1. 1 RADAR UNWANTED EMMISSIONS A personal view J R Holloway All data in this presentation comes from public domain sources ITU WP 8B Radar Seminar September 2005 GENEVA

2. 2 Unwanted Emissions Any emission outside the necessary bandwidth of the transmission 1.152necessary bandwidth: For a given class of emission, the width of the frequency band which is just sufficient to ensure the transmission of information at the rate and with the quality required under specified conditions. 1.144out-of-band emission * : Emission on a frequency or frequencies immediately outside the necessary bandwidth which results from the modulation process, but excluding spurious emissions. 1.145spurious emission * : Emission on a frequency or frequencies which are outside the necessary bandwidth and the level of which may be reduced without affecting the corresponding transmission of information. Spurious emissions include harmonic emissions, parasitic emissions, intermodulation products and frequency conversion products, but exclude out-of-band emissions. 1.146unwanted emissions * : Consist of spurious emissions and out-of-band emissions.

3. 3 The Early Years Although radars were very high power they operated on frequencies that did not interfere with other services. For many years the only interference was between radars this by its nature could be controlled or mitigated Their waveguide systems made them virtually immune to interference from lower frequency systems As a consequence radars were left essentially alone from a regulatory point of view.

4. 4 Current Situation Spectrum seen as a valuable saleable resource Mobile systems have developed technology to allow operation in the microwave bands Issues There is now pressure to share the exclusive radar bands with commercial comms. In order to share the amount of unwanted emissions needs to be controlled. To share the radar emissions must be more tightly controlled As from 2003 there will be a progressive tightening of requirements for unwanted emissions.

5. 5 Definitions Reference Bandwidth (RB) The Calculated –40dB Bandwidth of the signal Necessary Bandwidth (NB) The minimum bandwidth required to operate the radar Spurious Emission Region (SE) The region outside 10 X RB Out of Band Emission Region (OOB) The region between the NB region and the SE region

6. 6 Transmitted Spectrum Regions

7. 7 Unwanted Emission Limits Before 2003 no SE limit for radar From 2003 new radars must meet Cat A or Cat B SE limits Cat A -60 dB Cat B -100 dB Class B being proposed to be adopted in Europe. OOB Definition of the extent by the emission masks Current Mask Design Aim Status of Limits SE levels part of radio regulations Boundary part of regulation OOB mask is a recommendation Design aim for new OOB 2006/2012

8. 8 Current Unwanted Emission Limits Cat A&B Cat A Cat B

9. 9 Design Aim When the OOB Mask was introduced a design aim was also introduced. This proposed to increase the Roll off to 40 dB/dec If this is not agreed then the aim falls JRG is considering what should replace the design aim

10. 10 Design Aim Unwanted Emissions Cat A&B Cat A Cat B

11. 11 Problems With Current Mask Mask perceived to be too relaxed at estimating –40 dB Bandwidth Mask perceived to be too relaxed in terms of Roll-off for trapezoidal pulses Magnetron Radars find it difficult to meet current mask Impossible to meet design aim

12. 12 Problems With Mask Mask perceived to be too relaxed at estimating –40 dB Bandwidth Mask perceived to be too relaxed in terms of Roll-off for trapezoidal pulses. Magnetron Radars find it difficult to meet current mask Impossible to meet design aim

13. 13 FM Modulated (Driven Systems) The equation for a FM trapezoidal is given as: For FM-pulse radars, the 40 dB bandwidth is: where A is 0.105 when K = 6.2, and 0.065 when K = 7.6 The term A/tr adjusts the value of B–40 to account for the influence of the rise time, which is substantial when the time- bandwidth product Bct, is small or moderate and the rise time is short.

14. 14 Sensitivity of Equation for Bw-40 FM Pulsed Bw-40dB gets large when tr 0 Bc gets large

15. 15 FM Trapezoidal Pulses vs Mask Mask 15 MHz Value 10 MHz

16. 16 Practical Bandwidths Measured MHz Calculated 16 MHz 3 dB BW20 dB BW40 dB BW3 dB BW20 dB BW40 dB BW 2.53.6102.57.825.7

17. 17 Problems With Mask Mask perceived to be too relaxed at estimating –40 dB Bandwidth Mask perceived to be too relaxed in terms of Roll-off for trapezoidal pulses. Magnetron Radars find it difficult to meet current mask Impossible to meet design aim

18. 18 Rectangular Pulses The mask is based on a 20 dB/dec roll off which is the natural roll off of rectangular pulse. The mask accurately reflects the theory of rectangular pulses Theory only Does not include practical distortions

19. 19 Trapezoidal Pulse Two roll-off rates 20 dB/dec 40 dB/dec 20 dB/dec 40 dB/dec

20. 20 Trapezoidal Pulse Break Points

21. 21 Problems With Mask Mask perceived to be too relaxed at estimating –40 dB Bandwidth Mask perceived to be too relaxed in terms of Roll-off for trapezoidal pulses. Magnetron Radars find it difficult to meet current mask Impossible to meet design aim

22. 22 Magnetron: Difficult to meet current OOB Limits Failure

23. 23 Coaxial Magnetron: Cat B Limits Failure Zones

24. 24 JRG Work on New Mask Looking into how a better estimate of the reference bandwidth. Non linear chirps Limit excessive bandwidths due to Large Chirps Fast Rise Times Looking into what roll-off can be practically achieved How Roll-off Relates to RB Looking into the special problems associated with. Magnetron based radars FM CW radars

25. 25 Example of the Application of NL Chirp

26. 26 Trade Off Reference Bandwidth vs Roll-off If the Reference Bandwidth is accurately calculated 20 dB roll-off looks achievable 40 dB roll-off looks difficult These are theoretical however in practice distortions make things worse

27. 27 Practical Issues To Reduce Unwanted Emissions Use High Compression ratios Use slow rise and fall times Shape pulses to remove discontinuities Use Filters

28. 28 Practical Issues cont Magnetrons Below rotation can use high Q filters Multi pulse length systems have to use a filter wide enough to meet narrowest pulse Above rotation systems have limited space OOB match of filters could upset Magnetron and cause more emissions Cost

29. 29 Practical Issues Filters Are Lossy can contribute twice TX & RX Can cause wild heat (active arrays) Can take up space Can cause oscillation out of band if not well matched Can distort want signal if too narrow Limit the peak power due to arcing Costly

30. 30 Practical Issues Linear Beam Tube Transmitters Can use moderate compression ratios Difficult to control rise and fall times Single channel systems can use High Q channel Filters Agile systems can only use band limiting filters See Illustration

31. 31 Example Channel Filter S Band Radar 2700 to 2900 MHz Radar 4 MHz RB Currently meets mask ( has 20 dB/dec roll- off) Filter needed to meet new 40 dB/dec Tuned 2800 MHz Attenuation required 20 dB @ 2820 MHz

32. 32 Channel Filter

33. 33 Agile Band Limiting Filter S Band Radar 2700 to 2900 MHz Radar 4 MHz RB Currently meets mask (has 20 dB/dec roll-off) at band edge Filter needed to meet new 40 dB/dec at band edge Frequency Agile radar 200 MHz Attenuation required 20 dB @ 2918 MHz

34. 34 Band Limiting Filter

35. 35 Practical Issues cont: Solid State Lumped Transmitters Can use higher compression ratios Easier to control rise and fall times (slow down) Single channel systems can use High Q channel Filters Agile systems can only use band limiting filters of High Q

36. 36 Practical Issues cont: Solid State Distributed Transmitters Can use higher compression ratios Easier to control rise and fall times Agile systems can only use band limiting filters with a moderate Q

37. 37 Practical Issues Active Array Systems Can use very high compression ratios Difficult to control rise and fall times Agile systems can only use band limiting filters of very low Q Or Low pass filters

38. 38 Illustration: Solid State ATC Can make use off Fixed Operating Frequencies Long pulses Slow rise & fall times Many radar applications cannot make use of all these advantages

39. 39 Solid State ATC radar

40. 40 Conclusions to Date Currently there is some scope for improving the mask Solid State systems are better than linear beam devices and cross field devices Larger time bandwidth products There some scope for pulse shaping in Solid State transmitters OOB Filters are effective for fixed frequency systems Agile systems are more problematic Limited scope for OOB control OOB control not realistic in active arrays

41. 41 END Thank you John Holloway