1、22 Rec. ITU-R F.634-4 RECOMMENDATION ITU-R F.634-4 ERROR PERFORMANCE OBJECTIVES FOR REAL DIGITAL RADIO-RELAY LINKS FORMING PART OF THE HIGH-GRADE PORTION OF INTERNATIONAL DIGITAL CONNECTIONS AT A BIT RATE BELOW THE PRIMARY RATE WITHIN AN INTEGRATED SERVICES DIGITAL NETWORK (Question IT-R 13419) (198
2、6-1 990-199 1-1994-1997) The IT Radiocommunication Assembly, considering a) that the error performance objectives of a high-grade 2 500 km hypothetical reference digital path (HRP) at a bit rate below the primary rate for digital radio-relay systems are given in Recommendation ITU-R F.594, which app
3、lies to N x 64 kbit/s (1 5 Ne 24 ( or e 32, respectively) (see Note 1); b) that network performance objectives for digital sections are given in IT-T Recommendation G.921; cl that real paths which form part of the high-grade portion of an integrated services digital network (ISDN) sometimes differ i
4、n composition from the HRDP (see Recommendation IT-R F.556, Fig. 1). and share radio-relay links shorter than 2 500 km with other digital links; d) that conformity with digital radio performance Recomendations cannot be easily established by direct measurements on real systems due to the seasonal an
5、d annual variations in propagation conditions, and that practical advice on how to apply performance Recommendations is therefore necessary; e) and planning of real radio-relay links forming part of the high-grade portion of an ISDN; that, therefore, it is necessary to give objectives for allowable
6、bit error ratios (BERS) as a guide in the design f) that the Recommendations IT-R F.1092 and ITU-R F.1189, based on ITU-T Recommendation G.826, give error performance objectives for constant bit rate digital paths at or above the primary rate carried by digital radio-relay systems which may form par
7、t of the international and national portions, respectively, of a 27 500 km hypothetical reference path, recommends 1 that error performance should be assessed in terms of the events errored seconds (ES) and severely errored seconds (SES) and the parameters errored second ratio (ESR) and severely err
8、ored second ratio (SESR) as defined in ITU-T Recommendation G.82 1 (see also Recommendation ITU-R F.594); 2 that when establishing real digital radio-relay links with length, L (km), of between 280 km and 2 500 km providing connections at a bit rate below the primary rate and intended to form part o
9、f a high-grade circuit within an ISDN, the following error performance objectives should be respected for each direction of the N x 64 kbit/s (1 IN 2500 km), but the SESR evaluated for the total link length L should not exceed the objective 0.0005 + (L/2500) x 0.00004 in any month. Since the perform
10、ance of digital radio-relay systems is dependent upon fading, it is generally agreed that the behaviour of any section of the HRDP will be statistically independent. If this is assumed, then mathematically the HRDP performance could be determined by the convolution of the probability density functio
11、ns of all sections. This process is however, not a practical one, since the probability density function is not known in sufficient detail. The example given in Fig. 1 illustrates a further important principle coming from the application of ITU-T Recommendation G. 102. This Recommendation draws atte
12、ntion to the need for a margin between network performance objectives (NPO) and equipment design objectives (EDO) which are essentially the performance predictions for each hop (see also $ 2.5.5 of ITU-T Recommendation G.102). In Fig. 1, the predicted outage for the hop (see Recom- mendation ITU-R F
13、.1093) accounts for only a proportion of the objective, 10% is allocated to interference from the fixed-satellite service (FSS), whilst the remainder is used as a margin to take account of inaccuracies in the outage prediction method and for other non-deterministic effects. It is necessary to check
14、that all the objectives of this Recommendation are satisfied during route design including SESR, and ESR. Procedures for predicting performance are described in Recommendation ITU-R F. 1093 and the models generally evaluate the percentage of time that a BER of 1 x will be exceeded during a worst-mon
15、th for multipath fading. Performance measurement has shown that the majority of this time is composed of event durations which are less than 10 s. For this reason, as the worst-case, the predicted multipath outage could be considered as SES rather than unavailable time (see Recommendation ITU-R F.55
16、7 for the definition of unavailable time). 1.2 Network performance objectives (NPOs) and equipment design objectives (EDOs) ITU-T Recommendation G.80 1 describes the application of digital transmission models for the specification and sub-division of transmission parameters. It shows a way in which
17、a national reference model can be used to arrive at NPOs and EDOs for real transmission paths using objectives for hypothetical reference digital connections (HRDXs), hypothetical reference digital links (HRDLs) and HRDSs. This procedure is applicable to all transmission media and thus to radio-rela
18、y systems. COPYRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTD.ITU-R RECMN F-b34-4-ENGL 1997 m 4855232 0530944 302 m Rec. ITU-R F.634-4 FIGURE 1 Performance degradations due to the FSS Number of severely errored seconds (SES) in any mo
19、nth Network performance objective for hop e.g. 36 SES Total degradation allocated to interference from the FSS (1 0%) I l 1 PLrformance prediction , excluding interference from the FSS 25 Receiver thermal noise power (dB) Increasing interference power (dB) 0634-01 With the aid of HRDS, each administ
20、ration can establish its own reference model tailored to the prevailing national network conditions. It has to be ensured, however, that this model meets the objectives of ITU-T Recommen- dation G.82 1. To obtain a representative network model, the individual network parts should be composed of digi
21、tal sections such that as many as possible of the real links are taken into consideration. A block allowance can then be determined for each model network part, from the chosen arrangement of digital sections and their associated allocation percentages. This block allowance represents a NPO for the
22、network parts concerned and has to be met by real transmission paths. In terms of design, any radio-relay systems employed in these parts must be capable of fulfilling the applicable NPO. In this method, the radio-relay systems to be used (e.g. frequency band, bit rate, etc.) and the maximum possibl
23、e number of hops are defined for each part of the network. Equipment design objectives could then be specified by taking into account environmental effects, equipment ageing, etc. If, in practice, real links are installed which do not exhaust the values given for a particular part of the network mod
24、el - e.g have less hops - it is proposed that the link should still be allocated the full block allowance. The idea is to reduce the constraints at the time of the initial concept of the system. For example, the use of space diversity, use of adaptive time domain equalizers could be reduced, smaller
25、 sized antennas and lower power transmitters could be used and appropriate performance allowances for oversized hop lengths could be provided. 1.3 Intermittent failure mechanisms A system can experience an intermittent malfunction before total failure occurs. It is important, in maintaining good ser
26、vice, to detect such a condition as early as possible and take corrective action. COPYRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTDmITU-R RECMN FOb34-4-ENGL 3997 m 4855232 0530945 249 m 26 Rec. ITU-R F.634-4 Criterion 1) SESR 2) ESR
27、3) Unavailability 2 The criteria given in ITU-T Recommendation G.821 (and hence Recommendation ITU-R F.594 and this Recommen- dation) are normally too long for use as maintenance limits or for circuit provisioning tests. Measurements over much shorter periods (e.g. one day) may be necessary in order
28、 to determine whether a circuit is fit for service or should receive maintenance attention. Error performance objectives for periods shorter than one month Objective for 2 500 km HRDP ITU-R F.594 and ITU-R F.557) (Recommendations Allowance for band sharing 0.00054 in any month 0.0032 in any month 0.
29、3% of a period probably greater than one year 0.000054 in any month 0.00032 in any month 0.03% of a period probably greater than one year Radio systems differ from line systems in that the significant performance degradations (i.e. those due to fading) tend to be concentrated into a few days whereas
30、 for line systems they tend to be randomly distributed throughout the month. For this reason the mathematical models being examined by the ITU-T to describe the distribution of performance degradations within a month may not be suitable for radio systems. This subject requires active study. Field me
31、asurements have shown that radio systems cannot respect daily performance objectives which are linearly sub-divided with distance and time from the 2500 km HRDP. The measurements show that a 24-h performance objective of 20% of the one month objective can be met for SES and ES criteria. Further meas
32、urements are required to confirm the validity of these limits. However, the ITU-T Recommendation G.821 parameters are not ideally suited for monitoring radio system performance over short periods, particularly where fading causes bursty error distributions. Any performance assessment derived from a
33、single 24-h measurement period will contain a significant degree of uncertainty. It is therefore proposed that any limits used for periods shorter than one month should only be used in conjunction with other supporting performance related data; for example: - - local meteorological data; - - other s
34、ystem parameters (e.g. automatic gain control (AGC) levels and trends); performance of other radio links in the area; historical performance of radio system. 3 Interference from the FSS Many of the frequency bands used by radio-relay systems are now shared with the FSS. Interference from the FSS gen
35、erally may take two forms: - - to digital radio systems. The limits for the degradations to a 2 500 km HRDP quoted in Recommendation ITU-R SF.615 are given in Table 1. from earth stations in the bands used for up links; from satellites in the bands used for down links TABLE 1 Recommended limits for
36、performance and availabiliy degradations due to interference from sharing with the FSS Some allowance needs to be made for these degradations when setting real route design objectives for systems operating in the shared bands. It is therefore necessary to consider how to apportion the degradations i
37、n Table 1 to shorter links. COPYRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTDeITU-R RECMN F-634-4-ENGL I1777 4855212 053094b 185 Rec. ITU-R F.634-4 27 Sharing models have previously assumed that only one or two stations within a 2 50
38、0 km HRDP will be significantly affected by this kind of interference. This would imply that the majority of the degradations listed in Table 1 could be allocated to one or two hops and that sharing criteria (e.g. power flux-density limits, equivalent isotropic radiated power limits) can be derived
39、on this basis. However, this may in future be an unrealistic assumption since the number of earth stations is generally increasing (especially for systems operating above 10 GHz) and these now are often sited close to major cities. In consequence, the number of radio-relay receivers affected will al
40、so generally increase. This could lead to subsequent performance degradations which were not taken into account in the initial route design. The most prudent basis for apportioning these FSS degradations within the HRDP would be on a hop-by-hop basis. Thus the 10% degradation would be included in th
41、e performance budget at the route design stage as illustrated in Fig. 1 for the case of SES. ANNEX 2 Translation rules suggested for the normalization of error performance measurement results obtained at the system bit rate at or above the primary rate to the 64 kbit/s level (applicable only to digi
42、tal radio-relay links designed prior to the approval of Recommendations ITU-R F. 1092 (1994) and ITU-R F. 1 189 (1995) This Annex can be used for digital radio-relay links within the high-grade portion of an ISDN that were designed prior to the approval of Recommendations ITU-R F. 1092 (1 994) and I
43、TU-R F. 1 189 (1 995). In this case the relationship between objectives at 64 kbits and corresponding parameters at the system bit rate should be taken into account. 1 SESR objectives Various theoretical and experimental studies have indicated that the direct translation of SESR is accurate within a
44、 few per cent, or: The SESR normalized to 64 kbits can be assessed from measurements made at the system bit rate as follows: SESR64 = Y + where: Y: Z: SESR at the system bit rate non-SESR at the system bit rate containing one or more loss of frame alignment at the system bit rate. The factor Z repre
45、sents an allowance for error bursts which are extended or cause loss of frame alignment during the demultiplexing process between the system bit rate and 64 kbits. It should be noted that Z could be dependent on transmission equipment design and will not necessarily include all non-SES at the measur
46、ement bit rate which ultimately cause SES at the 64 kbits level. Certain error bursts which can pass transparently through demultiplexers at the hierarchical level of the source of the error burst without causing a loss of frame alignment at that level, can cause losses of frame alignment in demulti
47、plexers at lower hierarchical levels. These events cannot, therefore, be accurately assessed as SES in the measurement at the source of the event. In the case of digital radio-relay systems the factor Z is thought to lie between 0.01 and 0.05 of the measured , performance Y. COPYRIGHT International
48、Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTD-ITU-R RECMN Fmb34-4-ENGL 1797 4855232 0530947 033 28 Rec. ITU-R F.634-4 2 Relationship between ESR objective at 64 kbit/s and corresponding parameters at the system bit rate The following relationship is use
49、d. The ESR at 64 kbits is given by: J i=l where: n: number of errors in the zh s at the system bit rate N: system bit rate divided by 64 kbits J: integer number of 1 s periods (excluding unavailable time) within the total measurement period. The ratio (n/N)i for the ih s is: n/N if ON. This relationship is conservative, since it assumes that the errors occurring at system bit rate are uniformly distributed among the 64 kbits channels. In practice, due to the non-uniform distribution of errors the actual result at 64 kbits is better than the value calculated with the expressi
