ITU-R P 843-1-1997 Communication by Meteor-Burst Propagation《通过流行猝发传播的通信》.pdf

1、STD-ITU-R RECMN P-8q3-1-ENGL L777 48552L2 0527768 2Tl 1 Rec. ITU-R P.843-1 RECOMMENDATION ITU-R P.843-1 COMMUNICATION BY METEOR-BURST PROPAGATION (Question ITU-R 221/3) (1992-1997) The ITU Radiocommunication Assembly, considering that scattering from ionization caused by meteor trails can provide a

2、convenient means of communication a) at HF and VHF: b) ranges up to 1 800 km; cl rates up to 100 Bd when averaged over approximately one hour, that two-way telecommunication circuits are in operation with frequencies between 30 and 100 MHz over that communication relies on bursts of propagation duri

3、ng the occurrence of meteor trails and can support data recommends that the following information should be used in the design and planning of meteor-burst communications systems. 1 At certain times of the year, meteors occur in the form of showers and may be prolific over durations of a few hours.

4、There is, however, a general background of meteors incident upon the Earth from all directions and it is appropriate to consider only these sporadic meteors for communication-planning purposes. For sporadic meteors at mid-latitudes there is a roughly sinusoidal diurnal variation of incidence with a

5、maximum at 0600 h and a minimum at 1800 h local time. The ratio of maximum to minimum averages about four. In the Northern Hemisphere there is a seasonal variation of similar magnitude with a minimum in February and maximum in July. Considerable day-to-day variability exists in the incidence of both

6、 sporadic and shower meteors. The annual average flux of meteors incident per unit area and producing electron-line densities q exceeding a threshold qo per metre, I (q 40) is given as: Temporal variations in meteor flux By combining this overall meteor rate with a representative sinusoidal diurnal

7、variation and the seasonal factor, M, from Fig. 1 the average temporal changes in meteor flux can be estimated: 160 M 1 + 0.6 (sin E) m-2 s-l 40 where: T: local time (h). For planning purposes it may only be necessary to consider the worst combination of month and local time. 2 Meteors occur in all

8、parts of the world at all hours but statistical information is incomplete on their geographical distribution and trail directions. Spatial variation in meteor flux COPYRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling Services2 Rec. ITU-R P.843-1 FI

9、GURE 1 Month-to-month variation in sporadic meteor flux rate relative to the average value 1.8 1.6 1.4 6 c! x 1.2 E 1.0 r ?i c; 0.8 u 0.6 .* a 8 4-4 0.4 0.2 O JFMAMJJ ASOND hth 0843-01 Until such times as spatial variations are quantified it is recommended that flux estimates based on the method giv

10、en in 8 1 are used at all latitudes. 3 Underdense and overdense trails The ionized trails caused by meteors are classified as underdense or overdense according to the intensity of the ionization. The division between the two cases occurs for line densities of approximately 2 x 1014 electrons per met

11、re. The amplitude of signals scattered from underdense trails may be calculated by summing the scattered field arising from each individual electron. Overdense trails are those for which the coupling between electrons cannot be ignored, in which case, the reflecting properties are calculated as if t

12、he trail were a long metallic cylinder. At frequencies used in practice the echoes from underdense trails show an abrupt start followed by an exponential decay, whereas those from overdense trails have more rounded envelopes and are of longer duration. The relative proportions of underdense and over

13、dense echoes will depend on the system sensitivity. The relation between number of trails and peak amplitude, A, can be upproximdted by: Number of trails -(A)-v where w vanes from 1.0 at low signal levels to greater than 2.0 at larger signai levels where the majority of trails are overdense. For man

14、y links, the index y is of the order of 1.1 to 1.4. Results in the systems used so far indicate that echoes are predominantly from underdense trails. On this basis it is recommended that planning for a typicdl system should proceed on the basis that all meteor trziils are of the underdense type. COP

15、YRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesRec. ITU-R P.843-1 3 4 Effective length and radius of meteor trails 4.1 Effective length The ray geometry for a meteor-burst propagation path is shown in Fig. 2 between transmitter T and rec

16、eiver R. P represents the tangent point and P a point further along the trail such that (Ri + Ri) exceeds (RI + R2) by half a wavelength. Thus PP (of length L) lies within the principal Fresnel zone and the total length of the trail within this zone is 2L. Provided R1 and R2 are much greater than L,

17、 it follows that for practical cases: I” AR1 R2 = (Ri + R2) (i - sin2 cpcos2 ) where: cp : angle of incidence : angle between the trail axis and the plane of propagation h: wavelength. FIGURE 2 Ray geometry for a meteor-burst propagation path (3) C: Earths surface 0843-02 D: plane of propagation E t

18、rail F tangent plane : angle between the trail axis and the plane of propagation T transmitter R: receiver 4.2 Trail radius In order to evaluate the scattering cross section of the tmil it is usual to assume that ambipolar diffusion causes the radial density of electrons to have a Gaussian distribut

19、ion and that the volume density is reduced while the line density remains constant. The ionization trail immediately behind a meteor is formed near-instantaneously with a finite width. This is called the initial trail radius, 0. An empirical relationship between ro, and the meteor height is: logro =

20、 0.035 h - 3.45 (4) where: h : trail height (km) ro : initial trail radius (m). COPYRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling Services STDBITU-R RECMN P.843-L-ENGL 1977 W 4855232 0527773 87b 4 Rec. ITU-R P.843-1 5 Received power and basic tr

21、ansmission loss 5.1 Received power Since any practical meteor-burst communication system will rely mainly on underdense trails, the overdense formulae are of less importance. Satisfactory performance estimates can be made using formulae for the underdense case with assumed values of 4 in the range I

22、Ol3 to lOI4 electrons per metre according to the prevailing system pardmeters. The received power PK (t) after scattering from underdense trails at frequencies used in practice is as follows: wavelength (m) echoing area of the trail (m2) loss factor due to finite initial trail radius loss factor due

23、 to trail diffusion loss factor due to ionospheric absorption time in seconds measured from the instant of complete formation of the first Fresnel zone half the time taken for the meteor to traverse the first Fresnel zone transmitter power (W) power available from the receiving antenna (W) transmit

24、antenna gain relative to an isotropic antenna in free space receive antenna gain relative to an isotropic antenna in free space (Lossless transmitting and receiving antennas are assumed) RI, R2 : distances (m) see Fig. 2. The echoing area B is given as: where: r, : effective radius of the electron =

25、 2.8 x a : angle between the incident electric vector at the trail and the direction of the receiver from that point. m Since L2 is directly proportional to h, the echoing area, 0, is also proportional to h and hence the received power for underdense trails varies as h3. Horizontal polarization norm

26、ally is used at both terminals. The sin2 a term in equation (6) is then nearly unity for trails at the two hot spots. The loss factor al is given by: It represents losses arising from interference between the re-radiation from the electrons wherever the thickness of the trail at formation is compara

27、ble with the wavelength. COPYRIGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTD*ITU-R RECMN P-843-3-ENGL 3777 D 4855232 0529772 722 D Rec. ITU-R P.843-1 5 The factor a2(t) allows for the increase in radius of the trail by ambipolar diffus

28、ion, It may be expressed as: q(t) = exp - h2sec2cp 32x2Dt1 where D is the ambipolar diffusion constant in m2 s- given by: log D = 0.067 h - 5.6 (9) The increase in radius due to ambipolar diffusion can be appreciable even for as short a period as is required for the formation of the trail. The overa

29、ll effect with regard to the reflected power is equal to that which would arise if the whole trail within the first Fresnel zone had expanded to the same extent as at its mid-point. Since this portion of trail is of length 2L the mid-point radius is that arising after a time lapse of Ll V (s) where

30、Vis the velocity of the meteor in ms-l Calling the time lapse to gives, for trails near the path mid-point (RI = R2 = R): - For trails at right angles to the plane of propagation ( = 90“): - For trails in the plane of propagation ( = O“): ro (%I“ . seccp Substituting to from equation (10) into equat

31、ion (8) gives for the = 90“ case: For = O“ the exponent in this expression is sec p times greater. The ratio of the ambipolar diffusion constant D to the velocity of the meteor V (required in the evaluation of received power) can be approximated by: Dl V = 0.0015 h + 0.035 + 0.0013 (h - 9O)2 10-3 (1

32、3) a2(t) is the only time-dependent term and gives the decay time of the reflected signal power. Defining a time constant Tu, for the received power to decay by a factor e2 (Le. 8.7 dB) leads to: With reflection at grazing incidence sec2 p will be large and hence so is the echo-time constant. The ec

33、ho-time constant is also increased by the use of lower frequencies. 5.2 Basic transmission loss Basic transmission loss curves derived from equation (5) with q = 1014 electrons per metre are given in Fig. 3. As the angle can take any value between O“ and 90“ only these two extreme cases are shown. T

34、he advantage of lower propagation loss at the lower frequencies is clearly seen. Average meteor heights given from equation (15) have been used in deriving the curves. It should be noted that the prediction of system performance depends critically on the heights assumed. COPYRIGHT International Tele

35、communications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTD-ITU-R RECMN P-843-1-ENGL 1997 4855232 0527773 bb7 9 6 150 155 160 h 9 165 y: c 170 5 175 e 2 180 v I ._ w c u .- c9 185 I90 195 Rec. ITU-R P.843-1 FIGURE 3 Basic transmission loss for underdense trails given by

36、equation (5) with q = 1014 electrondm and horizontal polarization I I 1 I I I O 200 400 600 800 i000 1200 1400 1600 1800 2000 Ground range (km) Trails at right angles to the plane of propagation ( = 90“) - Trails in the plane of propagation ( = O) 0843-03 6 The underdense echo ceiling and average tr

37、ail height Both the initial trail radius, i-O, and the ambipolar diffusion constant, D, increase with altitude. Consequently, the loss factors al and a2 (to) combine to reduce the number of underdense meteors occurring near the top of the meteor region which are useful for communication purposes. Th

38、is effect is usually referred to as the underdense echo ceiling. Similar constraints have been observed to exist in the monostatic case. Figure 4 shows the measured height distribution of underdense echoes using various radar frequencies. It can be seen that the lowest altitude at which underdense e

39、choes occur is 85 km, and that the altitude distribution is approximately Gaussian at any frequency. The average trail height h (km) at frequencyf(MHz) is: The average trail height is a function of other system parameters in addition to frequency. However equation (15) is a good approximation. COPYR

40、IGHT International Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling ServicesSTD-ITU-R RECMN P.843-L-ENGL 1997 4855232 052977LI 5T5 Rec. ITU-R P.843-1 7 FIGURE 4 Height distribution of underdense meteors providing echoes at frequencies of 18,36 and 70 MHz 80 90 1 O0 11

41、0 Height (km) - Estimated distribution of meteor trails 120 i 30 0843-04 7 The scattering properties of straight meteor ionization trails are strongly aspect sensitive. To be effective, it is necessary for the trails approximately to satisfy a specular reflection condition. This requires the ionized

42、 trail to be tangential to a prolate spheroid whose foci are at the transmitter and receiver terminals (see Fig. 2). The fraction of incident meteor trails which are expected to have usable orientations is approximately 5% in the area of the sky which is most effective. Figure 5 shows the estimated

43、percentages of useful trails for a terminal separation of 1 O00 km. It may be seen that the optimum scattering regions are situated about 100 km to either side of the great circle, independent of path length. The fraction of usable trails, p, for any path length, D, can be estimated using the follow

44、ing formula: Positions of regions of optimum scatter 4 37c D 3 (e2 - q2) - (1 - q2) (e2 - 1) (t2 - q2) - 4c2 h2/ - 4q2 (c2 - 1) h2 I p=- (16) (t2 - q2)2 (k2 - i) (52 - i) (e2 - q2) - 4c2 h2 I D212 where: 5 = (RI + R2)lD 11 = (RI - R2)ID. 8 An assessment of the link power budget of a meteor-burst com

45、munication link can be made using the average trail height and other expressions presented above. Once a link appears viable, a more detailed analysis is required to estimate the rate at which useful meteor-burst signals will be passed. The most rigorous methods of estimating useful burst rate typic

46、ally involve the following stages: a) b) c) d) Estimating the useful burst rate establish the minimum useful received signal power; utilize the equations of 4 5 to describe the variation of system parameters; compute the fraction of useful trails as a function of scattering position using equation (

47、i 6); combine the estimated true height distribution of meteor trails with equation (2) to compute the volume density of meteor trails, as a function of q, in the atmosphere; integrate the product of c) and d) over the meteor region using at each point qmin derived at b). e) COPYRIGHT International

48、Telecommunications Union/ITU RadiocommunicationsLicensed by Information Handling Services8 STD-ITU-R RECMN P-q3-1-ENGL 1997 m 4855212 0529775 431 m Rec. ITU-R P.843-1 FIGURE 5 Estimated percentages of useful trails as a function of scattering position for a terminal separation of 1 O00 km O 100 200

49、300 400 500 600 700 800 Distance from path mid-point along great circle (km) 0843-05 9 Antenna considerations The effect described in 7, together with the fact that the trails lie mainly in the height range 90-1 10 km, serve to establish the two “hot-spot regions towards which both antennas should be directed. The two hot-spots vary in relative importance according to time of day and path orientation. Generally, antennas used in practice should have beams broad enough to cover both hot-spots. Thus the performance is not optimiz