1、 Rec. ITU-R M.1316-1 1 RECOMMENDATION ITU-R M.1316-1*Principles and a methodology for frequency sharing in the 1 610.6-1 613.8 MHz and 1 660-1 660.5 MHz bands between the mobile-satellite service (Earth-to-space) and the radio astronomy service (Question ITU-R 201/8) (1997-2005) Scope This Recommend
2、ation provides the principles and methodology that may be applied for the protection of the radio astronomy observations from emissions of land and maritime mobile earth stations in the 1 610.6-1 613.8 MHz and 1 660-1 660.5 MHz bands. Annex 1 describes the three steps to be followed, with Annex 2 de
3、scribing the calculation of the “separation distance by default” using the Monte Carlo methodology based on the principle of sampling random variables. Annex 3 calculates the restriction zones. Step 3 is the calculation of exclusion zones using the specific characteristics of the systems involved. T
4、he ITU Radiocommunication Assembly, considering a) that the World Administrative Radio Conference for Dealing with Frequency Allocations in Certain Parts of the Spectrum (Malaga-Torremolinos, 1992) (WARC-92) allocated the band 1 610-1 626.5 MHz on a primary basis to the mobile-satellite service (MSS
5、) in the Earth-to-space direction, and the band 1 610.6-1 613.8 MHz on a primary basis to the radio astronomy service (RAS); b) that the frequency band 1 610.6-1 613.8 MHz is used by radio astronomers to observe the spectral line of the hydroxyl molecule near 1 612 MHz; c) that No. 5.372 of the Radi
6、o Regulations (RR) states that “Harmful interference shall not be caused to stations of the radio astronomy service using the band 1 610.6-1 613.8 MHz by stations of the radiodetermination-satellite and the mobile-satellite services (No. 29.13 applies)”; d) that the mobile-satellite systems operatin
7、g in the 1 610-1 626.5 MHz band are likely to be utilizing mobile earth stations (MESs) with omni-directional antennas; e) that the 1 660-1 660.5 MHz frequency band is allocated to the RAS on a shared, primary basis with the land-mobile satellite service (LMSS) in the Earth-to-space direction; f) th
8、at the importance of the allocation at 1 660-1 660.5 MHz to the RAS was confirmed by Resolution 6 of the 20th General Assembly of the International Astronomical Union (IAU) (Baltimore, United States of America, August, 1988) and reconfirmed at the 21st General Assembly of the IAU (Buenos Aires, Arge
9、ntina, July, 1991) and at the 22nd General Assembly of the IAU (The Hague, The Netherlands, 1994); g) that Recommendation ITU-R RA.1031 does not fully take into account the statistical nature of the interference caused by mobile transmitters, *This Recommendation should be brought to the attention o
10、f Radiocommunication Study Group 7. 2 Rec. ITU-R M.1316-1ITU-R M.1316-1 recommends 1 that principles and methodologies similar to those described in Annex 1 may be used in coordination between radio astronomy stations and land and maritime MESs in the bands 1 610.6-1 613.8 MHz and 1 660-1 660.5 MHz;
11、 2 that further studies are needed by ITU-R, including studies for aircraft earth stations, in order to review the applicability of this Recommendation for detailed coordination between the MSS and the RAS; 3 that in any application of a methodology the input parameters should be agreed by the parti
12、es concerned during coordination; 4 that the ITU-R should, in cooperation with the Radiocommunication Bureau, jointly develop a computer program to implement the methodology given in Annexes 1 to 4. Annex 1 Assessment of the interference from MES/MSS into radio astronomy observations The protection
13、of radio astronomy observations can be provided through three different steps: Step 1: by setting a separation distance by default between an RAS site and an MES, which defines an area around an RAS site outside of which no restriction applies to the operation of MESs. Step 2: by setting a restricti
14、on zone around an RAS site, which defines an area within which there may be some restriction to the operation of MESs. These restrictions should be defined by the regulator and agreed by the radio astronomy community and MSS operator. Step 3: by setting an exclusion zone around an RAS site, defined
15、by means of detailed assessment of the characteristics of the systems involved and measurements if necessary, within which no operation of MESs should be allowed. Annex 2 and Annex 3 describe methodology which should be used for the calculations respectively for Step 1 and Step 2. Precise details fo
16、r the conditions for the operation of the mobiles in the restriction zone need to be agreed by the concerned parties, in order to arrive at an exclusion zone, following the definition above in Step 3. Annex 4 provides the list of the set of characteristics which are necessary for running a simulatio
17、n. Step 1 calculation is intended to provide separation distances by default. Annex 2 describes a general methodology of calculation which can be used for that purpose, using Monte Carlo methodology. The basis of this model is to calculate the statistics of the interfering power produced at an RAS s
18、ite by MESs in operation. Rec. ITU-R M.1316-1 3 In order to protect radio astronomy observations, it is stated that: “a 2000 second integration taken at any time of the day should have at least (100 x)% of being interference free, i.e. the mean interference power is below the levels specified in Rec
19、ommendation ITU-R RA.769. The figure of 90% (x = 10) has its origin in propagation calculations (ITU-R Handbook on Radio Astronomy, 4.2.4. See also Recommendation ITU-R RA.1031). The wider interpretation of this figure is under consideration within ITU-R”. Thus, Annex 2 methodology should be used wi
20、th the following assumptions: 2 000 s integration time (constant for all trials); peak traffic assumption; x% of time maximum interference criteria for the RAS (10% is the current value subject to revision by the ITU-R). In the case where different sources of interference are identified for the radi
21、o astronomy observations, further studies are required on the possible splitting of the maximum interference power level. Annex 2 Step 1 methodology: Calculation of separation distances by default between RAS sites and MESs 1 Introduction This Annex describes a general methodology which can be used
22、for the calculation of separation distances by default between RAS sites and the areas where MESs are allowed to transmit. These separation distances, based on calculations using a Monte Carlo methodology, should ensure the protection of radio astronomy observations. 2 General principles used in the
23、 methodology 2.1 Monte Carlo methodology In order to calculate the separation distances by default between RAS sites and MESs, it is necessary to evaluate the probability function of the interfering power produced by the mobiles and experienced by the RAS receivers. This can be done by using statist
24、ical modelling of interference, such as a Monte Carlo methodology. The Monte Carlo methodology is based on the principle of sampling random variables from their defined probability distributions. The variables to be sampled are often various and numerous, as the accuracy of the model usually increas
25、es with their number. In the particular case of the determination of separation distances by default, these variables may include the number of mobiles, the location of the mobiles, the propagation condition. 4 Rec. ITU-R M.1316-1ITU-R M.1316-1 The statistics of the interfering power produced at an
26、RAS site by MESs in operation is then derived from the calculation of interfering powers experienced for each sample. 2.2 Protection of radio astronomy observations Radio astronomy observations are performed by using time averaging, to significantly reduce noise fluctuations. In order to reflect suc
27、h practice, statistics of received interfering power are based on integration time samples used during the observations. The interference power coming from the MSS population is acceptable provided that no more than x% of the 2 000 s integration periods have mean interference power above the RAS det
28、rimental level. The following is based on this definition. 3 Presentation of the methodology of calculation As stated in 2, statistics of interfering powers are based on integration time samples: niter: number of integration time samples needed for the statistic. integr: duration of the integration
29、time sample. In the following, integr is supposed to be constant. During each integration time sample integr, the mean interference power produced by MESs is calculated by averaging “instantaneous” interfering powers produced within sub-time steps of t seconds duration. During each of the sub-time s
30、teps, interfering powers are determined by making random trials on the traffic load of the MSS system under consideration and on the location of each MES in operation. Rec. ITU-R M.1316-1 5 The outline flow chart of the calculation is given in Fig. 2: 4 Calculation of the interfering power experienc
31、ed during a sub-time step modelling of traffic The interfering power experienced during each sub-time step at the frequency is calculated by summing power produced by each mobile in operation during this time step. For each time step it, it is thus necessary to determine: the number of mobiles in op
32、eration during it (derived from a given traffic law); the channels the active MESs are using; the location of the mobiles around the RAS site distance, azimuth, etc. In order to keep the correlation between each sub-time steps, the number of mobiles in operation for sub-time step it is derived from
33、the number of mobiles in operation for it 1, by taking into account the number of calls dropped and initiated in-between. 6 Rec. ITU-R M.1316-1ITU-R M.1316-1 For the first sub-time step, the initial number of calls is calculated by making a random trial. Figure 3 gives the outline chart for calculat
34、ion of rec_poweriter,it( f ) (integration time sample iter, sub-time step it). 4.1 Monte Carlo random trials As stated in 2.1, the Monte Carlo methodology is based on the principle of sampling random variables from their defined cumulative distribution functions. Considering for example a variable x
35、, with the cumulative distribution function P(X), is then the probability p(x X). P(x) is uniformly distributed between 0 and 1. So, a random uniform trial of P = P(xi) between 0 and 1 leads to a single value of xi, enabling x = f 1(P) to be plotted. Rec. ITU-R M.1316-1 7 4.2 Calculation of the init
36、ial number of calls: iniiterAt the beginning of each integration time sample, the initial number of calls is calculated using the formula giving the cumulative distribution function of having iniitersimultaneous calls at any instant t: PEiEiiiiniiiNcalliter=!00(1) where: P : cumulative probability o
37、f having iniitersimultaneous calls at the instant t (iniiter Ncall) E : peak demanded load on the system measured (E) Ncall : maximum number of simultaneous calls the MSS system can support. Thus, iniitercan be derived from an uniform random trial of P by using formula (1) (see 4.1). 4.3 Calculation
38、 of the number of dropped calls: droppediter,itThe number of dropped calls for the iterthintegration time sample is calculated by determining the number of calls for which the call duration is less than or equal to it: If ncalliter,it 1is the number of calls of sub-time step it 1 (it 1), consider on
39、e specific call c of this sub-time step: if the call termination time of call c is less or equal than it then this call is dropped and is not retained for the calculation of rec_poweriter,it( f ). This call is added to the dropped call count (droppediter,it). if the call termination time of call c i
40、s more than it, then this call is retained for the calculation performed at sub-time step it. 4.4 Calculation of the potential number of attempted calls: borniter,itFor each time step, the potential number of initiated calls is calculated using the formula giving the cumulative distribution function
41、 of birth of calls over a specified interval of time: Ptiitiborniter it=()!,e0(2) where: P : cumulative probability of having borniter,itcalls attempted between sub-time steps it and it + 1 : is the mean call rate of the satellite system t : is the sub-time steps duration. Thus, borniter,itcan be de
42、rived from an uniform random trial of P by using formula (2) (see 4.1). 8 Rec. ITU-R M.1316-1ITU-R M.1316-1 4.5 Calculation of the effective number of new calls newiter,it, and of the number of calls ncalliter,itAmong the calls (attempted), not all will succeed because of the physical limitations of
43、 the system (maximum number of calls). If ncalliter,it 1is the number of calls at sub-time step it 1 (it 1) (used in the calculation of rec_poweriter,it( f ) then: droppediter,itis the number of calls dropped between sub-time steps it and it + 1, borniter,itis the number of calls attempted between s
44、ub-time steps it and it + 1, then the effective number of calls to be taken into account for the calculation of rec_poweriter,it( f ) is calculated using the following formula: ncalliter,it= min(Ncall, ncalliter,it 1+ borniter,it droppediter,it) (3) and the number of effective new calls is then: new
45、 ncall ncall droppednew borniter it iter it iter it iter ititer it iter it, ,()= + 1(4) If it = 1, ncalliter,1= iniiter. 4.6 Assignment of the (new) calls to the available traffic channels With both code division multiple access (CDMA) and time division multiple access (TDMA), several calls can be a
46、llocated to the same physical channel. We define a traffic channel each of the possible call slots (identified in the time domain for TDMA or by the code for CDMA), so that there are nmax traffic channels for any physical channel. The total traffic is uniformly distributed amongst the available traf
47、fic channels in an area having a radius of one spot beam. This means that: if it = 1, the iniitercalls are uniformly distributed over all the Ncall traffic channels, if it 1, the newiter,itcalls are uniformly distributed over the ncalliter,it 1 droppediter,itavailable traffic channels. If different
48、distributions for the assignment of traffic among the available traffic channels are provided by operators, these may be incorporated into the methodology. 4.7 Calculation of the call termination time of the (new) calls For each new call c, the call termination time is determined by using the formul
49、a giving the cumulative distribution function call duration: PTtc=1e()(5) where: P : cumulative probability of having a call duration of less than (Tc t) t : current sub-time step (date of birth of the call) Tc: call termination time : mean call length of the satellite system. Rec. ITU-R M.1316-1 9 Thus, Tccan be derived from an uniform random trial of P using the following formula (see 4.1): Tc= t .ln (1 P) (6) 4.8 Calculation of the other parameters linked to the (new) calls 4.8.1 Calculation o
