass filters which may have been removed in Step 2.

Presentation of Data (ENB Method) The result of Step 5, above, shall be given in dBm/Hz units and expressed with a precision of 0.1dB.

Setup (VSA Method) The power measuring instrument shall be an HP89441A Vector Signal Analyzer (VSA), and will be configured as shown in Table 4-5.

Table 4-5 - HP Vector Signal Analyzer Setup – Noise Measurements Parameter Setting Note: “Preset” first, and then set parameters below Center Frequency 97.9 MHz Span 1.0 MHz Range (sensitivity) Highest sensitivity before “over” LED Resolution Bandwidth 10kHz (auto-coupled)

Measurement Data PSD (Power Spectral Density) Averages 250 Marker Frequency 97.9MHz

Procedure (VSA Method) 1) Setup the HP 89441 VSA according to Table 4-5. 2) Note that the marker must be positioned to take measurements at 97.9MHz 3) Set the noise generator output attenuator to 0.0dB, and the FM bandpass filter inline/enabled. 4) Observe the power spectral density (PSD) marker reading on the VSA, ensuring that the running average has reached at least 250.

Presentation of Data (VSA Method) The resulting measurement will be given in dBm/Hz units, and expressed with a precision of 0.1dB. 4.1.6 Deviation and Channel Configuration Methodology A typical broadcast FM signal has numerous configuration parameters, which need to be regularly monitored and verified. Some of these parameters include main channel modulation, pilot injection, SCA injection and SCA modulation.

Throughout the testing process, there is a need to measure and verify these parameters. In recent years the market has seen the introduction of broadcast grade modulation monitors that utilize DSP technology for more accurate measurements. These units reduce calibration drift and uncertainty. In addition, the digital readouts remove much of the guesswork from directly reading an analog deflection meter. For these reasons, the testing process shall utilize broadcast grade DSP based modulation instruments to measure the various FM channel modulation parameters.

The modulation monitors to be used are a Belar FMMA-1 Modulation Analyzer, a Belar FMSA-1 Digital Stereo Monitor and a Belar SCMA-1 SCA Modulation Monitor in conjunction with a Belar RFA-4 Down Converter. The performance of these devices is documented in the proof of performance record, as referenced in section 1.4 of this document. Further verification of the Belar FMMA-1 accuracy will be obtained through comparison with a Modulation Sciences Inc. FM Modulation Monitor.

Setup The FM channel characteristics shall be measured using the instrumentation and setup shown in Figure 4-1. A complete listing of the setup of the Belar devices may be found in the test bed proof-of-performance record of test results, as referenced above in section 1.4.

Belar RFA-4 Belar SCMA-1 SCA Mod. Down Converter Mon. and Demodulator

FM BandPass Filter Belar FMMA-1 Mod. FM Splitter Analyzer (w/ Option 01) Belar FMSA-1 Stereo BandPass Demodulator Filter From Test Bed Modulation Sciences Amp FM Modulation Monitor

Figure 4-1 - Setup for Modulation & Configuration Readings

Notes: 1) The bandpass filter between the splitter and the Belar down converter blocks the L.O. signal, which would otherwise be conducted out of the RFA-4 input. 2) The amplifier is necessary to achieve an appropriate operating power level for the Modulation Sciences monitor.

Procedure The equipment configured as shown in Figure 4-1 shall be used to make measurements as follows:

1) Total Peak Modulation a) Belar FMMA-1 b) Modulation Sciences Inc. FM Modulation monitor 2) Pilot Injection a) Belar FMSA-1 b) Modulation Sciences Inc. FM Modulation monitor 3) Individual Left and Right Modulation a) Belar FMSA-1 4) Subcarrier Injection a) Belar SCMA-1 b) Modulation Sciences Inc. FM Modulation monitor 5) Subcarrier Percent Modulation a) Belar SCMA-1

The test engineer will verify that the Modulations Sciences Monitor and the Belar devices do not yield results that differ by more than a few percent when making the appropriate measurements (measurements 1, 2, and 4 listed above).

It is important to note that these measurements may only be performed on analog signals. In other words, all measurements must be made with the IBOC sidebands removed. These instruments are not designed to provide accurate readings in the presence of IBOC sidebands, and erroneous readings will result if such a measurement is attempted.

Presentation of Data The resulting data shall be expressed in percentage (%) units with a precision of 0.1%. 4.2 Baseband Audio Measurements This section details the methodologies that will be used to make audio measurements. 4.2.1 Signal-to-Noise Ratio (SNR) Methodology This procedure defines the method for finding the SNR of demodulated analog audio, as seen at the output of an FM SCA receiver.

SNR is the ratio of some reference signal to the baseband noise floor of the Device Under Test (DUT). In our case, the DUT is not only the receiver, but also the entire broadcast system and communications channel (including interferers). For our purposes, the reference signal is defined as the desired signal in the interference scenario under test. Refer to the subsections of 3.1 for the definitions of various desired signals.

The noise measurements shall comply with the ITU-R 468.4 standard2. This standard defines the characteristics of a “quasi-peak” voltmeter to be used for detection, and also a filter that is used to weight the response. The weighting filter is an attempt to model the human auditory system. The human ear, for example, is not as sensitive to low frequency noise as it is to mid frequency noise. Therefore low frequencies carry less weight in an SNR measurement that adopts the 468 method.

Setup The measuring instrument shall be an Audio Precision System One, with the analyzer configured as shown in Table 4-6.

Table 4-6 Audio Precision Setup for S/N Measurements Analyzer Settings Measure A Amplitude Range Auto BP/BR Freq. Auto Detector Q-Peak Bandwidth 10Hz 22kHz Filter CCIR-468 Channel A Input 100kW Range Auto Channel B Input 100kW Range Auto

2 International Telecommunications Union, ITU Rec.468-4, “Measurement of Audio-Frequency Noise Voltage Level in Sound Broadcasting”

Procedure 1) Apply the desired signal to the receiver. This will be the reference signal for the SNR measurement. 2) Set up the Audio Precision analyzer according to Table 4-6. 3) Use the Audio Precision analyzer to measure the value of the voltage generated across the appropriate output of the receiver. Record this value for use in step 7. 4) Remove modulating audio from the input to the SCA generator. 5) Use the Audio Precision analyzer to measure the value of the voltage remaining at the output of the receiver. Repeat this measurement until 40 readings are obtained. 6) Find the statistical mean of the 40 readings. 7) Find the ratio of the result from step 3 (in volts) to the result from step 6 (also in volts).

Presentation of Data The ratio resulting from step 7 (above) is a unit-less quantity, which should be expressed in dB. Since this ratio is derived from voltage units, the calculation shall be: 20*log(ratio). The result shall be rounded to the nearest tenth of a dB.

5 Test Procedures This section describes in detail, the test conditions that will be utilized during the testing process. 5.1 SCA Receiver Performance in the Presence of Analog and IBOC Adjacent Channels Objectives Characterize the performance of analog SCA receivers in the presence of both analog and hybrid adjacent channel interference. Quantify the difference between the impact of analog and hybrid interferers in various channel conditions.

Methodology In order to meet these objectives the test bed will be utilized to generate various adjacent channel interference scenarios. The performance of SCA receivers will be evaluated for each scenario with an analog and a hybrid adjacent channel interferer. For a list of the receivers to be tested, refer to section 6 of this report.

Evaluation of performance will be measured in terms of the SCA receiver’s audio signal-to- noise ratio (SNR). For all of the adjacent channel interference scenarios the SNR of each SCA receiver will be measured at varying adjacent channel power levels.

Upon completion of these tests, the test results data will be separated into cases with analog interferers and cases with hybrid interferers. The difference between hybrid and analog adjacent channel interference can then be objectively evaluated.

Test Conditions Table 5-1 illustrates all of the test interference conditions under which the SCA receiver performance will be tested in the presence of adjacent channel interference. The scenarios listed were chosen by the IAAIS and NPR. The following information may be helpful in understanding the table:

1. Each row of the table represents one test, and each test is assigned a unique identification number in the # column. 2. The Desired column indicates the mode, power level, and the SCA frequency to be tested on the desired channel. The SCA frequency to be tested is either 67kHz or 92kHz. The desired mode is Analog and the power level is Moderate (-62dBm) for all of these tests. 3. In the interferer columns, the mode of the interferer is indicated as Analog or Hybrid. Next to the mode of the interferer is the range of D/U to be used. D/U is the ratio of the power in the desired signal to the power in the undesired signal. This value is expressed in dB units. 4. The AWGN column indicates the presence or absence of a broadband noise floor in the test scenario. If additional noise is present in a test, the absolute power level is indicated in Kelvin units.

Table 5-1 Adjacent Channel Interference Scenarios

# Lower 2nd adj. Lower 1st adj. Desired Upper 1st adj. Upper 2nd adj. AWGN 8001 Analog: +30à0 Analog: 67kHz: Moderate None 8002 Hybrid: +30à0 Analog: 67kHz: Moderate None 8003 Analog: +30à0 Analog: 67kHz: Moderate 30,000K 8004 Hybrid: +30à0 Analog: 67kHz: Moderate 30,000K 8005 Analog: 67kHz: Moderate Analog: +30à0 None 8006 Analog: 67kHz: Moderate Hybrid: +30à0 None 8007 Analog: 67kHz: Moderate Analog: +30à0 30,000K 8008 Analog: 67kHz: Moderate Hybrid: +30à0 30,000K 8009 Analog: 0à-42 Analog: 67kHz: Moderate None 8010 Hybrid: 0à-42 Analog: 67kHz: Moderate None 8011 Analog: 0à-42 Analog: 67kHz: Moderate 30,000K 8012 Hybrid: 0à-42 Analog: 67kHz: Moderate 30,000K 8013 Analog:67kHz: Moderate Analog: 0à-42 None 8014 Analog:67kHz: Moderate Hybrid: 0à-42 None 8015 Analog:67kHz: Moderate Analog: 0à-42 30,000K 8016 Analog:67kHz: Moderate Hybrid: 0à-42 30,000K 8017 Analog: +30à0 Analog: 92kHz: Moderate None 8018 Hybrid: +30à0 Analog: 92kHz: Moderate None 8019 Analog: +30à0 Analog: 92kHz: Moderate 30,000K 8020 Hybrid: +30à0 Analog: 92kHz: Moderate 30,000K 8021 Analog: 92kHz: Moderate Analog: +30à0 None 8022 Analog: 92kHz: Moderate Hybrid: +30à0 None 8023 Analog: 92kHz: Moderate Analog: +30à0 30,000K 8024 Analog: 92kHz: Moderate Hybrid: +30à0 30,000K 8025 Analog: 0à-42 Analog: 92kHz: Moderate None 8026 Hybrid: 0à-42 Analog: 92kHz: Moderate None 8027 Analog: 0à-42 Analog: 92kHz: Moderate 30,000K 8028 Hybrid: 0à-42 Analog: 92kHz: Moderate 30,000K 8029 Analog: 92kHz: Moderate Analog: 0à-42 None 8030 Analog: 92kHz: Moderate Hybrid: 0à-42 None 8031 Analog: 92kHz: Moderate Analog: 0à-42 30,000K 8032 Analog: 92kHz: Moderate Hybrid: 0à-42 30,000K

Setup Please refer to the test bed schematics, located in the proof-of-performance record for details of the test platform setup. Note that SW7 through SW12 shall be set such that the multipath simulators are bypassed. SW1 through SW6 shall be used to set each channel to either analog, hybrid or off mode. AT1 through AT3 shall be used to set D, U1 and U2 signal power levels in 0.25dB increments. Noise power levels shall be set with the AWGN generator’s internal attenuator in 0.1dB increments.

Procedure 1) All of the analog SCA receivers (see section 6) shall be used for this test series. 2) Table 5-1 will be used as a guide for setting up each interference scenario. Each row of this table shall be executed individually. 3) Establish the noise floor according to the AWGN column of the test grid. Refer to section 3.1.9 for a definition of AWGN noise, and 4.1.5 for the procedures of AWGN noise power measurement. 4) Establish the signal strength, frequency, and mode of the desired signal as indicated in the test grid. For a complete definition of the desired analog and desired hybrid signals, refer to sections 3.1.2 and 3.1.5, respectively. The procedures for measuring FM power may be found in sections 4.1.1 and 4.1.2.

5) Establish the signal strength, frequency, and mode of the interfering signals as indicated in the test grid. For a complete definition of the undesired analog interferer and undesired hybrid interferer signals, refer to 3.1.7 and 3.1.8, respectively. The procedures for measuring FM power may be found in sections 4.1.1 and 4.1.2. 6) Modulate the desired main channel and the interferer with the appropriate CD recordings of Clipped USASI (3.2). 7) Begin taking SNR measurements, as described in section 4.2.1, at the appropriate D/U levels.

Presentation of Data A plot of the SCA receiver SNR as a function of the ratio of the desired channel power level to the undesired channel power level (D/U) will be generated.

5.2 SCA Receiver Performance in the Presence of an Analog or IBOC Host Objectives Characterize the performance of analog SCA receivers in the presence of both analog and hybrid “Host” signals. Quantify the difference between the impact of analog and hybrid host signals in various channel conditions.

Methodology In order to meet this objective the test bed will be utilized to generate Analog and Hybrid main channel signals under various channel conditions. The performance of SCA receivers will be evaluated for each condition. For a list of the receivers to be tested, refer to section 6 of this report.

Evaluation of performance will be measured in terms of the SCA receiver’s audio signal-to- noise ratio (SNR). For all of the host compatibility channel scenarios the SNR of each SCA receiver will be measured.

Upon completion of these tests, the test results data will be separated into cases where the host was analog and where the host was a hybrid signal. The difference between hybrid and analog host compatibility can then be objectively evaluated.

Test Conditions Table 5-2 illustrates all of the test interference conditions under which the SCA receiver performance will be tested for host compatibility. The scenarios listed were chosen by the IAAIS and NPR. The following information may be helpful in understanding the table:

3. Signal Type column indicates the signal type to be used for testing. For definitions of these signal types refer to sections 3.1.2 and 3.1.4. 4. The AWGN column indicates the presence or absence of a broadband noise floor in the test scenario. If additional noise is present in a test, the absolute power level is indicated in Kelvin units. 5. Table 5-2 Host Compatibility Scenarios

# Desired Signal Type AWGN 8101 Analog: 67kHz: Moderate Analog II None 8102 Hybrid: 67kHz: Moderate Hybrid II None 8103 Analog: 67kHz: Moderate Analog II 30,000K 8104 Hybrid: 67kHz: Moderate Hybrid II 30,000K 8105 Analog: 92kHz: Moderate Analog II None 8106 Hybrid: 92kHz: Moderate Hybrid II None 8107 Analog: 92kHz: Moderate Analog II 30,000K 8108 Hybrid: 92kHz: Moderate Hybrid II 30,000K

Setup Please refer to the test bed schematics, located in the proof-of-performance record for details of the test platform setup. Note that SW7 thru SW12 shall be set such that the multipath simulators are bypassed. SW1 thru SW6 shall be used to set each channel to either analog, hybrid or off mode. AT1 thru AT3 shall be used to set the Desired signal power level in 0.25dB increments. Noise power levels shall be set with the AWGN generator’s internal attenuator in 0.1dB increments.

Procedure 1) All of the analog SCA receivers (see section 6) shall be used for this test series. 2) Table 5-1 will be used as a guide for setting up each interference scenario. Each row of this table shall be executed individually. 3) Establish the noise floor according to the AWGN column of the test grid. Refer to section 3.1.9 for a definition of AWGN noise, and 4.1.5 for the procedures of AWGN noise power measurement. 4) Establish the signal strength, frequency, subcarrier frequency and mode of the desired signal as indicated in the test grid. For a complete definition of the desired analog and desired hybrid signals, refer to sections 3.1.2 and 3.1.5, respectively. The procedures for measuring FM power may be found in sections 4.1.1 and 4.1.2. 5) Modulate the desired main channel with the appropriate CD recording of Clipped USASI (3.2). 6) Begin taking SNR measurements, as described in section 4.2.1, at the appropriate D/U levels.

Presentation of Data The SNR measurement data will be presented in tabular form with a precision of 0.1 dB.

5.3 SCA Receiver Performance in the Presence of an Analog or IBOC Host in Multipath Conditions Objectives Characterize the performance of analog SCA receivers in the presence of both analog and hybrid “Host” signals under multipath impaired channel conditions. Quantify the difference between the impact of analog and hybrid host signals in these channel conditions.

Methodology In order to meet this objective the test bed will be utilized to generate Analog and Hybrid main channel signals under multipath impaired channel conditions. The performance of four SCA receivers will be evaluated for each scenario.

Evaluation of performance will be measured in terms of the SCA receiver’s audio signal-to- noise ratio (SNR). For all of the host compatibility with multipath test conditions the SNR of four SCA receivers will be measured. For a list of the receivers to be subjected to these tests, refer to section 6 of this report.

Test Conditions Table 5-3 illustrates all of the test interference conditions under which the SCA receiver performance will be tested for host compatibility. The scenarios listed were chosen by the IAAIS and NPR. The following information may be helpful in understanding the table:

1. Each row of the table represents one test, and each test is assigned a unique identification number in the # column. 2. The Desired column indicates the mode, power level of the desired channel. The desired mode is Analog or Hybrid and the power level is Moderate (-62dBm) for all of these tests. 3. Signal Type column indicates the signal type to be used for testing. For definitions of these signal types refer to sections 3.1.1, 3.1.3, 3.1.4, and 3.1.6. 4. The Multipath column indicates the multipath scenario that was used during testing. For a detailed description of the multipath scenarios used refer to section 7 of this document. 5. The Echo Phase column indicates the range of phases to be tested. The phase will be stepped over the range indicated in steps of 45 degrees. 6. The SCA Under Test column indicates the subcarrier that is being evaluated (either 67kHz or 92kHz).

Table 5-3 – Host Compatibility with Multipath Scenarios

# Desired Signal Type Multipath Echo Phase SCA Under Test 8201 Analog: Moderate Analog III StaticMP1 -187 à -7 67kHz 8202 Hybrid: Moderate Hybrid III StaticMP1 -187 à -7 67kHz 8203 Analog: Moderate Analog I StaticMP1 -187 à -7 67kHz 8204 Hybrid: Moderate Hybrid I StaticMP1 -187 à -7 67kHz

8205 Analog: Moderate Analog III StaticMP2 -187 à -7 67kHz 8206 Hybrid: Moderate Hybrid III StaticMP2 -187 à -7 67kHz 8207 Analog: Moderate Analog I StaticMP2 -187 à -7 67kHz 8208 Hybrid: Moderate Hybrid I StaticMP2 -187 à -7 67kHz 8209 Analog: Moderate Analog III StaticMP3 -16 à -164 67kHz 8210 Hybrid: Moderate Hybrid III StaticMP3 -16 à -164 67kHz 8211 Analog: Moderate Analog I StaticMP3 -16 à -164 67kHz 8212 Hybrid: Moderate Hybrid I StaticMP3 -16 à -164 67kHz 8213 Analog: Moderate Analog III StaticMP1 -187 à -7 92kHz 8214 Hybrid: Moderate Hybrid III StaticMP1 -187 à -7 92kHz 8215 Analog: Moderate Analog I StaticMP1 -187 à -7 92kHz 8216 Hybrid: Moderate Hybrid I StaticMP1 -187 à -7 92kHz 8217 Analog: Moderate Analog III StaticMP2 -187 à -7 92kHz 8218 Hybrid: Moderate Hybrid III StaticMP2 -187 à -7 92kHz 8219 Analog: Moderate Analog I StaticMP2 -187 à -7 92kHz 8220 Hybrid: Moderate Hybrid I StaticMP2 -187 à -7 92kHz 8221 Analog: Moderate Analog III StaticMP3 -16 à -164 92kHz 8222 Hybrid: Moderate Hybrid III StaticMP3 -16 à -164 92kHz 8223 Analog: Moderate Analog I StaticMP3 -16 à -164 92kHz 8224 Hybrid: Moderate Hybrid I StaticMP3 -16 à -164 92kHz

Procedure 1) Four of the analog SCA receivers, chosen by the IAAIS and NPR, shall be used in this test series. For a list of the receivers to be tested see section 6. 2) Table 5-3 will be used as a guide for setting up each interference scenario. Each row of this table shall be executed individually. 3) Setup the multipath scenario that the test grid indicates. 4) Establish the signal strength, frequency, subcarrier frequency and mode of the desired signal as indicated in the test grid. For a complete definition of the desired analog and desired hybrid signals, refer to sections 3.1.1, 3.1.3, 3.1.4, and 3.1.6 respectively. The procedures for measuring FM power may be found in sections 4.1.1 and 4.1.2. 5) Modulate the desired main channel with the appropriate CD recording of Clipped USASI (3.2). 6) Adjust the phase offset of the echo. 7) Take SNR measurements, as described in section 4.2.1. 8) Repeat steps 6 and 7 until measurements have been taken over the entire range indicated by the test grid.

Presentation of Data The results of these tests will be present in a plot of SNR versus the phase offset of the Echo.

6 - Appendix A - List of Receivers Under Test

Table 6-1 – Receivers to be tested for Adjacent Channel Compatibility and Host Compatibility without Multipath.

Radio Make/Model No: Type Serial No: Dayton AF210 (REF) 67 kHz and 92 kHz Settings 2101341 McMartin TR-E5/55M 67 kHz 286834 Norver 67 kHz 67 kHz A0012461 Dayton AF200 67 kHz 810239 ComPol SCA-RLA 67 kHz 1004 ComPol SCA-RLA 67 kHz 1005 CozmoCom 92 kHz 92 kHz 0073696 Dayton AF200 92 kHz 810238 ComPol SCA-RLA 92 kHz 1006 ComPol SCA-RLA 92 kHz 1007

Table 6-2 – Receivers to be tested for Host Compatibility with Multipath channel impairments.

Radio Make/Model No: Type Serial No: McMartin TR-E5/55M 67 kHz 286834 ComPol SCA-RLA 67 kHz 1005 ComPol SCA-RLA 92 kHz 1006 CozmoCom 92 kHz 92 kHz 0073696

7 - Appendix B - Multipath Profiles This appendix describes the multipath profiles that are to be used during the testing process.

7.1 Characterization of Static Multipath Profiles Since signal-to-noise ratio measurement in the presence of dynamic multipath may not be repeatable without averaging the results of an impracticably large number of measurements, the IAAIS and NPR agreed to the use of static multipath scenarios.

It was decided that there would be three static multipath scenarios developed for testing, each with one echo of varying phase offset. The echo was adjusted to have either 1.2ms or 3ms of delay.

Figure 7-1 illustrates the device setup used to develop the multipath scenarios, and produce the plots in the following sections. The tracking generator of the spectrum analyzer was amplified in order to simulate the power levels seen by the test bed during normal operation.

Spectrum Test Bed Analyzer Amp Pad

Tracking RF Input to Generator Test Bed Output

RF Output RF Input From Test bed

Figure 7-1 Device setup used to develop static multipath scenarios.

The plots on the following pages illustrate the frequency response of the channel with the worst case scenarios (echo with no attenuation) for both of the delays chosen. These plots are overlaid with an unimpaired IBOC signal. The pathological phase offset is defined as the phase offset that causes a null at 97.82MHz, which is between the lower sidebands caused by the 92kHz and 67kHz subcarriers. Four other phase offsets were chosen such that the null is near enough, in frequency, to the subcarrier sidebands that some interference is still discernable.

Figure 7-2 – Static Multipath, one echo, no attenuation, 1.2ms delay, 173 degrees phase offset

Figure 7-3 – Static Multipath, one echo, no attenuation, 1.2ms delay, -142 degrees phase offset

Figure 7-4 – Static Multipath, one echo, no attenuation, 1.2ms delay, -97 degrees phase offset

Figure 7-5 – Static Multipath, one echo, no attenuation, 1.2ms delay, -52 degrees phase offset

Figure 7-6 – Static Multipath, one echo, no attenuation, 1.2ms delay, -7 degrees phase offset

Figure 7-7 – Static Multipath, one echo, no attenuation, 3ms delay, -16 degrees phase offset

Figure 7-8 – Static Multipath, one echo, no attenuation, 3ms delay, 29 degrees phase offset

Figure 7-9 – Static Multipath, one echo, no attenuation, 3ms delay, 74 degrees phase offset

Figure 7-10 – Static Multipath, one echo, no attenuation, 3ms delay, 119 degrees phase offset

Figure 7-11 – Static Multipath, one echo, no attenuation, 3ms delay, 164 degrees phase offset

7.2 Multipath Profiles - StaticMP1 StaticMP1 is designed to have one type echo, with 6dB of attenuation, 1.2ms delay and a varying phase offset. The plots of this subsection illustrate the frequency response of the channel in the presence of this type of echo for all of the phase offsets to be used in testing. The frequency response is overlaid with an unimpaired IBOC signal.

Figure 7-12 – StaticMP1, 173 degree phase offset

Figure 7-13 – StaticMP1, -142 degree phase offset

Figure 7-14 – StaticMP1, -97 degree phase offset

Figure 7-15 – StaticMP1, -52 degree phase offset

Figure 7-16 – StaticMP1, -7 degree phase offset

7.3 Multipath Profiles – StaticMP2 StaticMP2 is designed to have one type echo, with 10dB of attenuation, 1.2ms delay and a varying phase offset. The plots below illustrate the frequency response of the channel in the presence of this type of echo for all of the phase offsets to be used in testing. The frequency response is overlaid with an unimpaired IBOC signal.

Figure 7-17 – StaticMP2, 173 degree phase offset

Figure 7-18 – StaticMP2, -142 degree phase offset

Figure 7-19 – StaticMP2, -97 degree phase offset

Figure 7-20 – StaticMP2, -52 degree phase offset

Figure 7-21 – StaticMP2, -7 degree phase offset

7.4 Multipath Profiles – StaticMP3 StaticMP3 is design to have one type echo, with 10dB of attenuation, 3ms delay and a varying phase offset. The plots below illustrate the frequency response of the channel in the presence of this type of echo for all of the phase offsets to be used in testing. The frequency response is overlaid with an unimpaired IBOC signal.

Figure 7-22 – StaticMP3, -16 degree phase offset

Figure 7-23 – StaticMP3, 29 degree phase offset

Figure 7-24 – StaticMP3, 74 degree phase offset

Figure 7-25 – StaticMP3, 119 degree phase offset

Figure 7-26 – StaticMP3, 164 degree phase offset