What is the difference between a far field and a near field antenna measurement system?
Far-Field vs Near-Field Antenna Measurement
The choice between far-field and near-field measurement depends on the antenna size, frequency, required pattern accuracy, and available facilities.
| Parameter | SOLT Cal | TRL Cal | eCal |
|---|---|---|---|
| Accuracy | Good | Excellent | Good-very good |
| Standards Needed | 4 (S,O,L,T) | 3 (T,R,L) | 1 (module) |
| Bandwidth | Broadband | Band-limited | Broadband |
| Setup Time | 5-10 min | 10-20 min | 1-2 min |
| Best For | Coaxial, general | On-wafer, waveguide | Production, speed |
Calibration Procedure
(1) Outdoor elevated range: the AUT and source antenna are mounted on towers or rooftops at sufficient height to avoid ground reflections. The separation distance > 2D^2/lambda. Advantages: unlimited quiet zone (any size antenna), wideband (no reflector or probe bandwidth limitations). Disadvantages: weather-dependent, ground and environmental reflections, security concerns (the antenna pattern is visible to outside observers), and the long distances require high-power sources or sensitive receivers. (2) Indoor far-field range (anechoic chamber): the chamber length must be > 2D^2/lambda. For a 30 cm antenna at 10 GHz: 2D^2/lambda = 6 m (feasible). For a 1 m antenna at 10 GHz: 2D^2/lambda = 67 m (impractical for most indoor facilities). Indoor ranges are limited to small/medium antennas or high frequencies. (3) Compact antenna test range (CATR): uses a reflector to create a plane wave in a short distance (5-15 m for most CATRs). The effective far-field distance is determined by the reflector quality, not the chamber length. Suitable for medium to large antennas at all frequencies.
Error Sources
(1) Planar near-field: the probe scans a flat plane in front of the AUT. The scan area must be larger than the AUT aperture + an overscan margin (to capture the main beam and first sidelobes). Overscan: the scan plane extends beyond the AUT edge by D_scan = z × tan(theta_max), where z is the probe-to-AUT distance and theta_max is the maximum pattern angle of interest. For theta_max = ±60° and z = 10 cm: D_scan = 10 × 1.73 = 17 cm beyond each edge. Scan step: lambda/2 (Nyquist sampling). At 10 GHz: step = 15 mm. At 60 GHz: step = 2.5 mm. At 300 GHz: step = 0.5 mm. Number of scan points: for a 30 × 30 cm AUT at 60 GHz with ±60° coverage: scan area ≈ 60 × 60 cm, step = 2.5 mm. Points = (600/2.5)^2 = 57,600 points. At 1 point per second: 16 hours of scanning. This is why planar near-field at mmWave is very slow. (2) Spherical near-field: the probe moves on a sphere surrounding the AUT. This captures the full 3D pattern (including back lobes). The AUT rotates in azimuth while the probe arm rotates in elevation (or vice versa). Scan step: lambda/2 in both theta and phi. Number of points: for a full sphere at 10 GHz: (360/1.5) × (180/1.5) ≈ 28,800 points (where 1.5 cm = lambda/2). Spherical near-field is the most complete measurement (full 3D pattern) but the slowest. (3) Cylindrical near-field: combines a linear probe scan (z-axis) with AUT rotation (phi). Good for antennas with 360° azimuth coverage and limited elevation pattern (e.g., omnidirectional antennas, base station antennas).
- Performance verification: confirm specifications against the application requirements before finalizing the design
- Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
- Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades
- Interface compatibility: verify impedance, connector type, and mechanical form factor match the system architecture
- Margin allocation: include sufficient design margin to account for manufacturing tolerances and aging effects
Fixture Considerations
Far-field: accuracy limited by range reflections, ground bounce, source antenna pattern knowledge, and path loss calibration. Typical pattern accuracy: ±0.5-1 dB for the main beam, ±2-3 dB for sidelobes. Near-field: accuracy limited by probe calibration, positioning accuracy, truncation effects, and the NFFT algorithm. Typical pattern accuracy: ±0.2-0.5 dB for the main beam, ±1-2 dB for sidelobes (better than far-field for high-precision measurements). CATR: accuracy limited by reflector quality and absorber performance. Typical: ±0.3-0.7 dB for the main beam, ±1-2 dB for sidelobes.
Frequently Asked Questions
Which method should I use for a 5G mmWave antenna?
For 5G mmWave (28/39 GHz) phased-array antennas (typical size 5-15 cm): (1) CATR: recommended by 3GPP for 5G NR FR2 OTA testing (TS 38.521-3). Far-field distance: 2 × 0.15^2 / 0.011 = 4 m (at 28 GHz). A CATR with 2-3 m reflector provides a quiet zone of 30-50 cm, sufficient for most 5G devices. Measurement speed: fast (direct far-field). (2) Planar near-field: suitable for high-gain, directive 5G antennas. Scan step: 5.4 mm at 28 GHz. For a 20 × 20 cm scan area: 1,369 points per cut (fast). The NFFT provides the full 3D pattern from a single planar scan (limited to the forward hemisphere). (3) Spherical near-field: provides the most complete pattern but slowest. Used for full characterization during development.
How do I measure a large antenna like a satellite dish?
For a 3 m satellite dish at 12 GHz: far-field distance = 2 × 9 / 0.025 = 720 m. Options: (1) Outdoor far-field range: mount the dish on a tower and use a source at > 720 m distance. Practical for permanent antenna test facilities. (2) CATR: requires a reflector > 4 m × 4 m. Such large CATRs exist (e.g., SELEX Galileo 8 m CATR, CNES 6 m CATR) but are very expensive. (3) Spherical near-field: mount the dish on a spherical positioner and scan with a probe at 1-2 m distance. The dish is large and heavy, requiring a heavy-duty positioner. Measurement time: many hours. (4) Satellite beacon measurement: for in-service satellite antennas, measure the pattern by tracking a satellite beacon and rotating the dish. This gives the actual operational pattern but only at the beacon frequency.
What is probe compensation in near-field measurement?
The near-field probe (sampling antenna) has its own radiation pattern that affects the measured near-field data: the probe preferentially samples field components aligned with its polarization and within its beamwidth. Probe compensation: the NFFT algorithm accounts for the probe pattern by: (1) Measuring or knowing the probe pattern (supplied by the manufacturer or measured separately). (2) Including the probe pattern in the transformation: the far-field pattern of the AUT = NFFT(measured_near_field) / probe_pattern. Without probe compensation: the measured pattern is convolved with the probe pattern, smearing the AUT pattern and reducing angular resolution. Error without compensation: typically 0.5-2 dB in sidelobes and 0.1-0.3 dB in the main beam. For standard gain horns (common probes): the probe pattern is well-known and the compensation is accurate.