ITU-R M 1314-1-2005 Reduction of unwanted emissions of radar systems operating above 400 MHz《ITU-R建议M 1314的草案修订版-运行在400 MHz以上的雷达系统的无用发射的减少》.pdf

《ITU-R M 1314-1-2005 Reduction of unwanted emissions of radar systems operating above 400 MHz《ITU-R建议M 1314的草案修订版-运行在400 MHz以上的雷达系统的无用发射的减少》.pdf》由会员分享，可在线阅读，更多相关《ITU-R M 1314-1-2005 Reduction of unwanted emissions of radar systems operating above 400 MHz《ITU-R建议M 1314的草案修订版-运行在400 MHz以上的雷达系统的无用发射的减少》.pdf（9页珍藏版）》请在麦多课文档分享上搜索。

1、 Rec. ITU-R M.1314-1 1 RECOMMENDATION ITU-R M.1314-1*Reduction of unwanted emissions of radar systems operating above 400 MHz (Question ITU-R 202/8) (1997-2005) Scope This Recommendation provides information on the design factors affecting unwanted emission characteristics of radar transmitters to b

2、e taken into account during the design of radars. It also recommends certain types of transmitter output devices that should be used when practicable to minimize unwanted emissions. The ITU Radiocommunication Assembly, considering a) that the radio spectrum available for use by the radiodeterminatio

3、n service is limited; b) that the radionavigation service is a safety service as delineated in No. 4.10 of the Radio Regulations (RR), and in addition that some other types of radar systems such as weather radars may perform safety-of-life functions; c) that the necessary bandwidth of emissions from

4、 radar stations in the radiodetermination service is large in order to effectively perform their function; d) that new emerging technology systems may use digital or other technologies that are more susceptible to interference from radars unwanted emissions due to their high peak envelope power; e)

5、that the ITU-R has been studying the question of efficient use of the radio spectrum by radar systems; f) that unwanted emissions from radar systems may in some cases cause interference to systems in other radio services operating in the adjacent and harmonically related bands; g) that RR Appendix 3

6、 specifies maximum permitted power levels for spurious or spurious domain emissions, and that Recommendation ITU-R SM.1541 specifies out-of-band limits for radiodetermination radars, recommends 1 that the information on radar transmitter design factors affecting unwanted emission characteristics of

7、radars contained in Annex 1 should be used to reduce unwanted emissions; 2 that, when practical, the best available output device technology should be used in radars to reduce non-harmonic radar spurious emission levels; *This Recommendation should be brought to the attention of the International Ma

8、ritime Organization (IMO), the International Civil Aviation Organization (ICAO), the International Maritime Radio Committee (CIRM), and the World Meteorological Organization (WMO) and Radiocommunication Study Groups 1 and 9. 2 Rec. ITU-R M.1314-1 3 that, when necessary and when possible, radar outpu

9、t filters should be used to reduce radar unwanted emissions. Annex 1 Reduction of unwanted emissions of radar systems 1 Introduction To maximize future efficiency of spectrum use, radar transmitters should be chosen, designed, and constructed such that the emission spectrum falls off as rapidly as p

10、ossible, given the constraints on radar performance, size, cost, weight, reliability, maintainability, etc. The emission-spectrum skirt fall-off rate (out-of-band emission characteristics), and the emission floor level (spurious emissions) are determined by the transmitter hardware and architecture

11、and by the transmitted waveform. Those influences are discussed below. 2 Radar design factors The function or mission of a radar largely determines the design of the radar. Radar missions are widely varied (such as: navigation, weather observation, wind velocity determination, surveillance, imagery

12、and mapping, terrain following, altimeter, etc.) and generally require unique performance specifications. These missions determine some parameters that are not under the control of the radar designer which directly impacts upon radar design factors such as: required transmitter power, transmitter wa

13、veform selection, transmitter output device selection, antenna gain, receiver sensitivity, range and azimuth resolution, and Doppler coverage. The judicious trade-off of radar design factors to improve emission spectrum control is key in enhancing compatibility between radar systems and other servic

14、es. 3 Waveform selection and shaping The choice of pulse waveform type and the way in which the waveform is shaped can also have important influences on spectrum control and hence on compatibility. Most radars, especially those using a single power oscillator or power amplifier, are constrained by c

15、onsiderations of energy efficiency and heat dissipation to use pulses having essentially constant-amplitude except during brief transitions between subpulses. That limits the types of waveforms that can be chosen. Even when that constraint applies, however, choices remain that can have a major effec

16、t on the emission spectrum. Radar waveforms can be categorized, at the first level, into plain-pulse, or unmodulated-pulse, waveforms (having the emission designator of “P0”) and intra-pulse-modulated waveforms. Intra-pulse modulation usually serves as a means for implementing pulse compression, alt

17、hough an exception occurs in the case of waveforms used to drive frequency-steered arrays. Intra-pulse modulations can thus be divided further into the following subcategories: continuous FM, or “chirp” pulses; stepped-chirp pulses; Rec. ITU-R M.1314-1 3 stepped-frequency pulses used in frequency-st

18、eered radars; discretely-coded pulses. From the standpoint of emission-spectrum control, a guiding principle in selecting and shaping a waveform is to remove discontinuities in as many derivatives of the waveform as possible, since that determines the ultimate spectrum fall-off slope, in dB/decade o

19、f frequency offset, that is achieved. The various pulse waveforms are therefore distinguished by the differences among their transitions of amplitude, phase, and frequency within the pulse. All pulse waveforms, of course, contain rise and fall ramps on the overall envelope. Other things being equal,

20、 it is desirable to have gradual and smooth rise and fall ramps. However, other things are not always equal. In particular, pulses generated in crossed-field devices require quick rise ramps to avoid excitation of spurious oscillatory modes that would worsen the spectrum. When amplifiers other than

21、crossed-field devices are used, smooth, gradual rise ramps are helpful to spectrum control when they can be implemented. Such implementation might still be difficult because power-amplifier dissipation is usually high when the amplifiers are not driven close to saturation; that can motivate use of f

22、ast rise and fall ramps even when spurious oscillations are not a concern. Continuous frequency modulation, or chirp, waveforms with high pulse-compression ratio, or bandwidth-pulsewidth product, have very steep spectrum fall-off rates. This applies to both linear FM and nonlinear FM waveforms. The

23、main contribution to undesired spectral components of these waveforms arise from the use of short rise ramps on the pulses. Stepped-chirp waveforms have piecewise-constant frequencies that increment or decrement monotonically throughout the pulse. They can be considered as a subset of continuous-fre

24、quency-modulation chirp. However, stepped-chirp waveforms, as well as non-monotonic stepped-frequency waveforms that are used with frequency-steered antenna arrays, have poorer emission spectra than continuous-FM chirp waveforms have. This is a consequence of discontinuities in the waveform. It migh

25、t be feasible to remove those discontinuities by implementing the stepped chirp in a way that maintains continuity of phase at the junctions between frequency steps. Even if that is so, however, discontinuities in the first derivative, which do not occur in true continuous-FM waveforms, will remain,

26、 so the spectrum will not be as good as that of a continuous-FM pulse with comparable pulse-compression ratio. There are also certain polyphase-coded waveforms, of which the Frank polyphase-coded waveform is the prototype, that effectively approximate chirp waveforms; i.e., they approximate “continu

27、ously-coded” waveforms1. However, these contain abrupt steps in phase, so their spectra do not fall-off nearly as steeply as those of continuous-FM chirp waveforms. Herein, discretely-coded radar waveforms refer to those that do not resemble continuous-FM waveforms in any way. Since that excludes th

28、e polyphase codes, most discretely-coded radar waveforms can be subdivided into bi-phase-coded and frequency-coded types. Waveforms in either of these categories can use Barker codes and pseudo-random binary sequence codes. In the absence of refinements, discretely phase-coded waveforms have abrupt

29、transitions between constant-phase “chips”. (The same is true of Frank codes and other polyphase codes.) As a consequence, their spectra fall off at only 20 dB/decade. However, some options are available that can improve the spectra of phase-coded waveforms. In principle, the spectrum of the RF driv

30、e (excitation) waveform can be made to fall-off arbitrarily fast by filtering the modulating waveforms or the modulated low-level drive waveforms (either IF or RF) themselves. However, those gains can be eroded in practice by spectral regrowth that occurs in both the transmitter power amplifier and

31、in receivers in the environment. When premodulation 1Chirp waveforms are sometimes referred to as coded waveforms even though their “coding” is not discrete. 4 Rec. ITU-R M.1314-1 filtering is used, the chip-to-chip transitions are gradual instead of abrupt, but on bi-phase waveforms and those polyp

32、hase waveforms that contain 180 phase transitions, nulls or dimples remain in the waveform envelope because that envelope passes through zero during transitions from one phase to the other. That is not a problem in itself, but the advantages gained are reduced by two factors. One factor is AM-to-PM

33、conversion that occurs in power-amplifier devices. The extraneous phase modulation that results widens the spectrum. Another disadvantage is that any limiting that occurs in either the power-amplifier transmitter stages or in victim receivers tends to reintroduce abrupt transitions into the dimpled

34、waveform. Those abrupt steps translate into unwanted spectral sidebands with spectrum skirts that again fall at only 20 dB/decade. It is possible to mitigate that spectral regrowth to a considerable extent. This can be done by constructing exciter (low-level, driver) waveforms that maintain a nearly

35、 constant envelope not only during the subpulse dwell intervals but also during phase transitions. In such waveforms, 180 phase transitions consist of rotations of carrier phase through a semicircle in the I-Q, or real-imaginary plane instead of movements along the I or Q axis that pass through the

36、origin. This can be implemented by means of quadrature modulators and suitable waveform-shaping circuitry. An alternative discretely coded waveform category is continuous-phase frequency-shift keying. These waveforms are essentially the same as so-called minimum-shift-key (MSK) waveforms used in som

37、e communication systems. Although sometimes referred to as phase-shift-keyed waveforms, these are really frequency-coded because the phase changes continuously while, in their basic unfiltered form, the instantaneous frequency changes abruptly and remains constant throughout each subpulse. There are

38、 no discontinuities in the waveform itself, but there are discontinuities in the first derivative. Consequently, the spectra approach asymptotes that fall at a rate of 40 dB/decade. Furthermore, these waveforms have constant envelope even during their subpulse transitions, so they are intrinsically

39、immune to the spectral-regrowth problems that occur with phase-coded waveforms. (Since swept-frequency waveforms have no subpulses, they too are immune from spectral regrowth due to limiting and AM-to-PM conversion.) In communication systems, premodulation-filtering of MSK waveforms is widely used.

40、It is expected that such filtering could also be applied in radars, in which event the emission spectrum fall-off would theoretically become steeper than 40 dB/decade. While steep fall-off of the emission spectrum is desirable, it cannot be pursued without regard for the consequences in range resolu

41、tion and Doppler coverage, usually expressed by the shape of the “ambiguity function”. That function represents the magnitude of the output signal evoked by return from a point target and produced by a filter matched to the transmitted signal. The ambiguity function is a function of both the range (

42、time delay) and the Doppler shift of the target return. As one extreme example, a linear-FM rectangular pulse waveform with infinite time-bandwidth product (i.e., infinite compression ratio) would have a perfectly rectangular spectrum, except for the contribution from the rise and fall ramps. But th

43、e response of a matched filter to such a waveform would have a sin(t)/t response for a constant-Doppler target return. Such a response has time (i.e., range) side-lobes only about 13 dB below the main response, which is inadequate in some applications that require a high degree of multiple-target re

44、solution. The matched-filter response is not simply the Fourier transform of the emission spectrum. However, there is a tendency for abrupt fall-off of emission spectrum to be accompanied by high range side-lobes in the response, much as abrupt steps in the time waveform are accompanied by high side

45、-lobes in the emission spectrum. To some extent, range side-lobe suppression can be improved by mismatching the receiver signal processor to the transmitted pulse, but this incurs a loss of sensitivity relative to that of a matched filter. It is therefore necessary to choose a waveform that makes a

46、good trade-off among spectrum control, range side-lobe suppression, and sensitivity. (Swept-frequency waveforms using a slightly non-linear FM profile are good compromises for some applications.) In general, however, the need for good resolution and sensitivity narrows the designers options. In addi

47、tion, many radar applications require nearly uniform response over a substantial span of Doppler frequencies; i.e., they are required to have low “Doppler sensitivity”. This introduces another constraint on the designers choice of waveform. Rec. ITU-R M.1314-1 5 In communication systems, the improve

48、ment of spectrum fall-off that is gained by premodulation filtering comes at the expense of worsening inter-symbol interference. Nevertheless, considerable improvement in spectrum control can often be achieved before intersymbol interference becomes unacceptable. In a radar, the spectrum improvement

49、 that can be gained by use of premodulation filtering comes at the cost of degrading the radars resolution. It is also to be expected that a slight loss of detection sensitivity will be incurred due to the difficulty of constructing a perfectly matched filter (or correlation process) for waveforms containing rounded corners (resulting from premodulation filtering) instead of sharp discontinuities. As with the analogous communication-system case, however, it is reasonable to expect that considerable improvement in spectrum control can often be achieved before the ambiguity func