iBiquity Digital Corporation 8865 Stanford Boulevard, Suite 202 Columbia, Maryland 21045 (410) 872-Independence Boulevard Warren, New Jersey 07059 (908) 580-7000
Impact of National Rollout of IBOC on Analog Radio Listenership

September 2000

2000 iBiquity Digital Corporation

Overview

iBiquity has demonstrated through the NRSC test program and its own internal testing that its system can provide robust digital performance throughout the market served by todays analog stations. The tests also show the iBiquity system eliminates most of the impairments that degrade existing analog broadcasting at the same time the system offers higher audio fidelity. This study provides supplemental support to the test findings that IBOC can be introduced without meaningful interference to existing analog stations. The objective of the study was to estimate the potential impact that a large-scale rollout of IBOC in the United States would have on FM analog radio station listenership.

Purpose

This study analyzes the potential impact, at various digital power levels, that an IBOC rollout would have on FM analog radio station listenership. Of paramount importance in the design of an IBOC system is the compatibility of the new digital signals with existing analog service. IBOC adds low-level energy to the sidebands of existing analog FM signals that could potentially impact analog listenership. iBiquitys design has optimized the spectral signature of the waveform, the spectral placement of the energy and trade-offs between digital coverage and analog interference to provide excellent digital coverage with minimal impact to the first adjacent channel analog reception.

Process

The study was conducted using the following process: 1. Select representative analog radio receivers for subjective listening tests Different classes of receivers vary dramatically with regard to interference. To account for the disparity in receiver functionality, four classes of receivers car stereo, home hi-fi, boombox and clock radio were selected for the study. A commercially available receiver was selected to best represent each class based on characteristics typical of that class and the number of units in use in the United States. The receivers selected were: Delco car stereo (16196204), Pioneer home hi-fi (SX203), Philips boombox (AX1020) and Sony clock radio (ICF-C390). Conduct point-of-failure (POF) subjective listening tests Subjective listening tests were used to determine analog points of failure for each radio receiver. The tests included failures due to weak signals, co-channel and first adjacent interference. The tests were conducted in a laboratory with listeners evaluating when they would change the radio station due to unacceptable listening conditions such as low signal-to-noise ratios. The point of failure was used to determine signal levels in dBu for weak signal conditions and desired-to-undesired (D/U) signal ratios in dB for interference. To determine the effect of introducing IBOC, digital carriers were added to first adjacent channel interferers generated in the laboratory. The power of the IBOC sidebands was increased until listeners indicated that they would change the radio station due to interference. This POF was converted to an equivalent D/U ratio. 3. Select a statistically valid, random-sample of 46 FM radio stations for analysis A random sample of 46 FM radio stations in Arbitron markets was selected for further analysis. The random sample was drawn from the population of Arbitron-rated FM stations. A statistically valid sample size of 46 stations was necessary to determine the mean percentage of impacted listeners with 95% confidence. 4. Model analog coverage on 46 study stations based on subjective listening tests

-2Signal levels from the analog POF subjective listening tests were used with propagation software to predict the weak signal coverage for the statistically selected sample of 46 FM radio stations. Propagation models used the D/U ratios from the subjective listening tests to predict areas of coverage lost to analog co- and first-adjacent channel interference. The areas lost to interference were subtracted from the weak signal coverage area to create an interference-free service area for each of the 46 analog FM radio stations. The results of the interference-free service areas were compared to actual Arbitron listener information for the study stations. The interference-free service areas captured substantially all listeners included in Arbitron ratings data. 5. Model potential first adjacent IBOC interference for 46 study stations based on subjective listening tests Using propagation software, the D/U ratio at the POF, determined in the subjective listening tests, was then used to calculate any losses to the existing analog coverage area for the study stations resulting from all first adjacent stations implementing Hybrid IBOC at a power level of 23 dB per sideband. Fan example coverage map demonstrating the impact to existing analog coverage is included in Supplement C. Using actual listener information from Arbitron, determine number of listeners potentially impacted Using actual listener data obtained from Arbitron Corporation, the number of listeners in areas subject to IBOC first adjacent interference for each study station was calculated. The potentially impacted listeners were divided by the stations total audience to generate a percentage estimate of potential impact.
Refer to Supplements A and B for a detailed description of each step in the methodology and other additional study-related information.

Results

The study found that, on average, 99.36% of an FM stations listeners will not be impacted by the introduction of IBOC, even in the event all first adjacent radio stations implement IBOC. 1

Conclusion

A large-scale national rollout of IBOC will have an immaterial effect on analog radio broadcasters and listeners. The study finding that, on average, 99.36% of an FM stations listeners will not be impacted by IBOC introduction is most likely overstated for three reasons. First, IBOC will be implemented over a number of years and the impact to a station will not reflect the study findings until all first adjacents adopt IBOC. Second, future advances in transmission and receiver technology will serve to reduce the listener impact. Finally, over time, listeners are expected to substitute digital receivers for analog receivers further reducing the analog listener impact. As a result, the actual percentage of listeners impacted is likely to be even less than the insignificant number observed in the study. This study provides additional support to the test findings that IBOC can be introduced without meaningful interference to existing analog stations.

At an IBOC power level of -23dB per sideband.
Supplement A Investigation to Predict Potential Losses in Analog Coverage Resulting From FM Hybrid IBOC Implementation
Overview Of paramount importance in the design of an IBOC system is the compatibility of the new digital signals with existing analog service. iBiquity has carefully designed its IBOC system to optimize the spectral placement, spectral character1 and power levels of its FM DAB sidebands to maximize digital audio coverage while minimizing the impact to existing analog coverage. Toward this end, iBiquity has performed a number of laboratory experiments to establish Point of Failure (POF2) metrics and their equivalent signal level in dBu or in the case of interferers Desired to Undesired (D/U) ratios. The signal strengths and D/U ratios were then used in conjunction with propagation modeling to predict the listenable coverage of 46 statistically sampled FM stations in the presence of receiver front-end noise, co- and first-adjacent channel interference and selective fading, multipath. These POFs were further used to predict any changes in the coverage of these stations if all first adjacent stations were to convert to hybrid FM broadcasting at various IBOC power levels. Finally, actual listener information from Arbitron for the 46 stations was used to assess the potential impact of any coverage changes on analog listenership. A more detailed discussion of the procedures in the study follows: 1. Select representative analog radio receivers for subjective listening tests Different classes of radio receivers vary dramatically with regard to susceptibility to interference caused by proximate stations on co- and first-adjacent channels. To account for the disparity in receiver functionality, four classes of receiversCar Stereo, Home Hi-Fi, Boombox and Clock Radiowere included in the study. iBiquity selected the following four commercial FM receivers to verify compatibility of FM Hybrid IBOC with existing analog services. (1) (2) (3) (4) Home HiFi Pioneer SX-203 Car Stereo Delco 16196204 Boombox Philips Magnavox AZ1020 Clock Radio Sony Dream Machine ICF-C390
A) Conduct point-of-failure (POF) subjective listening tests POF was determined by subjective, critical listening tests. For each of the four study receivers, a point-offailure metric was determined through a subjective evaluation program. Receivers were tested with 30 human test subjects in an acoustically treated listening room. Listening tests were conducted with the built-in speakers for the boombox and clock radio. The car radio was evaluated with external speakers and the home hi-fi was
Spectral character refers to the characteristics of the IBOC information when demodulated by an FM radio. iBiquitys IBOC signals are whitened and free of periodicities. Analysis has shown that a whitened recovered audio is a natural characteristic of FM reception and less annoying than a waveform with pure tones or tonal periodicities. Point of Failure as determined by subjectively evaluating the point where interference and/or noise will have caused 90% of the listeners to tune to another radio station or turn the radio off. The 90% figure was selected to eliminate statistical outliers in the subjective listening test thereby preventing an exaggerated estimate of the interference-free service area. The 90% POF provided for the interference-free service area that best matches a radio stations existing analog listening audience as reflected in its Arbitron ratings. iBiquitys hypothesis was that this area would capture nearly all of a radio stations Arbitron-rated listening audience. Subsequent analysis of Arbitron listening information confirmed iBiquitys hypothesis that a 90% POF captured nearly all of a radio stations listening audience, including listeners at or near the 90% POF contour, and therefore represents a reasonable estimate of analog coverage.

-4evaluated with headphones. Each subject was presented with a baseline, interference free, analog signal. Depending on the test being conducted, the level of the baseline signal was decreased or the interfering signal increased until the subject indicated that s/he would change stations or stop listening in a typical listening situation. In the case of the car radio, independent multipath fading3 was applied to the desired and interfering signals. The POF measurement for each receiver was defined as the point at which 90% of the test subjects indicated they would no longer listen to the station. This 90% POF was chosen to capture the vast majority of the Arbitron identified listeners and provide the most conservative assessment of IBOCs impact on existing listeners4. Once the analog POF was subjectively established, an objective SNR determination for each radio was made at its subjective POF for subsequent use. A) Analog POF Determination: This test records the D/U at which POF is reached for each of the receivers5. Since the FM spectrum is an interference-limited medium the test were performed with a co-channel interferer as the undesired signal. The audio used to modulate the desired and undesired FM signal was processed pop music. The first part of this test was subjective in nature employing 30 different listeners leading to a D/U ratio for a given subjective POF. Each subject was presented with a baseline, interference free, analog signal. The level of the interfering co-channel signal was increased until the subject indicated that s/he would change stations or stop listening in a typical listening situation. The subjective evaluation was repeated for the four types of radios. The tests were then reconfigured to determine objectively the SNR at the D/U ratio that has been identified as the subjective POF for each radio. Three types of analog POF metrics were determined through subjective evaluation. These included failures due to: 1) weak signal (weak signal POF) 2) co-channel (co-channel POF) 3) first-adjacent interference (first adjacent POF) B) Lab Test Procedure: The test setup is shown in Supplement B. The test was performed as follows: 1) 2) 3) 4) 5) Transmit a 54-dBu baseline FM desired signal. Increase the level of the interferer signal until POF. Record the associated D/U level. Change the desired audio to a 1kHz tone. Set the levels of the desired and undesired signal as measured in the POF determination. Using the AudioPrecision System 2 measure the SNR 30 times in 30 seconds, or 300 times in 5 minutes for faded cases. 6) Average the SNR readings and record the result. 7) Repeat (1) through (6) for all receivers under test using independent multipath fading for the automotive radio only. 8) Repeat (1) through (7) using all test subjects.
In 1993, the EIA conducted multipath characterization tests in Salt Lake City, and subsequently created four profiles that are descriptive of the multipath environment. The urban fast profile (simulates driving at approximately 35 mph through a city street) was used for this study. iBiquitys hypothesis was that this area would capture nearly all of a radio stations Arbitron-rated listening audience. Subsequent analysis of Arbitron listening information confirmed iBiquitys hypothesis that a 90% POF captures nearly all of a radio stations listening audience, including some listeners at or near the 90% POF contour, and therefore represents a reasonable estimate of analog coverage. It was determined that the low POF value of the Sony Clock Radio rendered it the least susceptible to IBOC interference because of the limited analog reception area. For reasons of efficiency, this receiver was omitted in the analysis. The remaining stationary receivers, the Pioneer Hi-Fi and Philips Boombox offered more critical, and therefore more conservative, results.

-5The POF information from above was used to convert the POF metric to signal levels, in dBu for weak signal conditions and desired-to-undesired (D/U) signal ratios in dB for interference. The test procedure to convert subjective POFs to signal strength level and interference D/Us follows: 1) Weak Signal POF: a. Transmit a 54-dBu 1 kHz tone modulated FM desired signal. Remove the tone and observe the resultant level. b. Decrease the level of the desired signal until POF SNR metric is achieved. Record the associated input power level. The signal level is subsequently converted to a field strength. 2) Co-Channel POF a. Transmit a 54-dBu 1 kHz tone modulated FM desired signal. Remove the tone and observe the resultant level. b. Add a co-channel modulated with processed pink noise and increase its level until the subjective POF SNR is reached. Record the associated D/U level. 3) First Adjacent POF a. Transmit a 54-dBu 1 kHz tone modulated FM desired signal. Remove the tone and observe the resultant level. b. Change the interferer to a first-adjacent channel modulated with processed pink noise and increase its level until POF. At POF metric record the associated D/U level. 4) Repeat (1) through (3) for all FM receivers under test using independent multipath fading for the automotive radios only. C) Digital POF Determination: Since digital IBOC energy from a first-adjacent IBOC signal is located in the upper and lower sidebands of the first-adjacent FM hybrid6 signal, the digital energy in the sideband closest to the desired analog signal is the predominant source of IBOC interference. This energy is coincident with the desired analog signal, therefore, interfering IBOC sidebands can be characterized as a cochannel analog interferer at a power level equal to the digital power in that sideband7. To determine the IBOC POF metric, IBOC digital carriers were added to a first adjacent channel interferer generated in the lab. Tests were conducted to establish the IBOC POF for a range of IBOC power levels at five fixed D/U ratios, +12, +6, 0, -6, and -12 dB. At each of these D/U ratios, the SNR of the receiver under test was measured in the presence of a first adjacent channel IBOC interferer. This test was designed to find the maximum acceptable level of digital portion of the hybrid waveform. While observing the SNR on the desired signal, the IBOC sidebands on the adjacent channel were increased in power until the equivalent POF for each radio was reached. This was labeled the IBOC POF, which was subsequently converted to an equivalent D/U ratio. The test setup is shown in Supplement B. The test is performed as follows: 1) Transmit a 54-dBu, 1-kHz-tone-modulated FM signal, remove the tone and observe the resulting level. 2) Add baseline FM first adjacent channel to the host FM signal at a D/U = +12 dB.

IBOC technology allows for two modes of operation, hybrid and all-digital. This study measured the effects of hybrid IBOC signals, in which audio information is broadcast as both analog (compatible with existing analog receivers) and digital (compatible with future IBOC digital receivers). For example, the added interference caused by a first-adjacent, 10,000 watt, IBOC station can be characterized as the interference that would be caused by a co-channel, 51.1 watt, analog station
-63) Add baseline DAB to the first-adjacent interferer. Increase the level of its DAB sidebands starting at 40dBc until the audio SNR of the POF metric is reached. Record the IBOC POF level, D/U, in dB relative to its host power. 4) Repeat (1) through (3) for D/Us of +6, 0, 6 and 12 dB. 5) Repeat (1) through (4) for all radios under test. 3. Select statistically valid, random-sample of 46 FM radio stations for analysis
A random sample of 46 FM radio stations in Arbitron markets was selected for further analysis. The simple random sample was drawn from a population of 3,452 Arbitron-rated FM stations in the BIA Research Media Access Pro Database as of May 2000. Using the formula prob(|m |) < D) > P, a statistically valid sample size of 46 stations was necessary to determine the mean percentage of impacted listeners within a maximum allowable difference of 1% with 95% confidence. The stations were selected by randomly assigning numbers to each of the 3,452 population stations, sorting by those numbers, then selecting every 77th station. The variance estimate observed in the actual test results closely approximates the sample size calculation methodology. As a result, the sample size of 46 FM radio stations is appropriate to determine with 95% confidence and with a +/- 1% margin of error, the mean potential impact on FM station listenership caused by energy added during nationwide implementation of IBOC technology. 4. Model analog coverage on 46 study stations based on subjective listening tests
The theoretical coverage areas were determined by using POF weak signal contour values in propagation software based on the Longley-Rice8 model with 50/10 criteria (a median location where the signal is received 90% of the time). These results represent regions in which a stations signal would be listenable on the given receivers without interference from any other co- or first-adjacent radio stations. Dataworld, Bethesda, MD, an industry-leading information services company specializing in propagation and mapping of the AM and FM radio services, employed Longley-Rice 50/10 propagation models to predict the interference limited service areas of the 46 stations in the study. Propagation studies for each of the three remaining study receivers (Delco Car, Pioneer Hi-Fi, and Philips Boombox) were conducted using the POF metrics established in the laboratory subjective listening tests. First, the coverage area based on the weak signal POF model for each of the three receivers was calculated. The next step used the co-channel POF D/U ratio to calculate the areas lost to all relevant co-channel interferers. The third step used the first adjacent channel interference POF metrics to calculate the areas lost to all relevant first adjacent channels. Finally, Dataworld predicted the analog interference free service area by removing the areas of co- and adjacent channel interference from the noise limited coverage area. 5. Model potential first adjacent IBOC interference for 46 study stations based on subjective listening tests

Longley-Rice 50/10 propagation software was used in conjunction with the IBOC POF D/U ratio metric to determine any losses to the existing analog coverage area resulting from full-scale conversion to Hybrid IBOC broadcasting. The predicted lost regions for each receiver type and IBOC power level were generated for the 46 stations. A sample map showing lost regions for each IBOC power level, on a Delco receiver, may be found in Supplement C. 6. Using actual listener information from Arbitron, determine number of listeners potentially impacted
The Longley-Rice model, which is based on electromagnetic theory and on statistical analyses of both terrain features and radio measurements, predicts the median attenuation of a radio signal as a function of distance and the variability of the signal in time and in space.
-7To convert the impacted regions to actual listeners, ratings data by zip code for each sample station was obtained from the Arbitron Company. The impacted area calculated by Dataworld was reduced to block level regions. Using block level US Census data, the number of persons residing in each block of the impacted area was summed to create a total number of potentially impacted persons per zip code. The percentage of potentially impacted persons was then calculated by dividing the total persons in the impacted area by the total persons in a given zip code. In the final step, Arbitron listening data for the zip code was multiplied by the percentage of potentially impacted persons to yield the number of impacted persons. The listening data was for the spring and fall 1999 rating periods.9 Arbitron provided rating information for listeners in three situations: in car, at home, and at work. It was assumed that listening behavior in each of these three situations is independent (e.g. impact to home reception does not affect listening at work). Therefore, each data point was treated as a separate listener. This may have resulted in understating the percentage of actual listeners impacted, but not in the percentage of listening events impacted. Mobile listeners, outside the listenable interference-free service area were assumed commuters and were moved to the closest location within the service area. Mobile listener loss was calculated by comparing the impacted regions of the Delco car receiver to Arbitron in-car listeners. Stationary listener loss was calculated by comparing the impacted regions of both the Pioneer Home Hi-Fi and the Philips Boombox to the aggregate of Arbitron at-home and at-work listener. Stationary listener loss was weighted by assuming 38% of stationary listeners were using Pioneer Hi-Fi receivers and 62% were using Philips Boombox receivers. These weightings were based on the proportion of annual sales of high-end and low-end receivers obtained from Intellect NPD Research. Study Results The test results are shown in tables 1, 2, 3, and 4. Table 1 shows the laboratory POF Metric determinations for weak signal, co-channel, first adjacent channel and SNR. Table 1: Laboratory POF Metrics Receiver Under Test Multipath SNR POF (dB) 11.664 13.63 17.76 16.435 Weak Signal POF (dBuV/m) Co-Channel POF D/U Ratio (dB) 1st Adjacent POF D/U Ratio (dB) 18 12

Urban Fast Delco 10 None Sony Clock Radio None Pioneer Home Philips Boombox None
Table 2 shows the levels of analog and IBOC signal power, per sideband, to achieve POF in the four receivers, at six different D/U ratios. The colored dots along the 0 level indicates respective levels of analog first adjacent channel interference that has driven the respective receivers to analog POF. The four colored lines demonstrate the performance of each of the four receivers when IBOC is added to the analog first adjacent interferers. For example, the Pioneer radio reaches its POF when the analog first adjacent channel interferer is 18 dB higher than the desired signal, -18 dB D/U. In the presence of a Hybrid IBOC first adjacent channel interferer,
Ratings data supplied by Arbitron Corporation and are used with their permission. The actual performance of this radio was measured with wideband test equipment while the subjective evaluations were observed through the limited frequency response of the built in speakers.
-8operating at -20 dB per sideband, the Pioneer radio reaches POF at a D/U of -12 dB. Therefore, the Pioneer radio did not fail (reach POF) in the presence of an IBOC first adjacent interferer 12 dB stronger.
Table 2 POF vs. DAB Power Level
DAB PowerLevel, Per Sideband to achieve POF
0 -5 -10 -15 -20 -25 -30 -35 -40 -18

Delco DELCO Analog POF

Analog 1st Adjacent D/U dB
Pioneer(hifi) Pioneer Analog POF Philips(boombox) Philips Analog POF

Sony Clk Sony Analog POF

Table 3 depicts the results of the Dataworld propagation models and Arbitron listener data to show the percentage of potentially impacted listeners verses the family of power ratios used in the lab test program. For example, the graph shows that at an IBOC power level of -23 dB per sideband, on average, 0.64% of an FM stations listeners would be potentially impacted. Table 3 Power Level vs. Listener Impact
Listener Impact vs. IBOC Power per Sideband
2.50% % of Listeners Impacted 2.00% 1.50% 1.00% 0.50% 0.00% -25dB -23dB -21dB -19dB -17dB -15dB Relative Adjacent Digital Power Level Per Sideband
Conclusions The methodology to evaluate POF subjectively to determine the listenable analog signal and the LongleyRice prediction models yield a reasonable interference-free analog service area when compared with actual Arbitron listening data. Furthermore, it serves as a useful tool for modeling IBOCs effects to adjacent channel analog reception. As can be seen in Table 3, at a power level of -23 dB per sideband, 99.36% of an average FM stations listeners will not be impacted by the introduction of IBOC, even with all first adjacent radio stations broadcasting in Hybrid. The study finding that, on average, 99.36% of an FM stations listeners will not be impacted by IBOC introduction is most likely overstated for three reasons. First, IBOC will be implemented over a number of years and the impact to a station will not reflect the study findings until all first adjacents adopt IBOC. Second, future advances in transmission and receiver technology will serve to reduce the listener impact. Finally, over time, listeners are expected to substitute digital receivers for analog receivers further reducing the analog listener impact. As a result, the actual percentage of listeners impacted is likely to be even less than the insignificant number observed in the study. This study provides additional support to the test findings that IBOC can be introduced without meaningful interference to existing analog stations. Refer to Supplement B for additional information regarding the laboratory procedures and methodology used in the study.

- 10 -

Supplement B Additional Information
(A) Receivers iBiquity selected four commercial FM receivers to verify compatibility of FM DAB with existing analog services. They are representative of four major classes of commercially available FM radios: (1) (2) (3) (4) Home HiFi Car Stereo Boombox Clock Radio Pioneer SX-203 Delco 16196204 Philips Magnavox AZ1020 Sony Dream Machine ICF-C390
Where feasible the receivers were modified to accept an RF signal delivered over coaxial cable through a BNC connector. (B) Exciters The host analog FM signals were generated by a Harris THE-1 FM exciter, and the DAB digital sidebands were generated using an iBiquity FM DAB Exciter. (C) Audio Measurement Equipment Recovered analog audio signal-to-noise ratios (SNR) were measured using an Audio Precision System 2 model 2333. (D) Fading Channel Simulation The Home Hi-FI, Boombox, and clock radios are not typically used in a mobile environment, therefore, were tested in a static environment. The car stereo was tested in selective fading using the EIA Urban Fast Rayleigh Multipath Profile1 described below. Measuring audio SNR in the constantly changing multipath environment was accomplished by sampling the receivers SNR 300 times over a 5 minute time period then the readings where averaged to achieve repeatable results. The Electronic Industries Association (EIA) 9-ray Urban Fast Rayleigh multipath fading profile was generated in the laboratory using a Noise/Com MP2700 Multipath Fading Emulator. Both desired and interfering signals were independently faded per this profile. Urban Fast Rayleigh Multipath Profile Ray 9 Delay (microsec) 0.0 0.2 0.5 0.9 1.2 1.4 2.0 2.4 3.0 Doppler (Hz) 5.2314 5.2314 5.2314 5.2314 5.2314 5.2314 5.2314 5.2314 5.2314 Attenuation (dB) 2.0 0.0 3.0 4.0 2.0 0.0 3.0 5.0 10.0
In 1993, the EIA conducted multipath characterization tests in Salt Lake City, and subsequently created four profiles that are descriptive of the multipath environment. The urban fast profile simulates driving at approximately 35 mph through a city street. Impact of National Rollout of IBOC on Analog Radio Listenership

- 11 (E) dBu to dBm Conversion Conversion from dBu into dBm at the receiver input is as follows. Assuming a mid-band carrier frequency of 100 MHz and a dipole antenna of unity gain (due to ground-plane losses, etc.), electric field intensity E (V/m) can be converted to carrier power C (W) at the input to the receiver using

C= E2 Ae 120

In this case, Ae = 0.716 m2 is the effective aperture of the unity-gain dipole antenna. Using this formula, a 54dBu field strength corresponds to a 63.2 dBm carrier power. Likewise, 74-dBu and 34-dBu field strengths correspond to 43.2 dBm and 83.2 dBm carrier powers, respectively. The tests were performed at a desired signal contour of 54 dBu2. (F) Lab Test Setup
Figure 1 Compatibility setup with fading for non-faded cases set the FCS to through Mode.

Desired waveform

Path 1

CD Player

IBOC DAB Exciter

Path 2

Audio Precision

Atten.

Path 3 Attn Path 4 Path 5 Path 6 Path 7
FM Audio Processor FM Analog Exciter Atten.

Car Stereo

Path 8 Path 9
Fading Channel Simulator #1 SCA Modulation Generator CRL Systems SCA-300B
First adjacent waveform IBOC DAB Exciter
Path 2 Path 3 Path 4 Path 5 Path 6 Path 7 Attn Path 8 Path 9 Attn Attn
CD Player Sencore Fm Modulator Atten.
Fading Channel Simulator #2

Figure 1 Lab Test Setup

The 54 dBu signal level was used to place the signal at a sufficient level to mask the effects of receiver front end noise

- 12 -

G. Additional information (1) The baseline DAB signal occupies digital subcarriers in the spectral region +/- 128 to 199 kHz. (2) The baseline FM signal consists of a pilot channel and a main audio channel modulated by processed pop music. (There are no SCAs). (3) The combination of baseline DAB and baseline FM signals produces the baseline hybrid signal. (4) The processed pop music used on the desired signal is The Wallflowers, I Wish I Felt Nothing and for the interferer is The Blues Traveller, But Anyway. The audio processing for the desired and interferring signal is provided by an Orban Optimod 8200 with a Rock-Dense setting. (5) A quiet FM signal is defined as modulation by the pilot only (10%), with no left or right audio inputs. (6) The processed clipped pink noise is simply white gausian noise that has been filtered with a 3 dB/octave rolloff and is subsequently processed using an Orban FM Optimod 2200D. The pink noise was generated by the Audio Precision System Two 2322. The Optimod processing settings were : EQ

30HzHPF = In LOW BASS = +2 dB HF ENHANCE = +2 FULL CONTROL GATE THR = -40 dB AGC = On AGC DRIVE = 10 dB 2B DRIVE = 12 dB REL TIME = +1 dB/sec BASS COUPL = 0 % HF LIMIT = 0.0 CLIPPING = +0.5 FINAL CLIP = 0.0 (7) POF is defined as the point of failure, where the analog signal of interest degrades to a point where the listener would tune to another station. (8) SNR is defined as the signal to noise ratio of the recovered audio where the signal is a 1k tone @ 90% modulation. With the 10% iniection of the stereo pilot the total modulation is 100%.

- 13 -

Supplement C Sample Coverage Map
The sample coverage map for WWMX-FM which was not analyzed in the study. The green area reflects the interference-free analog service area for a Delco car stereo. The interference-free analog service area accounts for lost coverage due to analog co- and adjacent channel interference. The four other colors (yellow, red, brown and blue) reflect coverage areas lost to digital first adjacent interference at various digital power levels. These areas are indicative of the lost coverage areas where actual listener impact was calculated in the study. Coverage losses at successively higher digital power levels include the losses identified at lower power levels. For example, analog coverage loss at the highest power level, 14 dB relative DAB sideband power, includes the areas identified in yellow, red, brown and blue.
Relative Host DAB Sideband Power

Interference Free

-dB dB

-14 dBdB 14

-25 dB

-22 dB

Modification of Sony ICF-SW 7600GR for DRM
This receiver (like the old model SW7600G) has a 2nd IF of 455kHz, so the SATSchneider crystal mixer can be used for downconverting the IF to 12 kHz. As there is not enough space inside the receiver, I put the mixer board, together with a 9V battery and a filter CFW455F in an extra case. I tapped the signal before it enters the IF filter on the board (CF201, see picture below) and connected it to the receivers line out jack (for listening to AM you can still use the earphone jack).
Pic 1: schematics showing CF201
Pic 2: showing CF201 on the PCB For the connection use a shielded wire and connect the shield to ground next to the signal. The other end of the wire can be connected to the line out after removing the capacitors C248 and C250.
Pic 3: showing the line out connection
In the extra box put the filter CFW 455F in font of the mixer board. All you would need for this is: 1 filter CFW 455F 1 crystal mixer board (both available from SAT-Service Schneider) a piece of coax cable and a 9V battery Its a simple modification but it has been working great, even with the telescopic antenna, on some transmissions I achieved up to 27 dB SNR and I especially like it for receiving DRM on medium wave with the built-in ferrite antenna. For the old model SW7600G it is nearly the same, except the point for tapping the IF is CF401 instead of CF201, and the capacitors that need to be removed are C 421 and C423.
If you are interested in the complete service manual of the SW7600GR as a pdf file or for further questions, send me an email to: stoeppler@yahoo.com
Simone Stoeppler stoeppler@yahoo.com www.stoepplernet.de/DRM/

Feb 2004