Presentation on theme: "Interference Effects and Interference- Limit Criteria for Radar Receivers IEEE Electromagnetic Compatibility Conference Ft. Lauderdale, FL, 26 July 2010."— Presentation transcript:
Interference Effects and Interference- Limit Criteria for Radar Receivers IEEE Electromagnetic Compatibility Conference Ft. Lauderdale, FL, 26 July 2010 John Carroll, Frank Sanders, Robert Sole Institute for Telecommunication Sciences, NTIA, U.S. Department of Commerce, Boulder, CO And Office of Spectrum Management NTIA, U.S. Department of Commerce, Washington, DC ) ( ) ( ) ( ) ( ) ( )
NTIA Involvement Radar systems have historically been allocated in bands where they do not share with communication systems. But in recent years proposals have been made for communication systems to operate in radar bands. These include: Satellite Signals; Radio local area networks (RLANS); Various new mobile radio systems including WiMax; These proposals have raised the question: At what thresholds do various types of radio interference degrade the performance of radar receivers? Institute for Telecommunication Sciences – Boulder, Colorado
Sharing proposals may directly affect US Government radar systems. Therefore it is critical that technical parameters for sharing between radar receivers and non-radar systems be accurately quantified. NTIA Involvement Starting in FY-2002, NTIA Office of Spectrum Management and NTIA Institute for telecommunication Sciences undertook a joint effort to determine thresholds for interference to radar receivers. PROBLEM: Lack of quantitative data regarding thresholds at which various types of interference degrade performance of radar receiver systems. Airport Surveillance Air-Search Radar
Institute for Telecommunication Sciences – Boulder, Colorado Communication receivers typically experience interference effects as a function of the ratio of the interference level to the level of the desired signal (S/I or C/I). Radar Interference Criterion: I/N Fixed ground-based weather surveillance radar display in the presence of strong interference. Radar receivers, in contrast, normally operate against their internally generated noise. They are noise limited and thus the critical interference level parameter is the ratio I/N within the radar receiver IF stage. But what criteria should be used to determine radar receiver performance at a given I/N level?
Institute for Telecommunication Sciences – Boulder, Colorado OSM and ITS decided to assess radar performance on the basis of probability of detection (P d ) of desired targets as a function of interference level (I/N) in the radar IF stage. Radar Performance Criterion: Probability of Detection of Controlled Targets Air surveillance radar ppi display with internally generated targets Target levels would have to be controlled during measurements. Therefore the targets had to be generated and injected along with interference. Baseline P d needed to be marginal: targets strong enough to be easily observed, but nevertheless at the edge of radar’s capability. Decision was made to set baseline P d of desired targets at P d = 90%. Then interference effects would be measured relative to that performance level.
Institute for Telecommunication Sciences – Boulder, Colorado Although procedures have varied slightly from radar to radar, the essential procedure that NTIA (OSM and ITS) has used is: Radar Interference Measurement Procedure OSM engineer preparing a maritime radar for signal injection at Curtis Bay Coast Guard Station near Baltimore. 1) Disconnect radar receiver and inject desired (controlled) targets at the RF stage so that they are handled the same way as regular returns. 2) Adjust target level until P d is as close as possible to 90%. 3) Inject interference signals at the radar RF stage. For each interference modulation, start at I/N=-12 dB, and successively raise the level to: -12 dB, -10 dB, -9 dB, -6 dB, -3 dB, 0 dB, +3 dB, +6 dB, on up to as much as +60 dB if necessary. Interference types tested include: CW; pulsed CW; CDMA; TDMA; BPSK; OFDM; and UWB; and a variety of simulated radar emissions. Not all modulations have been used on all radars. 4) Count at least 200 desired targets to obtain P d for each level of interference.
Institute for Telecommunication Sciences – Boulder, Colorado Typical Radar Interference Testing Block Diagram Note: Some live-target tests have also been performed.
Institute for Telecommunication Sciences – Boulder, Colorado Typical Setup (DoD/FAA/DHS Radar)
Institute for Telecommunication Sciences – Boulder, Colorado Radars Types Measured ( ) Several maritime radars in S and X-bands (in both the US and the UK); Airport surface detection radar in X-band; Two models of DoD/FAA/DHS long-range radars in L-band; Airport surveillance radars in S-band. Weather surveillance radars in S and C bands. Airborne weather surveillance radar in X-band. Precision approach radar in X-band. Wind profiler radar in UHF.
Institute for Telecommunication Sciences – Boulder, Colorado Example of P d Data Curves
Institute for Telecommunication Sciences – Boulder, Colorado Target Losses at Low I/N Levels Target losses at low I/N levels are insidious because there are no ancillary effects to indicate that interference is occurring or that targets are being lost.
Institute for Telecommunication Sciences – Boulder, Colorado Summary of Results to Date Interference at high duty cycles (above about 2%), such as from communication signals, typically causes target losses to begin at I/N levels between -10 to -6 dB.; Interference at low duty cycles (less than 2%), such as from other radars, can often be sustained at I/N levels as high as +30 dB to +60 dB without degrading receiver performance. Target losses due to low levels of interference are insidious because no visible effects are associated with the interference. The targets simply fade away. Target losses can occur at any range. Losses do not just occur at the edge of radar coverage, but rather anywhere that the targets are close to radar receiver noise.
Institute for Telecommunication Sciences – Boulder, Colorado Summary of Results to Date (cont.) Meteorological radars show effects at the lowest levels: Wind profilers show effects at I/N levels as low as -18 dB; Interference thresholds go lower as the number of pulses used to integrate radar targets increases. Target loss curves (P d as a function of I/N) are analytically predictable based on the number of pulse echoes used to integrate targets. ITU-R criteria of -6 dB I/N threshold for interference analyses are barely adequate for many radar types.
Institute for Telecommunication Sciences – Boulder, Colorado Further Reading: Sanders, Sole, Bedford, Franc, and Pawlowitz: Effects of Interference on Radar Receivers, (NTIA Technical Report TR ) Available: