Why OTA is Needed
OTA (Over-The-Air) refers to testing wireless device performance without cables, directly through antennas under over-the-air conditions. Even if a cellular module shows good conduction indicators on the evaluation board, once enclosed in a case, factors like antenna efficiency, human body blockage, metal structure reflection, and multipath environments can significantly change the actual link. Radiation testing can measure TRP (Total Radiated Power) and TIS (Total Isotropic Sensitivity), which are common requirements for operator certification and high-end IoT devices.
Typical Test Environment
A fully anechoic chamber (FAC) or reverberation chamber is used to create controlled multipath or statistical environments. Test equipment rotates on a turntable to sample performance from multiple directions, simulating random orientations in real use. Multi-probe systems can quickly scan spatial distribution. Compared to conduction testing, OTA is more expensive and time-consuming but can better detect whole-device integration issues, such as antenna clearance being occupied by screws or frequency shifts caused by changes in the dielectric constant of plastic enclosures.
Relation to IoT Scenarios
Fixed industrial gateways and handheld devices have different sensitivity to antenna patterns: weak coverage in indoor corners relies more on uplink margins, and when TRP is insufficient, the dropout rate increases. Vehicle-mounted terminals need to consider how glass and metal car bodies reshape antenna directivity. Low-power wide-area network (LPWAN) devices with limited uplink power can benefit from OTA surveys to determine antenna type and installation orientation.
Project Coordination
During the R&D phase, it is recommended to establish a dual benchmark of "Conducted + OTA": Conducted tests are used to locate RF front-end issues, while OTA is used to verify full-device integration. On production lines, couplers are usually used instead of far-field OTA for quick screening, but correlation with OTA gold samples should be established. After changing metal components such as batteries, speakers, or displays, OTA should be re-evaluated.
Engineering Practice Supplement
Starting from "Radiation and OTA," common on-site issues often focus on "lab-compliant metrics are difficult to replicate in the field." Besides RF and protocols, power-ground bounce, connector contact, and firmware race conditions should also be checked. On the cellular side, it is recommended to collect RSRP/SINR, cell reselection, and PDP reconstruction records; on the application side, clear strategies for retry, idempotency, and queue persistence should be in place. When data security and compliance are involved, the scope of collection, storage location, retention period, and third-party sharing should be specified in privacy and contract annexes to avoid high costs of patching after deployment.
Spherical Average Power and a Null in One Direction
A radiated test looks at what the whole device sends into space from the antenna and what it can receive back from space. It does not go through an RF cable. Total radiated power is the average of transmit power over directions. Total isotropic sensitivity combines, over directions, the level at which reception is still correct. One direction can be very strong and the rest can have a null, and the average can still pass. If the device always faces the base station with that null, it drops in the field. A conducted test cables the module to the instrument. It can find problems in the chip and the match. It cannot show how the housing, screws, and a human body twist the pattern. OTA is a test after integration. It is not a repeat of the module's factory figures.
The condition of use is a whole device close to the state that will be sold: housing, battery, cables, and mounting bracket in place, and the antenna keep-out not blocked by temporary copper foil. A chamber or a reverberation room makes the environment controllable. A turntable's samples deal with direction being unknown. A handheld product and a fixed installation should not share one pass criterion on the pattern. After a display, a speaker, or a metal bracket changes, the old pattern is void, even if the circuit did not change.
When the Production Coupler and the Chamber Gold Sample Diverge
The usual break is a production screen that uses a coupling fixture sensitive to only one face. A null that fails in the chamber always passes on the fixture. Without a correlation curve between the fixture and the chamber gold sample, the line limit is set by feel. The second miss is to report only an average at one frequency, so a keep-out problem on a high band is covered by a good result on a low band. The third is to infer radiated pass from VSWR pass. VSWR says only how the port reflects. It does not say efficiency. Loss can become heat instead of radiation. If a body phantom or a handheld pose is not placed the way the product is actually worn, the over-the-air result for a device used close to the body will not match the certification posture.
Close to the body, also check whether absorbed energy exceeds the limit; see SAR and exposure limits. If poor radiation comes from a power template that was never calibrated, go back to the line's frequency data first; see production RF calibration. Do not change the mechanical parts first.
A report should at least show the directional distribution in each band, not only one omnidirectional average. If the average passes and one mounting face has a null, judge again in the actual fixed pose. A fixed gateway does not turn at random the way a handheld device does. The pass limit of a coupling fixture should be derived by comparing the same gold-sample lot in the chamber and on the fixture. If the fixture reading drifts, check the fixture and the cable before judging the product. After a battery, a display, or a metal bracket is replaced, sample the over-the-air result again even if the schematic did not change, because keep-out and ground both changed. VSWR, port power, and over-the-air power are three columns. They can explain each other. They cannot stand in for each other: a good VSWR can mean energy turned into heat, and enough port power can still fail to radiate. Put the three columns and the mechanical revision in one record. Before certification, that is what shows whether to change the match, the calibration, or the housing. If one column is missing, do not infer an over-the-air pass from another column's pass.