Coverage aspects of digital terrestrial television ... · 2. Digital terrestrial television...

19EBU Technical Review Winter 1996Weck

Coverage aspects of digitalterrestrial television broadcasting

C. Weck (IRT)

The specification for digital terrestrialtelevision, DVB-T, offers a wide rangeof potential applications: single trans-mitter and single-frequency networks,prohibited channel operation, port-able reception, hierarchical trans-mission, etc. The network operatorcan select technical parameters suchas the number of OFDM carriers, thelength of guard interval, the degree oferror protection and the modulationmethod. The last two parameters inparticular allow the operator to reacha compromise between the numberof programmes carried and theirtransmission reliability. This raisesthe question of which results, withregard to coverage, need to beachieved in each case.

The main applications have beenstudied at the IRT using Monte Carlosimulations of regular networksituations. The results are docu-mented here in the form of predic-tions for the coverage probabilities.The reasons for the choice of para-meters for the terrestrial specificationare dealt with specifically. Ways torestrict the basically very wide choiceof system states (modes) can also bederived from these simulations.

1. Introduction

The intention is to make terrestrial broadcasting oftelevision programmes a more attractive proposi-tion to meet competition from cable and satellitetransmission. The new specification for digitalterrestrial television, DVB-T [1], will make it pos-sible to provide television services which can holdtheir own, even when digital transmission via sat-ellite and cable is introduced.

The chief advantages of terrestrial broadcastingare to be found in regional and local services(pictures, sound and data). A further plus pointfor digital systems is the marked improvement inportable reception when ordinary antennas areused. It is also possible with digital systems totransmit within a standard (7 or 8 MHz) televisionchannel either (i) more programmes concurrentlyor (ii) programmes with a higher image resolution(EDTV, HDTV).

As far as coverage planning is concerned, thereare important advantages in the design of thetransmitter network structure and the energy bal-ance. An important aspect in this respect is thepossibility of using single-frequency networks(SFNs) combined with the possible use of “gap-fillers” (local relay transmitters) on the same car-rier frequency in regions where the service is poor.With a digital system it is also possible to usedomestic gap-fillers, at the same transmission fre-quency as the main transmitter, inside houses orflats where the signal attenuation is frequentlyhigh, in order to supply adequate field strengthsfor portable appliances.

Original language: GermanManuscript received 25/9/96.

20 EBU Technical Review Winter 1996Weck

2. Digital terrestrial televisionbroadcasting

Because of their high data-rate (ITU-R Recom-mendation BT.601 [2] specifies a rate of 216 Mbit/sin the studio), digital video signals can only betransmitted via existing television channels afterthe data-rate has been effectively reduced. TheMPEG-2 source-coding method envisaged forDVB achieves a data-rate of 5-6 Mbit/s for digitalvideo signals of a quality at least comparable withPAL. This data-rate is assumed for a standard-definition programme (SDTV) although an ade-quate image quality (comparable to that of theVHS tape format) could be achieved with lowerdata-rates, depending on the picture content; higherdata-rates (8-11 Mbit/s) are also of interest forhighly-critical image presentations (e.g. sport).

There are still bottlenecks with the available trans-mission channels for DVB-T, despite the largedata reduction achieved by source coding. Thefrequency resources in the VHF and UHF rangesare already largely exhausted by analogue ser-vices (e.g. in Germany alone, there are currently578 analogue television transmitters and 8707gap-fillers). The situation is exacerbated by the

Abbreviations

agl Above ground level

C/I Carrier-to-interference ratio

C/N Carrier-to-noise ratio

CCETT Centre Commun d’Etudes de Télé-diffusion et de Télécommunications(France)

COFDM Coded orthogonal frequencydivision multiplex

DLR Deutsche Forschungsanstalt fürLuftund Raumfahrt (Germanaerospace research establishment)

DVB Digital Video Broadcasting

EDTV Enhanced (extended) definitiontelevision

FFT Fast Fourier transform

HDTV High-definition television

MPEG Moving Picture Experts Group

OFDM Orthogonal frequency divisionmultiplex

PSK Phase shift keying

QAM Quadradure amplitude modulation

SDTV Standard-definition television

SFN Single-frequency network

fact that the introduction of DVB-T will generateinstant demand for several digital television pro-grammes or additional multimedia services.

A transmission method for digital video signalswith a high spectral efficiency is therefore re-quired which, at the same time, has a high resist-ance to the many different types of interference onthe channel (e.g. interference due to multipath re-ception, signal interference from other radio orbroadcasting services, high signal attenuation inthe case of portable reception in buildings, etc.).

In the development and definition phase of theDVB-T specification, the advantages of theCOFDM transmission method [3][4] outweighedthose of single-carrier methods in terms of broad-casting requirements (as was also the case withDAB). The crucial factor in favour of COFDM isthe ability to set up single-frequency networkswhich offer network planners a higher networkefficiency. However, a digital system also allowsgreater planning leeway for regional broadcastingservices in the existing transmitter network (e.g.possible use of prohibited channels such as adja-cent and image channels) because, with an ap-propriate choice of parameters, networks can alsobe planned with lower signal-to-noise ratios.

3. Coverage probability

The abrupt degradation of a DVB system at thecoverage periphery makes it easy to establishwhether or not reception is possible at a location.A single location can be deemed to be covered ifthe required signal-to-noise ratio is achieved for99 % of the time. On the other hand, it is notpracticable to demand a similar coverage for alllocations in a “sub-area” (e.g. 100 m x 100 m).The coverage criterion for a sub-area is also in-fluenced by the extent to which the signal-to-noiseratio varies as a function of the location (“log-normal fading” component). A high probability ofinterference-free reception within a particular sub-area results in high costs for the network operator(higher transmission power, additional main trans-mitters and gap-fillers). In an initial approach bythe EBU, the coverage was described as “good” ifat least 95 % of the sub-area is served for 99 % ofthe time and as “acceptable” if at least 70 % of thesub-area is served for 99 % of the time [5].

Another quantity of interest for the planning ofDVB-T is the global coverage probability withinthe target service area, which is derived from thecoverage probability of all the sub-areas con-tained within the target area.


4. Reception situations

A distinction is drawn between two particular re-ception conditions for digital terrestrial televisionbroadcasting: stationary and portable reception.In both cases, typical conditions found in practiceare assumed, not the worst conditions; for exam-ple, it is assumed that the position of an antennacan be optimized over a range of ± 50 cm.

A directional antenna 10 m above ground level(agl) is assumed for stationary reception. For por-table reception, a non-directional antenna on top ofor inside the equipment – positioned 1.5 m abovethe ground or above floor level – is assumed. Theleast favourable case, on the ground floor of abuilding, is generally considered separately.

The data in the DVB-T specification which relatesto the efficiency of the system are based on simula-tion results using a Rayleigh channel model forportable reception (omnidirectional reception in amultipath channel) and a Rice channel model forstationary reception with a directional antenna. Inthe latter case, a Rice factor of K = 10 dB is app-lied, i.e. the received field strength of the directsignal is 10 dB higher than that of all the reflectedsignal components. The DVB-T specification thususes known values for the theoretically requiredsignal-to-noise ratio for various operating modes.However, during coverage analyses an imple-mentation loss of at least 3-4 dB must be expected.

5. Simulation model forcoverage analyzes

The IRT’s contribution to the DVB-T systemspecification included an assessment of the cover-age probabilities which result when different sys-tem parameters are applied. For this purpose, asimulation model was created to analyze cover-age, thus allowing a fundamental study to be madeof reception conditions in an SFN. The resultsdescribed in this article do not therefore refer to aspecific planning study of DVB-T.

The propagation of a broadcast signal from asingle transmitter to a receiving location is consid-ered as a statistical process, which takes accountof the various propagation conditions that have tobe considered at a specific distance from the trans-mitter. The parameters for this statistical processare based on the ITU-R propagation curves [6]which represent the results of numerous fieldmeasurements. An average unevenness has beenapplied as the parameter for the topology (e.g.

50 m); in other words, the actual terrain of thecoverage areas is not taken into account. It is thuspossible to simulate the main reception situationswith a “Monte Carlo” simulation (explicit study ofrandom events). First and foremost, this allowsthe fundamental characteristics of the newtechnology to be identified, whereas practical usecalls for the planning of coverage for specificcases. Simulations such as these are based onpropagation models whose parameters are notclearly established. The influence of critical prop-agation parameters therefore have to be taken intoaccount. In the case of single-frequency net-works, the statistical propagation processes fromall the active transmitters are superimposed.

The fading component at a specific distance fromthe transmitter exhibits a log-normal distribution.When investigating the coverage probabilities bymeans of Monte Carlo simulations, the basic stan-dard deviation (σ) for the fading component mustbe selected carefully. The use of a mean variationof σ = 9.5 dB is recommended for the planning ofanalogue services in the UHF range. However,propagation measurements with digital signals ina bandwidth of 1.5-8 MHz show a significantlylower variation (e.g. 5 dB) in the measured fieldstrengths across the location [7][8]. This meansthat the value on which local planning calcula-tions are based could also be lower. The variationof the field strength prediction error also plays acrucial role in the calculation of coverage proba-bilities in an SFN. This quantity also has a log-normal distribution but, with an average value of14 dB, it is significantly higher [9] than the varia-tion measured in an area, because it is highly de-pendent on the terrain. In the IRT studies, asdescribed in this article, a value for σ of approxi-mately 9 dB was applied, taking into account boththese quantities, i.e. the measured variation of thefield strength and the variation of the fieldstrength prediction error.

The signals from different transmitters in a single-frequency network arrive at the receiving locationwith different delays, to form a sum signal whichis a function of the individual path delay differ-ences. The signal components which are withinthe guard interval make a constructive contri-bution to the signal-to-noise ratio (C/N) or signal-to-interference ratio (C/I) which can be deter-mined in the case of COFDM simply by addingthe powers of the signal components. All signalcomponents received outside the guard intervaldisturb the usable signal, an effect known as theinherent interference of a single-frequency net-work. The impact of inherent interference on thecoverage probability depends on the amplitude of


all the signal components involved and the lengthof time they exceed the guard interval.

The computation model used for evaluating inter-ference by signal components outside the guardinterval was the one on which the planning forDAB was also based [10]. However, more recentstudies by RAI [11] indicate that this is toooptimistic for DVB-T. In contrast to DAB, achannel estimate for the coherent demodulation isrequired for DVB-T which is impaired by echoesoutside the guard interval, especially with64-QAM. This is, however, of lesser importancefor the comparative studies of the DVB-T systemvariants presented in this article, whereas the ab-solute results will require further investigation.

The ITU-R propagation curves for 1 % of the timemust be used for the most distant transmitters in asingle-frequency network; they must thus beconsidered as interfering transmitters [12]. In thelocal area around an SFN transmitter, the curvesfor 50 % of the time have to be taken into accountbecause this corresponds to the less favourablereception conditions. However, in view of therelatively small distances to the nearest transmit-ter in an SFN (which results in little difference be-tween the propagation curves for 50 % and 99 %

of the time), the curves for 1 % of the time can beused for all transmitters in an SFN.

6. Transmitter networksstudied

Two transmitter network structures are of particu-lar interest with regard to local and regionalDVB-T services:

– an extensive SFN with a large number of trans-mitters;

– a minimal SFN with only two transmitters,which is also applicable to the reception situa-tion when gap-filling transmitters are used.

The SFN on which the results presented here arebased covers an area of about 400 km x 240 km(Fig. 1). It comprises 33 transmitters arranged ina regular structure, with a transmitter spacing of60 km which corresponds to the average spacingbetween transmitters of present-day TV networks.

All transmitters broadcast the signal synchro-nously at the same power. A standard transmitterheight of 150 m agl is selected, and the height ofthe receiving antenna is taken as 10 m agl for

Figure 1Structure of theanalyzed regularSFN: 33 transmitterswith an interspacingof 60 km.


Duration of guard interval (T g ) at theuseful symbol time (T u ) of:

Resulting transmission capacity(%) at the useful symbol time of:

Tu = 896 µs Tu = 224 µs Tu = 896 µs Tu = 224 µs

1/4 224 µs 100 %

1/8 112 µs 111 %

1/16 56 µs 1/4 56 µs 118 % 100 %

1/32 28 µs 1/8 28 µs 122 % 111 %

1/16 14 µs 118 %

1/32 7 µs 122 %

directional reception [13] and 2 m for omni-directional (portable) reception. The coverageprobabilities were investigated for both direction-al and omnidirectional reception in a rectangulararea (area A1), measuring 60 x 52 km, which isrepresentative of a large-area SFN. In order tosave computing time, reception conditions at apoint (P2) – about 25 km from one of the trans-mitters – was also considered as an alternativebecause experience has shown that the resultsobtained there correspond approximately to thearea average within the network.

The minimal SFN considered in the IRT studiesconsists of just two transmitters, spaced 60 kmapart, which emit identical power from the sameantenna height. The coordinates (in km) of thetwo transmitters are: Transmitter 1: (0, 0); Trans-mitter 2: (60, 0). The coverage probabilities fordirectional and omnidirectional reception wereanalysed in a rectangular area, 150 km by 60 km,around these two single-frequency transmitters.

7. Influence of the DVB-Ttransmission parameters

The DVB-T specification has been formulated soflexibly that ideal parameters for transmission canbe selected for each different application, be it aspecific source data-rate (e.g. SDTV or HDTV) ora specific transmission reliability (error protectionor modulation for portable or stationary recep-tion), a particular transmitter network structure(single-transmitter or SFN) or compliance withdifferent signal-to-noise ratios. This allows thebest possible use to be made of frequency re-sources when introducing DVB-T.

8. Length of the guard interval

The standardization bodies working on DVB-T inEurope called for the option of using SFNs so thatfrequency resources could be used efficiently.The COFDM modulation scheme was therefore

based on 8K-FFT which allows a guard interval ofTg = 224 µs at 25 % overhead and an effectivesymbol time of Tu = 896 µs. This means that thepermissible signal delay times are outside the sig-nal delay between adjacent transmitters, whenthese are situated less than 67 km apart.

Choosing a large guard interval costs transmissioncapacity (e.g. 25 %). Also, the cost of the hard-ware at the receiving end is increased because ofthe large OFDM symbol times (8K-FFT). Com-promise solutions for the length of the guard inter-val are either (i) at the expense of frequency econ-omy from the network planning aspect (lowcoverage probabilities in the SFN) or (ii) at the ex-pense of transmission efficiency (low data-rate).

A long guard interval will not be necessary for allcoverage concepts of a future digital terrestrial tele-vision system. Local or regional services can beimplemented with considerably shorter guardintervals than large-area SFNs. Whereas it mightbe possible to achieve an increase in frequencyeconomy with a large guard interval in an SFN, areduction in transmission efficiency by the amountof the guard interval would have to be accepted forlocal services with conventional network planning.

The DVB-T specification therefore makes provi-sion, firstly, for different lengths of the guard inter-val (1/4, 1/8, 1/16 and 1/32 of the effective symboltime) and, secondly, for two different symbol timesTu = 896 µs and Tu = 224 µs (Table 1). The shortersymbol time means that there is a correspondinglysmaller number of carriers and that a 2K-FFT sys-tem is used instead of 8K-FFT. The DVB-T speci-fication thus allows for six different values for theguard interval in the range from 7 to 224 µs.

Flexible setting of the guard interval provides ameans of optimizing the transmission efficiencyaccording to the network structure. Although thefrequency efficiency is reduced as the guard inter-val increases, this case basically only arises whenvery high frequency economy is achieved throughthe use of a large-area SFN.

Table 1Flexible settings of

the DVB-T guardinterval.


9. Influence of the guardinterval on coverageprobability

In a wide-area SFN, the coverage probabilities de-crease when a shorter guard interval is applied (seeFig. 2). Of the four guard intervals provided for inthe DVB-T 8K-FTT specification, three of them(28, 112 and 224 µs) are shown on this graph, aswell as two non-specified cases (0 and 448 µs) forillustration purposes. These results indicate clearlythat the longest guard interval selected for theDVB-T 8K system (i.e. 224 µs) was a compromisein order to keep the symbol overhead within ac-ceptable limits (25% of the active symbol time,896 µs). Fig. 2 shows that the 8K system provideseven better coverage probability if the guard inter-val is set to 448 µs: despite the higher symbol over-head (50% in this case), there may exist some ap-plications of very large SFNs where the frequencyefficiency is still superior to that of conventionally-planned networks.

Fig. 2 also shows that the relatively large activesymbol time of the 8K-FFT system (896 µs) resultsin a certain coverage probability even with noguard interval. This is due to the fact that the inter-fering echoes largely decay within the first quarterof the symbol.

In a 2K-FFT system, signal delays that exceed theguard interval are very much more conspicuous

due to the considerably shorter usable symboltime of 224 µs. The signal delays occurring in thewide-area SFN considered here, which has atransmitter spacing of 60 km, are significantlyhigher than 56 µs. Consequently, increasing theguard interval from 1/32 to 1/4 of the symbol timeresults in hardly any positive effect (Fig. 3). Inthat case, a coverage of 90 % of the locationswould only be possible for a single-programmeservice based on 4-PSK.

10. Comparison of coverageprobabilities in a two-transmitter SFN

Assuming that DVB-T will concentrate particu-larly on local and regional services, it becomesnecessary to decide whether an 8K-FFT system isonly required in large single-frequency networksor whether an 8K-FFT system and a guard intervalof 1/4 would also provide coverage advantages forsmall transmitter networks. For this purpose, theIRT studied an SFN with just two transmittersspaced 60 km apart and with the same antennaheight, radiating the same transmission power.

The coverage probabilities were investigated fordirectional and omnidirectional reception in a rec-tangular area, 150 km by 60 km, around the twosingle-frequency transmitters. The result ob-tained across the area indicates that the influenceof inherent interference of the transmitters in the8K system is more than 17 dB lower than in the

Figure 28K-FFT system witha useful symbol timeof 896 µs: influenceof the length of guardinterval; directionalreceiving antenna.


case of the 2K system. Fig. 4 shows the coverageprobabilities across the receiving location when asignal with four television programmes is broad-cast (equivalent to a C/I of 23 dB for a 64-QAMsystem). One half of the reception area, which issymmetrical for 30 km, is shown and the resultsfor the 2K and 8K systems are compared.

The results for the 2K-FFT system shown in Fig.4 (upper) indicate that considerable impairment ofsignals is to be expected with omnidirectional re-ception, especially between the two transmitterswhere slight path delay differences occur. Withdirectional reception, the attenuation at the rear ofthe receiving antenna can noticeably reduce theeffect of the interfering signal. However, an ex-ceptionally steep drop in the coverage probabilitycan be seen in the area outside the two transmit-ters, in the case of both omnidirectional and direc-tional reception. The results for 8K-FFT in Fig. 4(lower) illustrate the clear superiority of the 8Ksystem, also – or particularly – in a small SFNsuch as the one used in the IRT simulation.

11. Modulation parameters

Provision is made in the DVB-T specification forthree different phase/amplitude constellations,64-QAM, 16-QAM and 4-PSK , in order to meetthe different requirements in terms of spectralefficiency and the reliability of the broadcast ser-vice. The reduction in the data-rate by a factor of

about 2 from stage to stage is matched by a simul-taneous increase in the reliability.

Since the bandwidth of the transmitted COFDMsignal has been based on an 8 MHz-wide televi-sion channel, it is possible with 4-PSK to transmitexactly one digital programme with a data-rate ofapproximately 7 Mbit/s. This means that an ana-logue television service can be replaced by a digi-tal service, thereby allowing a considerable im-provement in the reception conditions for portablereceivers. Conversion to a digital terrestrial tele-vision system becomes even more attractive as thenumber of programmes offered increases in rela-tion to the analogue services. This speaks infavour of using 64-QAM with which, for exam-ple, four programmes can be accommodated inone 8 MHz channel. This form of modulationagain requires significantly higher signal-to-noiseratios than 4-PSK.

Apart from the correspondingly higher effectivefield strength required at the receiving location,this also means greater protection ratios withrespect to adjacent services in the same or theadjacent frequency range. The availability ofthese channels is, however, extremely limited.COFDM with 16-QAM offers a compromisewhereby there is room for two to three pro-grammes in an 8 MHz channel and there is lowersensitivity to interference from other transmit-ters. This mode permits the use of “prohibited”channels for local services, at the same time

Figure 32K-FFT system witha useful symbol timeof 224 µs: influence

of the length of guardinterval; directionalreceiving antenna.


allowing the conventional planning of a DVB-Ttransmitter network with less stringent require-ments in terms of the protection ratio.

Fig. 5 shows the coverage probability at point P2of the single-frequency network under consider-ation (COFDM with 896 µs effective symboltime and 224 µs guard interval) for both station-ary and portable reception. Ranges for C/I valuesare marked in which one, two or four pro-grammes, each at approximately 6 Mbit/s, can bereceived depending on the modulation method,or more programmes if the source data-rate and/or error protection is reduced. Fig. 5 shows thatabout 90 % of locations can be supplied with fouror more programmes (64-QAM) in the case of di-rectional reception (unbroken curve). On theother hand, only about two programmes(16-QAM) can be broadcast for portable recep-tion with an omnidirectional antenna if 90 % ofthe locations are to be served (broken curve). Forportable reception of four programmes, the cov-erage probability is reduced to as little as 25 %(in the C/I range, 25-28 dB).

12. Hierarchical modes

When COFDM carriers are modulated with4-PSK, the information (2 bits each) is transmittedin the phase quadrant. With 16- or 64-QAM, an

additional 2 or 4 bits are transmitted in each phasequadrant, by spacing possible amplitudes andphase values closer together. An obvious solu-tion is therefore to organize the modulation intwo steps. In the first step, the phase quadrant isselected on the basis of two bits of a (channel-encoded) datastream, HP (“high priority”). Thenthe momentary phase and amplitude constellationis formed inside the preselected quadrant, de-pending on the 2 or 4 remaining bits of another(channel-encoded) datastream, LP (“low prior-ity”). This form of modulation, known as hier-archical modulation, enables 4-PSK to be inte-grated in the constellation diagram of a 16- or64-QAM system. Fig. 6 shows the case for16-QAM. The HP datastream can be demodu-lated independently by a single evaluation ofthe quadrant used.

The separate transmission of the two data-streams, HP and LP, each with their own errorprotection, means that transmission of the inte-grated 4-PSK is less susceptible to interferencethan is the case with non-hierarchical 16- or64-QAM. It also results in a higher coverageprobability for the HP datastream, which is ad-vantageous for a broadcasting system geared toportable reception. However, a slightly highersignal-to-noise ratio is required for the LP data-stream compared with normal 16- or 64-QAM,because for the purposes of error correction the

Figure 4Comparison of

coverage probabili-ties of the 2K and 8K

systems in a half-area of 75 x 60 kmaround one of two

single-frequencytransmitters spaced60 km apart, based

on a signal with fourprogrammes (23 dB).

(Upper):Omnidirectional

receiving antenna(Lower):

Directionalreceiving antenna.

Figure 5Coverage

probabilities in anSFN (at point P2)

using COFDM withan effective symbol

time of 896 µs(8K-FTT) and a

guard interval of 1/4,equivalent to 224 µs,

for both directionaland omnidirectionalreceiving antennas.


mean distance between successive constellationpoints is lower when transmission occurs onlywithin one quadrant. This difference is small1 inthe case of uniform hierarchical modulation(uniform QAM) where the constellation diagramremains unaltered compared with non-hierarchical16- or 64-QAM.

Decoding of the HP datastream can be furtherimproved by increasing the distance of theconstellation points from the coordinate axes.Fig. 7 shows the constellation diagram for non-uniform hierarchical 64- or 16-QAM with α = 2(α = 2 means that the distance between twoconstellation points in different quadrants istwice as large as the distance within a quadrant).The DVB-T specification provides for a value ofα = 4 in addition to the values α = 1 (uniform)and α = 2. However, the improvement in trans-mission reliability for 4-PSK with a value ofα > 1 causes a further increase in the requiredsignal-to-noise ratio for the LP datastream. Hier-archical modulation does, however, represent aninteresting application with respect to the cover-age probabilities.

1. Approximately 0.1 – 0.5 dB with code rates 1/2 – 3/4.

Fig. 8 shows the coverage probabilities in an SFNwith additional examples of hierarchical and non-hierarchical modulation. These are based on therequired signal-to-noise ratios2 for portable andstationary reception which are given in the DVB-Tspecification. This example shows 4-PSK in a64-QAM system. The same error protection with acode rate R = 2/3 is used for both datastreams, LPand HP, so the ratio of the data-rates between LPand HP is exactly 2 to 1. With the guard interval of1/4 considered here, this gives a data-rate of6.6 Mbit/s for the HP datastream.

The coverage probability for the three pro-grammes with non-hierarchical modulation isabout 65 % for portable reception and about 98 %for stationary reception. The coverage probabilityfor the LP datastream is reduced only minimallywith hierarchical transmission where α = 1,whereas it is significantly higher for the HP data-stream (90 % for portable reception). This meansthat the coverage probability for a service whichcan be received portably (HP: 90 %) is approach-ing the values for the services intended for station-ary reception (LP: 98 %). In the case where α = 2,a value of 98 % is produced for services intendedfor portable reception, compared with 95 % forservices intended for stationary reception. In oth-er words, owing to hierarchical modulation, thecoverage probability for at least one programmecan be increased significantly and there may evenbe no loss in practice for the LP datastreams.

The DVB-T specification also allows many otherpossible combinations, apart from the exampleshown in Fig. 8, by varying the error protectionfor datastreams HP and LP. For example, if4.98 Mbit/s-per-programme is adequate for anapplication, it is possible to accommodate asmany as three programmes in the LP datastream.Increasing the error protection for the HP data-stream from R = 2/3 to R = 1/2 (owing to the lowerdata-rate of 4.98 instead of 6.6 Mbit/s) results in acoverage probability of nearly 99 % for a porta-bly-received service at a value of α as low as 1,and 96 % for the remaining three programmeswith stationary reception. In this way a “good”supply is achieved for all services.

13. Conclusions

The definition of the DVB-T system parameters isnaturally subject to various conflicting interests.The impact of the system parameters on the cover-age probability is of particular interest for a futurebroadcasting service. Initially, the parameters

2. Simulation results of DLR and CCETT.

Figure 6 (upper)Hierarchicalmodulation: 4-PSKintegrated in16-QAM.

Figure 7 (lower)Hierarchicalmodulation: Improvedconstellation for4-PSK integrated in64-QAM or 16-QAMwith α = 2.


required for a system specification and their rela-tive effects on the coverage will be of greater in-terest than the results for a concrete applicationwith a real transmitter network and real signal-to-noise conditions (a task that must be performed bynetwork planners).

The study of the DVB-T coverage aspects pre-sented in this article has been based on the use ofa single-frequency network. This is because, inGermany, there is great interest in the use ofSFNs for reasons of frequency and power econo-my. The results are also fundamentally transfer-able to other network structures. More preciseresults about system behaviour, which are impor-tant for the actual planning of DVB-T services,are expected to be obtained from the EuropeanACTS project “VALIDATE” which will alsoverify the DVB-T specification in laboratory andfield tests.

Bibliography

[1] Draft ETS 300 744: Digital broadcastingsystems for television, sound and dataservices; Framing structure, channelcoding and modulation for digital terres-trial televisionETSI, November 1996.

[2] ITU-R Recommendation BT.601-4:Encoding parameters of digital televisionfor studios .

[3] Alard, M., Lassalle, R.: Principles of Modu-lation and Channel Coding for DigitalBroadcasting to Mobile ReceiversEBU Review – Technical, No. 224, August1987, pp 168-190.

[4] De Bot, P.G.M., Le Floch, B., Mignone, V.,Schütte, H.-D.: An Overview of the Modula-tion and Channel Coding Schemes Devel-oped for Digital Terrestrial TelevisionBroadcasting within the dTTb ProjectProceedings of IBC-94, Amsterdam, May1994.

[5] CEPT Interim Report: Planning andIntroduction of Digital Terrestrial Tele-vision (DVB-T) in EuropeCEPT/FM-PT24, WGA 11 Rev. 2, WGB 14,May 1996.

[6] ITU-R Recommendation PN.370-6: VHF andUHF propagation curves for the frequencyrange from 30 MHz to 1000 MHz; Broad-casting services .

[7] Böer, A.: Leistungsübertragung im Mehr-wegekanal (Power transmission in multipathchannels)IRT Technical Report No. B 118/91.

Figure 8Comparison of the

coverage proba-bilities in an SFN with

hierarchical andnon-hierarchical

modulation: COFDM,4-PSK in 64-QAM,

R = 2/3, effectivesymbol time 896 µs,

guard interval 1/4,directional andomnidirectional

receiving antennas.


[8] Groβkopf, R., Schramm, R.: Der DAB-Funk-kanal bei mobilem Empfang. Messungenin München. Vorhersagemodelle (TheDAB radio channel in mobile reception. Mea-surements in Munich. Prediction models)IRT Technical Report No. B 132/93.

[9] Groβkopf, R.: Prediction Error of Field-Strength Calculations after Recommen-dation ITU-R PN.370-5 for the VHF andUHF RangeIRT, Munich, March 1994.

[10] Meilhoc, M.: Interfering Effects in COFDMCCETT, Rennes, 1992. EBU documentR1/DIG 081.

[11] Mignone, V., Morello, A.: Simple DVB-T Sys-tem Models for MFN and SFN PlanningRAI, Turin, August 1996.

[12] EBU document Tech 3254 (1988): Planningparameters and methods for terrestrialtelevision broadcasting in the VHF/UHFbands.

[13] ITU-R Recommendation BT.419-3:Directivity and polarization discriminationof antennas in the reception of televisionbroadcasting.

Dr Chris Weck graduated in Electronic Engineering from the Technical University of Berlin in 1986 andjoined the “Systems for Sound and Data Transmission” section of the IRT in Munich. First, he worked onchannel coding for digital audio broadcasting with regard to subjective criteria. In the framework of theEureka-147 DAB Project, he was involved in the system specification and contributed especially to thedefinition of the unequal error protection scheme.

Since 1993, Chris Weck has participated in various national and European projects dealing with thedevelopment and specification of digital terrestrial television (e.g. HDTV, RACE dTTb, DVB task forces).He has been engaged in the field of channel coding and modulation, and also in the coverage aspects ofSFNs. In the ACTS Project VALIDATE (verification and launch of the DVB-T specification), he iscurrently Chairman of the “Field Tests” Task Force.

In 1996, Chris Weck obtained a Ph.D. from the Technical University of Berlin, in the field of source-adaptive channel coding for digital audio broadcasting.

Mobile DVB-T – important results

Cologne, January 1997.

RTL and Deutsche Telekom AG are carrying out a series of trials using DVB-T compliantequipment supplied by the UK company, DMV. The trials, covering the area of Cologne inGermany, are designed to test the reception of DVB-T in a variety of conditions.

Initial results have been extremely promising, with the DVB-T system perform-ing better than predicted. The preliminary tests consisted of transmitting 2kDVB-T signals from the Cologne transmitter of Deutsche Telekom with an ERP(equivalent radiated power) of 1 kW on Channel 40. A variety of DVB-Tparameters were chosen including QPSK, 16-QAM and 64-QAM modulationwith guard intervals of 1/4 and 1/8. Both the QPSK and 16-QAM FEC (forwarderror correction) options of 2/3 and 3/4 were tested. Reception was based onthe use of an omnidirectional set-top antenna and a short antenna designed formobile telecommunications reception in a car. The initial trials investigated thestationary, portable and mobile reception of the various DVB-T signals.

Most interesting were the results for mobile reception. A car was fitted with the set-top anten-na and a DMV DVB-T receiver. The RTL/Deutsche Telekom team found that reception fromthe transmitter was perfect for QPSK and 16-QAM at normal traffic speeds in the centre ofCologne. Furthermore, the 16-QAM DVB-T signals were received perfectly on a motorwayat speeds of up to 140 km/h. QPSK DVB-T was perfectly received at speeds in excess of 170km/h. In fact, according to Dr Paech (RTL), the limiting factors were the received power fromthe transmitter, the traffic and the fact that the car used for the trials had a maximum speedof about 170 km/h!

More extensive trials are underway, and the results are expected to be at least as good asthese preliminary findings.

Coverage aspects of digital terrestrial television ... · 2. Digital terrestrial television...

Documents

Transcript of Coverage aspects of digital terrestrial television ... · 2. Digital terrestrial television...