Single RF connection measurements

Your task

RF component characterization typically involves measuring modulation accuracy as well as matching or full S-parameter measurements. Modulation accuracy measures the transmission performance of a device with regard to certain parameters such as error vector magnitude (EVM) or bit error rate (BER). Characterization also determines whether the RF component complies with regulatory requirements such as out-of-band emissions and the adjacent channel leakage ratio (ACLR). Matching measurements ensure that the component will perform in its system as designed, e.g. the transmission of rated power at a given antenna impedance. Test time is always a critical parameter, so any increase in testing speed reduces costs.

Rohde & Schwarz solution

Modulation accuracy measurements require fully modulated signals to stimulate the device under test (DUT) so that it performs in the same way as it does in its real-world application. A wideband vector signal generator (VSG) in the test setup provides the input signal to the DUT.

Realistic performance metrics, such as EVM or BER require wideband vector signal analyzers (VSA) with a measurement application fitting the DUT’s application or set of applications (standard compliant or custom defined). Even though EVM can be estimated based on narrowband measurements, a standard compliant EVM, BER or realistic digital predistortion (DPD) generally requires a wideband VSA with a matching measurement application.

Regulatory measurements, such as the ACLR, often require only narrowband spectrum analyzers, which provide unmatched dynamic range. For some testing scenarios, higher testing speed is preferable to dynamic range. Combined signal and spectrum analyzers can provide the exact tradeoff between speed and dynamic range required for a given scenario. Matching and S-parameter measurements generally require the same tradeoff between dynamic range and speed as regulatory measurements.Vector network analyzers (VNA) provide precisely this flexibility.

Application

As mentioned above, RF component testing requires at least three different test and measurement functions (EVM, ACLR and S-parameters), often with different performance requirements. In addition, test time is always a critical parameter. Changing RF cabling is time-consuming, often requires manual interaction and is a common source of measurement errors.

Reducing the number of RF connections therefore significantly reduces test costs. The figure below shows a test setup that combines all these requirements:

• Single RF connection
• VNA for highly flexible matching measurements
• VSG/VSA combination for real-world measurements, including EVM, DPD and BER

Two off-the-shelf couplers complete the setup. Couplers can be selected according to dynamic range and frequency range requirements.

Coupler 1 connects the VSG and one VNA port to the input of the DUT, and coupler 2 connects the VSA and another VNA port to the DUT’s output. The through connections of the couplers are used for the modulated wideband measurements because they are most sensitive to signal-to-noise ratio (SNR). With their continuous wave (CW) stimulus, VNAs can compensate for lower SNR by reducing the filter bandwidth, for instance.

VNA measurements and calibration refer to plane 3/4 on the input side and plane 5/6 on the output side. Deembedding (e.g. for the external couplers) is available on both the VSG and the VSA.

Additional flexibility can be added to this setup by using three independent instruments: VSG and VSA could be high-end instruments with maximum bandwidth and performance, while the VNA measurements might only require a mid-class instrument or vice versa.

Summary

In summary, this setup is highly flexible with respect to testing requirements and provides the further advantage of a single RF interface to the DUT.