Medium Wave DRM Field Test Results in Urban and Rural Environments

Transcription

1 Medium Wave DRM Field Test Results in Urban and Rural Environments David Guerra, Gorka Prieto, Igor Fernández, José M. Matías, Pablo Angueira, Juan Luis Ordiales Department of Electronics & Telecommunications Bilbao Engineering College (University of the Basque Country) Alameda Urkijo s/n Bilbao. SPAIN Abstract This paper presents the results of the first Spanish field trial carried out to analyze a DRM (Digital Radio Mondiale) system in the medium-wave band. A 4-kW average power omni directional ground-wave experimental DRM transmission at a frequency of 1359 khz was surveyed by means of a measurement vehicle for fixed and mobile reception. Several radial routes starting from the transmitter site provided rural and suburban behavior features of the system. Urban reception trials were performed in several dense and open streets of Madrid, within the expected coverage area. Field strength threshold values were determined for the tested transmission configurations and compared with the AM ground-wave ITU model predictions. Reliability versus distance from the transmitter is stated in this paper for different transmission configurations and the causes of dropouts for different reception conditions are explained. This analysis took into account subjective quality features of each configuration, providing practical planning parameter values. Keywords Digital AM, digital audio broadcasting, digital radio, DRM, medium wave. I. INTRODUCTION DRM (Digital Radio Mondiale) is the only universal non-proprietary digital radio system for broadcasting audio in the frequency bands below 30 MHz. It is designed to fit in with the existing AM broadcast band frequency plans, based on signals of 9 khz or 10 khz bandwidth. DRM radio system is capable of providing near-fm quality audio and it has the capacity to integrate data and text. In addition, DRM OFDM-based modulation along with guard interval techniques allows the implementation of single frequency networks (SFN) [1]. Several broadcasters, manufactures, research institutions and regulatory bodies integrate nowadays the international DRM consortium formed in In the different consortium working groups, technical, promotional and standardization tasks are carried out. One of the main results of this collaborative effort is the successful launch of commercial emissions in June 2003 by 16 of the leading international broadcasters. The actual, much longer, DRM services schedule can be found in [2]. The achievements of the status of ETSI Standard [3] and IEC International Standard [4], and the ITU procedural change [5] which allows the DRM use of the current analogue assignments in Regions 1 and 3, have also been important milestones.

2 Despite the present availability of several DRM services and the achieved standardizations, an intense DRM analysis work is being performed. There is still a need for network parameter planning values measured on the field and also a need to analyze the behavior of the DRM system under several reception conditions. These tasks are mainly focusing on short-wave ionospheric long-range [6] transmissions [7] whereas there is a lack of testing of medium-wave ground wave propagated DRM signals. This medium frequency band propagation, traditionally of high importance in the AM radio broadcasting in Spain, has been chosen to carry out the DRM measurement campaign presented in this paper. First in this paper the objectives are explained. Then, the measurement campaign is presented by describing the broadcasting network, the reception and measuring systems, the campaign planning and the methodology used in these tests. After the measurement campaign description, the data processing method is detailed and the associated results are described in several graphs and tables. This results section is divided into subsections following the objectives initially proposed. Finally the most relevant conclusions are summarized. II. OBJECTIVES The main objective of this work has been to put under test the reception of DRM audio services in the medium-wave band broadcasted with different DRM transmission modes [3], under different reception conditions. This overall target was divided into the following specific objectives: Study of the performance of different DRM transmission configurations in medium-wave reception. Analysis of the DRM coverage range and field strength threshold in static reception for each analyzed mode comparatively with the corresponding AM range. Study of the mobile DRM reception with different transmission modes and determine on the field the critical reception factors. III. MEASUREMENT CAMPAIGN Broadcasting Network The tested DRM broadcast features are shown in Table 1. The AM transmitter facilities used for the tests are located in Arganda del Rey which is 25 kilometers from downtown Madrid.

3 The DRM standard [3] provides several configurable transmission parameters that allow many different DRM transmission modes with different robustness against noise, multipath and interference. The more robust the mode, the less maximum subjective audio quality can be achieved due to a lower useful bit rate available. In order to evaluate the influence of each parameter the modes in Table 2 were chosen for the tests. The OFDM configurations A and B of the DRM standard are the only ones that fulfill the ITU constraints of medium-wave signal bandwidth in Spain [9]. A broader explanation of the number of carriers, guard interval, protection and interleaving algorithms, and Main Service Channel (MSC), Service Description Channel (SDC) modulations is out of the scope of this article and can be found in detail in [3]. The first column of Table 2 is a reference code included for the sake of briefness when referring to a combination of parameters in the rest of this paper. The DRM modes in Table 2 are ordered from less to more robust as can be deduced from the corresponding useful bit rate which decreases from the top to the bottom of the table. The maximum subjective audio quality provided by each DRM mode depends on the audio coding parameters, which are in turn determined by the available useful bit rate. Thus it is possible to compare the maximum subjective audio quality that the modes in Table 2 can provide. Reception and Measurement System The measurement vehicle was equipped as shown in Figure 1 where three sections are distinguished: The acquisition and distribution system, the measurement system and the control system. The acquisition system was composed of the fully characterized short monopole active antenna R&S HE010 mounted on top of the van with a specific ground plane. The signal received by the antenna was distributed to the measurement equipment by means of a splitter system formed by two power dividers. A first divider fed the received signal on the one hand to a DRM monitoring receiver and on the other hand to the second divider that in turn fed the received signal to a second measurement block. The DRM monitoring receiver was composed of an AR7030 analogue frontend modified to be capable of tuning the received DRM signal an down convert it to a 12 khz IF channel. A SB Extigy sound card sampled this IF signal into a PC which ran the professional Windows

4 based Fraunhofer Software Radio DRM demodulator. The second measurement block was made up of a Vector Signal Analyzer which provided RF spectrum captures and a Field Meter to measure the RF power level field strength in the DRM signal bandwidth. The last section was based on a laptop computer running a control software over a GNU/Linux platform which had the tasks of configuring and controlling the rest of the equipment, calculate some on-the-run statistics, coordinate the time/position/trip measuring, and conveniently store all the captured and calculated data. Ancillary data were provided by a GPS receiver and a trigger system which was generated every wheel turn. The control section was completed with a GSM remote control system for the DRM modulator located in Arganda which allowed the quick on-board changing of the DRM transmission mode via GPRS (General Packet Radio System) data calls between two identical GPRS modules, one located in the van, the other in the transmitter site. This remote control system for the modulator allows the analysis of all the proposed DRM modes with the same propagation conditions i.e. in the same point and consecutively. Measurement Methodology The measurement campaign included measurements in static locations and mobile measurements along routes which were planned following environmental criteria. The distinction made between static and mobile reception is based on the two kinds of radio listeners: the on board ones and the on foot ones and it reflects the most difficult reception conditions featured by mobile reception mainly due to fast signal variations. Radial Routes The first set of measurements was captured along radial routes from the transmitter, as shown in Figure 2, featuring mainly rural and suburban environments and thus, locations where man-made noise level was expected to remain quite low. The sporadic presence of power plants, high-voltage power lines and urban cores in certain locations of these mainly rural-suburban routes is an exception to the aforementioned lack of noise and must be taken into account in the subsequent data analysis. Modes 1, 4 and 5E were measured in 42 static locations and along a total number of 2200 km in order to determine their coverage ranges. The less robust mode 1E was not measured in these routes as its expected coverage range was too low.

5 Madrid Routes The second set of measurements was planned in the urban environment of Madrid as shown in Figure 3. The routes and static locations were selected to represent from wide avenues to the densest urban environment in Madrid. Distance from the transmitter and the direction of the route (radial and circular) with respect to the transmitter were also taken into account. Modes 1, 4 and 5E were tested in 19 static locations, along 470 km overall. Mode 1E, which was not initially considered, was also measured along these routes and points in order to confirm that a code rate less robust than 0.6 (mode 1) could not guarantee service availability neither for static nor mobile reception. Stored Data For both fixed and mobile receptions a set of DRM signal parameters and ancillary data was captured by the measurement system and was conveniently stored in plain text format files. A 400-point 20-kHz bandwidth spectrum centered on 1359 khz was also captured and stored every 5 seconds. These base parameters were stored over each whole route for mobile measurements or over a threeminute interval at each static chosen location and for each tested mode. Using these continuously captured parameters, online statistics were generated every wavelength, which approximately corresponds to eleven wheel turns of the measurement van in mobile measurements. The statistical analysis window was one minute for fixed reception measurements. The on-the-run statistics were stored in one plain text file for each of the measurements performed. As well as the base parameters and the calculated statistics, the sampled input signal down-converted to an intermediate frequency of 12 khz was stored in binary files. The received DRM IQ samples of the whole measurement campaign are thus stored for any subsequent processing. IV. RESULTS Three magnitudes are mainly analyzed in this section: audio quality, received field strength and manmade noise. The audio quality is measured as the quotient of the number of correctly decoded audio frames over the total number of transmitted audio frames in a certain interval for each measured mode. This figure is taken from the statistics so it is calculated every wavelength for mobile reception or every minute for

6 fixed reception. An audio quality threshold value of 98 % is considered to provide no audible dropout i.e. the best subjective audio quality achievable with the DRM mode in use. The field strength is calculated from the signal strength measurements by means of the antenna K-factor and the fully characterized reception chain for the working frequency. The analysis of this parameter was done in terms of median values provided by the stored statistics. Man-made noise is one of the main factors that affect the reception in the MW band. A median noise level value was calculated by analyzing the 5 khz lower and upper bands of the captured DRM spectra and recorded in the statistics files. In order to do so, the 400-point 20-kHz bandwidth spectra centered on 1359 khz captured and stored every 5 seconds by the Vector Spectrum Analyzer were used. The basic noise values were calculated by integrating the captured samples of each spectrum at both sides of the 9 khz DRM spectrum, that is approximately in the 5 khz lower and upper bands of each captured DRM spectrum. Finally the median value for each analysis interval was considered. It should be taken into account that these noise measurements are not intended to give accurate noise ratings but to give a qualitative indication of the man-made noise level in both static and mobile measurements. In this way, this noise estimation was used to detect noisy areas and thus identify possible causes of bad reception rather than to give absolute C/N values. Predicted field strength for an AM transmission was also calculated to study the correspondence between today s available ITU-R recommended AM coverage prediction procedures with the actual measured coverage of the DRM transmission under test. Audio Quality for Static Reception First and as an indication of the general performance of both the DRM system and the transmitting infrastructure, the audio quality measured in static locations over the whole measurement campaign was analyzed. This DRM service analysis consisted of calculating the percentage of locations without noticeable audio errors, i.e., locations where the percentage of correct audio frames received is higher than or equal to 98% of the total number or received audio frames. Previous to the service quality analysis, a data filtering was done in order to identify fixed locations where measurements would not be representative of standard receiving conditions, i. e., points where

7 AM reception would not have been possible either. From all the initially chosen locations in the radial rural-suburban routes ten points were discarded: One point was discarded due to the noise caused by a very close power plant. In another location, 120 km far from the transmitter site, an AM interference at 1341 khz had much higher level than the DRM measured signal. Eight locations far away from the transmitter (more than 160 km) were also eliminated due to the low received signal. Regarding the urban environment, one very dense urban location was rejected also due to the very low field strength received. After this filtering the results shown in Table 5 were obtained. A very good service quality is achieved in most locations with modes 1, 4 and 5E. In a deeper analysis of these previously filtered fixed locations, they were divided into two categories: points located in rural-suburban environments and points located in urban areas. Two studies were done in turn with the rural-suburban locations: one considering the 11 points within a radius of 50 km from the transmitter site, and a second one with the 22 points in the range of 100 km. The results shown in Table 6 lead to a 100 % of the fixed rural-suburban locations receiving a perfect DRM audio quality within a radius of 50 km from the transmitter. Considering a 100-km radius, again a 100 % of locations show good quality values for DRM modes 1, 4 and 5E. The analyzed urban locations show a good reception behavior in more than 94 % of the cases with DRM modes 1, 4 and 5E. Coverage Range in Rural-Suburban Fixed Locations In this section a comparison between the analogue predicted field strength values and the DRM measured ones in rural-suburban points is presented.

8 The predicted variation of the analogue field strength values with distance from the transmitter site was obtained from a simulation performed with the computer version GRWAVE of the ITU-R Recommendation P [11]. The simulation was done with the same features as the transmission under test, assuming a homogeneous curved earth with decreasing refractive index (value selected: 315 N-units) and considering the typical terrain in central Spain having a ground conductivity value of 5 ms/m and a relative permitivity of 6. The transmitter output power value used for the simulation was 10 kw peak power which is the corresponding AM power value to the 4 kw average DRM transmitted power. In order to estimate the corresponding analogue coverage range, the minimum field strength value recommended by the ITU-R [12] for AM coverage (60 dbµv/m) is compared with the prediction. Both the predicted analogue field strength values versus distance from the transmitter and the field strength threshold level for AM are depicted in the three graphs of Figure 4, Figure 5 and Figure 6. The DRM measured field strength values versus distance are represented in Figures 4, 5 and 6 by dots. Each dot corresponds to the median field strength obtained from the statistics files, which in static reception included one median value in each time slot of one minute. A previous filtering of locations is applied and it eliminates two of the above-mentioned points, the one affected by a power plant and the one that was interfered by an AM signal. The chosen locations ensure the study of the DRM field strength behavior with distance only, and leave well known anomalous reception conditions for subsequent specific studies. This allows the calculation of a field strength threshold for DRM static reception by means of differently depict dots corresponding to good audio quality minutes and dots with an audio quality value lower than a 98 % (squared crosses). Using the explained representation for the DRM modes 1, 4 and 5E tested in rural-suburban environment, the graphs of Figures 4, 5 and 6 show that with DRM there is an increase of the coverage area with respect to AM with all tested modes. While predicted analogue field strength provides coverage within a radius of around 60 km from the transmitter, DRM modes 1, 4 and 5E reach ranges higher than 100 km. Another remarkable conclusion is the fact that the ground wave analogue prediction fits well the tendency followed by the field strength measurements.

9 Analyzing the first graph in Figure 4, which represents mode 1, the DRM threshold level is indicated by the dot without perfect audio quality with the highest field strength value that can be considered 45 dbµv/m. Every measured minute with a median field strength value higher than 45 dbµv/m had an audio quality value higher than or equal to a 98 % of error free received audio frames. The mode 4 in Figure 5 did not present a relevant decrease in the threshold. In Figure 6 the mode 5E showed an excellent audio quality for values higher than 38 dbµv/m, which remarkably improves the threshold level achieved by modes 1 and 4. Mobile Reception in Rural-Suburban Environments The five radial rural-suburban routes are discussed in this section. The statistics in mobile measurements were calculated and stored continuously over a wavelength. Considering DRM mode 1, the least protected one of the rural-suburban tests, the distribution shape of the reception dropouts led to distinguish two circular areas around the transmitter. A perfect audio quality mobile reception in rural and suburban environments was ensured for distances up to 35 km from the transmitter. Very few audio dropouts were present within the next circular zone which reached distances of 75 km from the transmitter site. The audio dropouts within this area were due to power lines, power plants and tunnels. The received audio quality was improved by selecting the more robust mode 5E. The results for this mode are shown in the southwest route mobile reception profiles presented in Figures 7, 8, and 9 for modes 5E, 1 and 4 respectively. Three magnitudes are depicted in each graph: the audio quality values in percentage calculated every travelled wavelength (statistics in mobile reception), the corresponding median field strength and the median estimated noise values for each represented wavelength section of the route. On the x-axis the distance from the transmitter site is represented. The three graphs show high field strength values up to a distance of 30/35 km, medium values from 30/35 km to 60/65 km and low ones from 60/65 km on. Only three significant dropouts disturbed mode 5E reception: one at a distance a bit higher than 30 km and other two at the end of the route. All of them were caused by well known long tunnels. This conclusion is driven by the fact that they were accompanied by a noticeable fading of the received field strength. Along the route there were four sections with different lengths in which man-made noise level was high but none of them showed audible reception artifacts with the DRM mode 5E. The first three

10 ones were due to a high voltage power line, a voltage transformation centre and a power plant. The wide fourth section was caused by the dense industrial environment surrounding the city of Toledo. Despite some little differences due to different measuring days, field strength and noise behaviors are very similar in Figure 7 (mode 5E) and Figure 8 (mode 1) as the same route is represented. As expected, the mode 5E leads to a wider error free coverage but both modes 5E and 1 were useless for inside-tunnel reception. Mode 1 was more sensitive to high noise levels and high field strength variations. However the increase in reliability of mode 5E might not be considered enough taking into account the subjective audio quality decrease caused by the lower available bit rate. Mode 4 improved slightly the DRM service availability against low-intensity isolated impairments such as little field strength variations mainly due to bridges, but it did not show an improved performance against high man-made noise levels. This kind of reception disturbances caused short audible drop outs to mode 1 for short (35 to 65 km in the graph) and long (65 km on) distances from the transmitter (i.e. medium to low signal levels). Moreover, due to the long interleaver used in mode 4, long-lasting or highintensity reception difficulties, and even little impairments when receiving low field strength signal levels caused longer dropouts to mode 4 than the ones observed for mode 1. As a conclusion DRM mode 4 did not improve significantly the performance for mobile reception. Mode 5E did increase this performance but at the cost of noticeable loss of subjective audio quality related to its low bitrate. The critical situations for ground-wave DRM reception in the medium wave band were tunnels and high-level man-made noise sources. V. CONCLUSIONS DRM offers a good opportunity to broadcast digital quality audio services in the medium wave band. While keeping the AM spectrum international standard restrictions, it offers good subjective audio quality, very close to FM, and significantly higher coverage ranges than the ones predicted for AM with the same transmission power. This advantages are achieved by means of chosing the right DRM transmission mode what implies the configuration of several parameter values.

11 The DRM transmission mode referenced as mode 1 in this article (OFDM configuration A, code rate 0.6, short interleaving, MSC 64 QAM, SDC 16 QAM, bit rate 23.5 kbps) ensured all the former specifications. Neither OFDM mode B nor long interleaving increased the DRM medium-wave ground wave coverage area in a significant way. This was true where the main reception impairments were high field strength fading (due mainly to tunnels) and high man-made noise levels. The overall reliability improvement of mode 4 (OFDM configuration B, code rate 0.5, long interleaving, MSC 64 QAM, SDC 16 QAM, bit rate 15.2 kbps) over mode 1 lies upon its more robust code rate. On the one hand, mode B is intended to deal with ionospheric propagation conditions which lead to time and frequency dispersion. Groundwave propagation of the transmission under test is not characterized by high time nor frequency dispersion values so mode B features are not very helpful in this case. On the other hand, long interleaving length can be useful in DRM medium-wave broadcasting as it eliminates short audible dropouts caused by low-intensity impairments. This long interleaver slightly increases the audio availability in the medium ranges of the coverage area but it implies a slower recovery from more important impairments. In the latter case of these severe impairments, the effect of this long interleaver is a considerable decrease of the audio availability in the vicinity of the coverage area limits. The remarkable robustness demonstrated by DRM mode referenced with the code 5E (OFDM configuration A, code rate 0.5, short interleaving, MSC 16 QAM, SDC 4 QAM, bit rate 13.0 kbps) confirmed the importance of appropriate code rate values. Mode 5E also included the two more robust 16QAM and QPSK which also helped in increasing the reliability. The above mentioned two parameters (code rate and constellation) have a noticeable influence on the reliability of DRM medium-wave ground wave reception. Regarding the code rate, a value of 0.78 proved to be insufficient as mode referenced as 1E (OFDM configuration A, code rate 0.78, short interleaving, MSC 64 QAM, SDC 16 QAM, bit rate 30.8 kbps) didn t behave in the same stable way as

12 the others. A code rate value of 0.5 (mode 4) worked slightly better than a value of 0.6 (mode 1) but a value of 0.5 seems to be excessive for the code rate in DRM medium-wave ground wave reception. The constellation choice had a very critical influence on the DRM reception reliability as shown in the case of mode 5E. On the other hand the robust QAM constellations used by mode 5E implied a considerably low bit rate which reduced the achievable subjective audio quality. As a conclusion, MSC 16 QAM and SDC 4 QAM might be considered too robust constellations for DRM medium-wave ground wave reception. With regard to the DRM 4-kW average power medium-wave tested transmissions: All the DRM modes tested provided quasi error free static reception in rural-suburban environments within a 100 km radius from the transmitter site. DRM modes with reference codes 1, 4 and 5E proved a quasi error free reception in more than a 94 % of static urban locations. Mobile reception is excellent within a 35 km from the transmitter and very good over a distance of 75 km in rural-suburban environments. Further tests are being carried out to continue the process of planning parameter tuning. VI. ACKNOWLEDGMENTS This paper is a result of the collaboration of Radio Nacional de España, VIMESA, TELEFUNKEN SENDERSYSTEMS and the University of the Basque Country. It has been partly financed by the Spanish Ministry of Science and Technology under the MCYT project with code number TIC and by the University of the Basque Country, UPV-EHU. VII. REFERENCES [1] [2] [3] ETSI, ES V2.1.1, DRM ETSI Standard, European Telecommunications Standards Institute, 2004

13 [4] IEC, International Standard , Digital Radio Mondiale (DRM) Part 1: System specification, International Electrotechnical Commission, 2003 [5] ITU-R, Recommendation BS , System for digital sound broadcasting in the broadcasting bands below 30 MHz, International Telecommunication Union, 2002 [6] Hofmann, F.; Hansen, C.; Schafer, W. Digital Radio Mondiale (DRM) digital sound broadcasting in the AM bands. IEEE Transactions on Broadcasting,Volume: 49, Issue: 3 pp , Sept [7] Freeman Roger L., Radio System Design for Telecommunications, 2 nd ed., Wiley-Interscience, 1997 [8] Giefer, A., Digital Radio Mondiale longterm tests results, BBC R&D White Paper WHP032, 2002 [9] ITU-R, Recommendation BS , Planning parameter for digital sound broadcasting at frequencies below 30 MHz, International Telecommunication Union, 2003 [10] Dietz, M., CT-aacPlus a state-of-the-art audio coding system, EBU Technical Review, 2002 [11] ITU-R, Recommendation P , Ground-wave propagation curves for frequencies between 10 khz and 30 MHz, International Telecommunication Union, 1992 [12] ITU-R, Recommendation BS. 703, Characteristics of AM sound broadcasting reference receivers for planning purposes, International Telecommunication Union, 1996

14 FIGURES AND TABLES

15 Table 1. Transmission data Transmitter station Broadcaster Modulator Arganda del Rey (Madrid), Spain Radio Nacional de España (RNE) TELEFUNKEN DMOD2 Amplifier TELEFUNKEN TRAM 10 Transmission frequency Output power Nominal bandwidth Antenna system 1359 khz 4 kw average DRM power 9 khz 1.1 dbi gain vertical monopole Transmission timetable 08:30 14:00 Table 2. DRM modes tested REF. OFDM Code rate 1E A 0.78 Short (S) 1 A 0.60 Short (S) 4 B 0.50 Long (L) 5E A 0.5 Short (S) Interleave MSC SDC 64 QAM 64 QAM 64 QAM 16 QAM 16 QAM 16 QAM 16 QAM 4 QAM Bit rate: kbps Table 3. Subjective audio quality provided by the tested DRM modes [3] [10] Mode Best audio coding supported Sampling frequency Provided subjective audio quality 1E AAC with SBR Parametric stereo 24 khz FM-like audio High quality music 1 AAC with SBR Parametric stereo 24 khz Close to FM audio Good quality music 4 AAC with SBR Mono 24 khz Better than AM audio Good quality voice 5E CELP with SBR Mono only voice 16 khz AM-like voice Good quality voice

16 Table 4. Base measurements Provider Type Parameter Fs Receiver signal strength RF Delay spread DRM receiver Field strength meter GPS receiver IF Baseband RF Ancillary data Doppler spread QAM constellation measurements Corrupted audio frames distribution Signal strength Time GPS positioning Speed 400 ms 400 ms 1 s Table 5. Service availability in static measurements Mode 1E Mode 1 Mode 4 Mode 5E % Of points with audio quality 98 % Number of locations Table 6. Service availability in static measurements in rural-suburban and urban environments (% = Percentage of locations with audio quality 98 %) % Rural Suburban 50 km % Rural Suburban 100 km Mode 1E Mode 1 Mode 4 Mode 5E % Urban

17 Antenna / Distribution Measurement Equipment Control Antenna Field Meter GPS GPRS Atennuator DRM Professional Rx AOR Sound Card 10BaseT Splitter Vector Signal Analyzer Wheel Trigger Figure 1. Reception and measurement system

18 Figure 2. Rural and suburban routes.

19 Figure 3. Urban routes.

20 Field Strength (dbuv/m) Field Strength vs Distance: Ground Wave DRM Mode 1 Minutes with Audio Quality >= 98% AM ITU Predicted Field Strength AM ITU Minimum Field Strength Threshold Minutes with Audio Quality < 98% Distance from tx (km) Figure 4. Simulated AM and measured DRM field strength variations with distance from the transmitter (DRM mode 1). Static reception.

21 Field Strength (dbuv/m) Field Strength vs Distance: Ground Wave DRM Mode 4 Minutes with Audio Quality >= 98% AM ITU Predicted Field Strength AM ITU Minimum Field Strength Threshold Minutes with Audio Quality < 98% Distance from tx (km) Figure 5. Simulated AM and measured DRM field strength variations with distance from the transmitter (DRM mode 4) Static reception.

22 Field Strength (dbuv/m) Field Strength vs Distance: Ground Wave DRM Mode 5E Minutes with Audio Quality >= 98% AM ITU Predicted Field Strength AM ITU Minimum Field Strength Threshold Minutes with Audio Quality < 98% Distance from tx (km) Figure 6. Simulated AM and measured DRM field strength variations with distance from the transmitter (DRM mode 5E). Static reception.

23 SouthWest Route DRM Mode 5E 100 Audio Quality (%) Field Strength (dbuv/m) Noise (dbuv) 100 Correct Audio Frames in a Wavelength (%) Distance (km) Field Strength (dbuv/m) Figure 7. Received field strength and audio dropouts along one of the routes (DRM mode 5E). Mobile reception.

24 SouthWest Route DRM Mode Audio Quality (%) Field Strength (dbuv/m) Noise (dbuv) 100 Correct Audio Frames in a Wavelength (%) Distance (km) Field Strength (dbuv/m) Figure 8. Received field strength and audio dropouts along one of the routes (DRM mode 1). Mobile reception.

25 SouthWest Route DRM Mode Audio Quality (%) Field Strength (dbuv/m) Noise (dbuv) 100 Correct Audio Frames in a Wavelength (%) Field Strength (dbuv/m) Distance (km) Figure 9. Received field strength and audio dropouts along one of the routes (DRM mode 4). Mobile reception.