What is the difference between transmitter-side and receiver-side radio tests?

Transmitter-side tests measure what the product radiates: maximum power, OBW, spectrum emission mask (SEM), ACLR, spurious emissions. Receiver-side tests measure what the product tolerates in the presence of an interfering signal: reference sensitivity, adjacent channel selectivity (ACS), in-band and out-of-band blocking, IM3 intermodulation, spurious response. Under RED article 3.2 both families are mandatory to declare conformity. Under FCC Part 15.247 and 15.249, TX tests are the rule, and RX tests can be required through KDB 558074 for certain product classes.

What is the receiver "spurious response"?

The receiver spurious response refers to the frequencies outside the useful band at which the receiver reacts as if they were in band, for instance through an intermediate frequency image, an internal mixing product, or an local oscillator harmonic. A strong signal at these frequencies degrades sensitivity or generates false detections. ETSI EN 300 328 (Wi-Fi/BLE 2.4 GHz), EN 300 220-1 (sub-GHz SRD) and 3GPP TS 38.521-1 tests quantify the spurious response by imposing a threshold on the tolerated interferer level before degradation.

How is receiver blocking measured?

The bench includes a wanted signal generator set on the tested channel, an interferer generator set on an offset frequency, an RF combiner, the DUT and an error-rate measurement device (BER, PER or BLER depending on the protocol). The wanted signal is set 3 dB above reference sensitivity, then the interferer is raised until the error rate crosses the pass threshold (typically BER 0.1% for SRD protocols, BLER 5% in 3GPP). The pass threshold and interferer frequencies are defined in the applicable standard (ETSI EN 300 328 clause 5.4.6 for Wi-Fi/BLE for example).

What does ACS (Adjacent Channel Selectivity) measure?

ACS measures the receiver's ability to receive a wanted signal on its nominal channel in the presence of a modulated signal of the same nature on the immediately adjacent channel. The result is expressed as a ratio in dB between the tolerated adjacent interferer level and the wanted signal level, for a target error rate. A low ACS signals insufficient RF filtering or a limited-quality mixer. The ACS threshold is defined in 3GPP TS 36.521-1 and TS 38.521-1 per band and channel bandwidth, and in the applicable ETSI standards for Wi-Fi and SRD.

Why test IM3 intermodulation?

When two strong interferers at frequencies f1 and f2 are present at the receiver input, its non-linearity generates mixing products. The third-order product 2*f1 - f2 (and its symmetric 2*f2 - f1) can fall inside the useful band and disrupt reception. The IM3 test verifies robustness to this scenario, common in a dense multi-radio environment (IoT gateway colocated with a Wi-Fi access point, smartphone in the presence of neighbour base stations). Thresholds appear in 3GPP TS 36.521-1, TS 38.521-1, ETSI EN 300 328 and EN 300 220-1.

Does 5G FR2 mmWave force an OTA methodology for these tests?

Yes. 3GPP TS 38.521-2 mandates OTA measurement in an anechoic chamber for FR2 tests (bands n257, n258, n260, n261, n262 above 24 GHz), because mmWave antennas are integrated within the RFIC module and no externally accessible antenna port exists at these frequencies. Blocking, ACS and IM3 tests are then performed with OTA sources oriented towards the UE, rigorous chamber calibration and precise angular positioning. The FR2 OTA fleet remains small, which becomes a calendar factor for mmWave products.

How do TX-side OBW and SEM relate to RX tests?

OBW (Occupied Bandwidth) and SEM (Spectrum Emission Mask) are TX tests that constrain how much the product radiates out of channel. On the RX side, blocking and ACS constrain how much the product tolerates from neighbouring radios. A realistic multi-radio environment is compliant if each product respects its TX limits (OBW, SEM, ACLR) and each product tolerates the interferers expected by the RX standards. The TX/RX pair is complementary and inseparable in ETSI and 3GPP frameworks.

Pre-compliance vs full-compliance, what changes for these tests?

Pre-compliance refers to tests run in-house or in a non-accredited lab with a simplified bench (often a signal generator, a spectrum analyser and a combiner), to reduce the risk of failure before the official campaign. Full-compliance refers to tests run in an ISO/IEC 17025 accredited lab recognised by the target scheme (RED notified body, FCC TCB, PTCRB or GCF lab), with traceable instrumentation and complete procedures. Only the full-compliance report is admissible in the final conformity file.

Radio: RX blocking, selectivity and intermodulation tests

Author Hugues Orgitello Last updated 27 May 2026 ~13 min read

Guide · Receiver-side radio tests

Receiver-side radio tests quantify what the product tolerates in the presence of interfering signals: blocking by a high-level out-of-channel signal, adjacent channel selectivity (ACS), IM3 intermodulation generated by two simultaneous interferers, spurious response to an image frequency. Under RED article 3.2 in the European Union, these tests are mandatory through the ETSI harmonised standards (EN 300 328, EN 300 220-1, EN 301 893, EN 303 687 for Wi-Fi 6 GHz) and 3GPP specifications (TS 36.521-1 for LTE, TS 38.521-1 and TS 38.521-2 for 5G NR). Under FCC Part 15.247 and 15.249 in the United States, the historical focus stays on transmitter-side tests, but OET KDB 558074 introduces RX blocking elements for certain product classes. This page details the bench methodology, the standard-referenced thresholds, the OTA specifics for 5G FR2 mmWave, the sub-GHz LPWA case (LoRa, Sigfox), and the recurring pitfalls observed in pre-compliance.

Why separate TX and RX tests

A regulated radio product complies with two distinct sets of requirements. The first constrains what it radiates: power, spectral occupancy, out-of-channel leakage. The second constrains what it tolerates receiving without degradation: interfering signals in its frequency neighbourhood or products generated by non-linearity in its own input stage.

Transmitter-side tests (TX) protect other spectrum users. Receiver-side tests (RX) protect the product's own end user. Without RX tests, a product may radiate cleanly yet be unusable in the presence of a neighbouring Wi-Fi or a colocated cellular base station.

Under the RED article 3.2 framework, both families are mandatory. Under FCC Part 15 in the United States, TX tests are the rule and RX tests stay less systematic, except when KDB 558074 is invoked. 3GPP, via TS 36.521-1 and TS 38.521-1, treats TX and RX with equal weight in a cellular terminal test plan.

Category	Measured quantity	Reference type
TX power	Maximum power, EIRP, EVM	ETSI EN 300 328, 3GPP TS 36.521-1
TX spectrum	OBW, SEM, ACLR, spurious emissions	ETSI EN 300 328, ETSI EN 300 220-1
RX sensitivity	REFSENS, BER or BLER target	3GPP TS 38.521-1, ETSI EN 300 220-1
RX selectivity	Adjacent channel ACS	3GPP TS 36.521-1, ETSI EN 300 328
RX blocking	In-band and out-of-band blocking	3GPP TS 38.521-1, FCC KDB 558074
RX intermodulation	IM3 by dual interferer	3GPP TS 36.521-1, ETSI EN 300 328
RX spurious response	Image frequency and LO harmonics	ETSI EN 300 220-1

The receiver spurious response

The "spurious response" refers to frequencies outside the useful band at which the receiver reacts as if they were in band. The main sources are:

Intermediate frequency (IF) image. A superheterodyne receiver down-converts the RF signal to an IF frequency via mixing with a local oscillator (LO). The image frequency, symmetric to the wanted signal around the LO, is converted to the same IF. Without sufficient image filtering, a strong interferer on the image frequency enters the IF chain and disrupts reception.
Local oscillator harmonics. LO harmonics produce parasitic conversions at distant frequencies. A strong interferer at one of these frequencies folds back into the useful band.
Internal mixing products. Mixings between LO, wanted signal and interferers generate combinations f_LO +/- f_signal +/- f_interferer that can fall in the useful band.
Front-end parasitic resonances. Mistuned filters, badly terminated lines, insufficient decoupling create unforeseen coupling paths.

The spurious response test characterises these frequencies by injecting a modulated interferer and sweeping the frequency over a wide range (typically DC to a few harmonics of the carrier). The ETSI EN 300 220-1 (sub-GHz SRD) and 3GPP TS 36.521-1 / TS 38.521-1 standards define the tested frequencies, the interferer level and the tolerated degradation threshold.

RX blocking methodology

The receiver blocking test (also blocking response, desensitisation) verifies that the presence of a strong out-of-channel interferer does not degrade sensitivity beyond a threshold. The standard method uses a three-source bench:

                  +------------------+
   Generator 1 -->|                  |
   (wanted)       |    RF combiner   |---> DUT (receiver)
   Generator 2 -->|                  |        |
   (interferer)   +------------------+        |
                                              v
                                       BER/PER/BLER measurement
                                       (software or callbox)

Typical sequence for ETSI EN 300 328 clause 5.4.6 or 3GPP TS 38.521-1:

Calibrate the bench, measure DUT reference sensitivity (REFSENS) on the tested channel.
Set generator 1 to the wanted channel frequency, level at REFSENS + 3 dB (exact margin specified by the standard).
Set generator 2 to the blocking frequency defined by the standard, modulation specified (CW or modulated depending on the case).
Raise the generator 2 level to the threshold specified by the standard.
Measure the error rate, verify it stays below the pass threshold (target BER, PER or BLER).
Repeat for each blocking frequency defined by the standard.

The pass threshold and the absolute interferer level are defined in the applicable standard and vary by band, channel bandwidth and product class. Citing a threshold without normative reference is meaningless.

In-band blocking. Interferer at a frequency close to the wanted channel, within the product's operating band (for instance another Wi-Fi channel in the 2.4 GHz band).

Out-of-band blocking. Interferer at a frequency outside the operating band (for instance a cellular signal on 1800 MHz interfering with a 2.4 GHz Wi-Fi receiver).

Narrow-band blocking. Variant with an unmodulated CW interferer at a close frequency, which exposes precise resonances or mixing products.

ACS, adjacent channel selectivity

ACS (Adjacent Channel Selectivity) verifies that the receiver sufficiently rejects a modulated signal of the same nature on the adjacent channel. The bench is identical to the blocking bench, but generator 2 is set on the adjacent channel frequency (offset equal to the nominal channel bandwidth), with a modulation representative of the wanted signal (for instance an OFDM signal if the DUT is OFDM).

The result is expressed as a ratio in dB between the wanted signal level and the tolerated adjacent interferer level for a target error rate. The higher the ACS (in dB), the better the selectivity.

ACS is very sensitive to:

RF input filter quality (SAW or BAW filter),
LNA and mixer linearity,
local oscillator phase noise (a noisy LO spreads adjacent interferer energy onto the wanted channel via reciprocal mixing),
ADC dynamic range.

ACS thresholds appear in 3GPP TS 36.521-1 for LTE, 3GPP TS 38.521-1 for 5G NR FR1, and in the receiver sections of ETSI EN 300 328 (Wi-Fi/BLE 2.4 GHz) and EN 301 893 (Wi-Fi 5 GHz).

IM3 intermodulation

The intermodulation test uses two simultaneous interferer generators at frequencies f1 and f2. Front-end non-linearity produces mixing products, the most problematic being the third-order product 2f1 - f2 (and its symmetric 2f2 - f1). The standard picks f1 and f2 such that 2*f1 - f2 falls inside the useful band, sometimes exactly on the tested channel.

The bench becomes:

   Wanted signal generator -----+
                                |
   Interferer #1 generator -----+--> Combiner ---> DUT
                                |
   Interferer #2 generator -----+

Sequence:

Calibrate, measure REFSENS.
Set the wanted signal at REFSENS + 3 dB.
Set interferer #1 at f1, interferer #2 at f2, levels per the standard.
Measure the error rate, verify it stays below the pass threshold.

IM3 is dimensioning in dense multi-radio environments, where several colocated transmitters generate strong signals at different frequencies simultaneously. A cellular IoT product colocated with a Wi-Fi access point, or a smartphone in line of sight of several cells, are the typical cases.

Standards mapping by band and protocol

The following table summarises the receiver standards used per product family.

Band / protocol	Primary RX standard	Covered tests	Regulatory framework
Wi-Fi / BLE 2.4 GHz	ETSI EN 300 328	Blocking, ACS, intermodulation	RED article 3.2
Wi-Fi 5 GHz	ETSI EN 301 893	Blocking, ACS, DFS	RED article 3.2
Wi-Fi 6 GHz	ETSI EN 303 687	Blocking, ACS, AFC	RED article 3.2
Sub-GHz SRD (433/868 MHz)	ETSI EN 300 220-1	Blocking, ACS, spurious response, intermodulation	RED article 3.2
LTE UE	3GPP TS 36.521-1	REFSENS, ACS, blocking, IM3, spurious	PTCRB, GCF, RED article 3.2
5G NR FR1	3GPP TS 38.521-1	REFSENS, ACS, blocking, IM3, spurious	PTCRB, GCF, RED article 3.2
5G NR FR2 mmWave	TS 38.521-2	OTA REFSENS, OTA blocking, OTA ACS	PTCRB, GCF, RED article 3.2
FCC Part 15.247	FCC Part 15.247 + KDB 558074	Class-dependent selective blocking	FCC tests
FCC Part 15.249	FCC Part 15.249	TX dominant, limited RX	FCC tests

The mapping does not exempt the file from citing the exact version of the standard (ETSI release, 3GPP version, FCC KDB edition). A standard without version in a conformity file is insufficient for a notified body or a TCB.

Bench architecture

The minimum instrumentation for a full-compliance RX bench includes:

Element	Role	Typical range
Wanted signal generator	Generates the modulated signal on the tested channel	Covers DUT operating band, native modulation
Interferer generator(s)	Generates the interferer signals	Covers the blocking-frequency sweep range
RF combiner / coupler	Combines sources without parasitic intermodulation	Isolation specified by the standard
Spectrum analyser	Verifies source purity and levels	Covers the extended band
Wattmeter / power meter	Absolute power calibration	Head matched to the band
Error rate measurement	BER, PER or BLER on the DUT side	DUT software, 3GPP callbox, Wi-Fi/BLE sniffer
OTA chamber (if required)	OTA measurement for portless products	3GPP/CTIA anechoic, FR2 if mmWave

The combiner is an underestimated item. A poorly isolated combiner re-injects the interferer into the wanted-signal source, which modulates generator 1 and biases the measurement. ETSI and 3GPP standards mandate a minimum isolation between combiner input paths, generally specified in the methodology annex of the document. A bench assembled with basic splitters exposes its measurements to a systematic bias.

Pass thresholds: where to find them

RX test pass thresholds are defined in the applicable standard and are not invented by the lab. For each test, the standard specifies:

the wanted signal frequency (tested channel),
the interferer frequency or frequencies,
the wanted signal level (often expressed as REFSENS + delta dB),
the interferer level (in absolute dBm or as a delta dB relative to the wanted signal),
the source modulation,
the target error rate (BER, PER, BLER, minimum throughput),
the tolerance margin (measurement uncertainty deducted or added).

Practical reading: a conformity report cites the exact clause of the standard (for instance "ETSI EN 300 328 V2.2.2 clause 5.4.6" or "3GPP TS 38.521-1 V17.x.y clause 7.7B") and reproduces the results table with normative references. Citing "RX blocking compliant" without precise clause is insufficient.

See RED tests for the ETSI-clause mapping per band, and RED standards for the list of harmonised standards accepted by notified bodies.

Sub-GHz LPWA: LoRa, Sigfox, SRD

Sub-GHz LPWA products (LoRa Alliance, Sigfox) operate in the 433 MHz and 868 MHz ISM bands in Europe, 902 to 928 MHz in North America. The ETSI baseline is ETSI EN 300 220-1 for the general RF part (TX and RX), with the EN 300 220-2, EN 300 220-3-1 and EN 300 220-3-2 variants for equipment classes.

RX specifics in sub-GHz LPWA:

Narrow band. LoRa and Sigfox signals use narrow channels (down to a few kHz for Sigfox). Adjacent channel ACS is therefore very sensitive to RF filtering and LO phase noise.
Low data rate, long integration. The receiver integrates the signal over a long duration, which improves sensitivity but extends test duration (a single BER point can take seconds to minutes).
Regulatory duty cycle. In sub-GHz Europe, TX duty cycle limits emission duration, which constrains the bench on the TX side but not on the RX side. The RX test can transmit continuously from the generator; the DUT is in receive and is not bound by the TX duty cycle.
Very wide spurious response. Frequencies tested for the spurious response typically cover 25 MHz to 2-3 times the carrier frequency, with a step size defined by the standard. Sweep duration is significant.

For Sigfox and LoRa, the RED article 3.2 file cites EN 300 220-1 plus the relevant sub-part (EN 300 220-2 for non-specific SRD). FCC Part 15.247 and 15.249 apply to the North American equivalents with a different methodology.

5G FR2 mmWave: OTA mandatory

For 5G NR FR2 (bands n257 26 GHz, n258 24 GHz, n260 39 GHz, n261 28 GHz, n262 47 GHz), TS 38.521-2 mandates OTA measurement in an anechoic chamber: mmWave antennas are integrated within the RFIC module, no external port exists. Blocking, ACS and IM3 tests use OTA sources oriented towards the UE, rigorous per-measurement-cell chamber calibration (source-to-UE path loss characterised for each tested frequency), and angular positioning defined by the standard, often in line of sight of the dominant beam. Interferer generators drive a separate OTA source (a second antenna, polarised or oriented per the standard).

The fleet of OTA chambers capable of running these FR2 RX tests stays small in Europe and North America, which becomes a calendar factor whenever an FR2 product enters PTCRB or GCF scope.

Multi-radio coexistence and TX/RX coupling

IoT products often run several radios in parallel (Wi-Fi 2.4 GHz, Wi-Fi 5 GHz, BLE, LTE, NB-IoT, GNSS), which generates additional RX tests: self-interference (the 2.4 GHz Wi-Fi PA entering the colocated BLE LNA), insufficient inter-chain filtering leading to failed IM3 or blocking, firmware scheduling on combo chipsets. The Bluetooth SIG mandates coexistence tests for qualified products. ETSI EN 300 328 and EN 301 893 include LBT (Listen Before Talk) tests, TX-side but conditioning RX quality in the presence of another transmitter in the band.

OBW and SEM (TX-side) bound how much each product radiates out of channel. ACS and blocking (RX-side) bound how much each product must tolerate. The pair is complementary and inseparable: a product that passes TX but fails RX is unusable in deployment, and a product that passes RX with large margin but fails TX pollutes others' spectrum. The ETSI and 3GPP frameworks are built on this parity.

Pre-compliance vs full-compliance

Aspect	Pre-compliance	Full-compliance
Lab	In-house or non-accredited	ISO/IEC 17025 accredited
Recognition	None, indicative value	RED notified body, FCC TCB, PTCRB/GCF lab
Bench	Basic generator, analyser, combiner	Complete bench with traceable calibration
Report	Internal note	Formal report, file-admissible
Cost	Low, internalised	High, per bench day
Use	Design iteration, risk reduction	Final submission to file

A correctly run RX pre-compliance campaign divides the risk of full-compliance failure by a significant factor. Gaps between pre-compliance and full-compliance typically come from:

lower-quality combiner in pre-compliance (parasitic intermodulation),
simplified interferer modulation (CW instead of modulated OFDM),
reduced BER integration time (false pass through too-short measurement),
absence of an OTA chamber in pre-compliance for portless products.

Common pitfalls in RX tests

Wanted signal level mis-set. The blocking test requires the wanted signal at REFSENS + 3 dB. If the bench raises it to REFSENS + 10 dB, the DUT has more margin and the pass is misleading. The standard specifies the exact delta to apply.

Simplified interferer modulation. An ACS test run with a CW interferer instead of a representative modulated signal underestimates reciprocal mixing impact from a noisy LO. The pre-compliance pass does not survive full-compliance.

Basic combiner. A 6 dB basic coexistence splitter re-injects the interferer into the wanted-signal generator and modulates source 1. Result: false fail or false pass depending on phase. The standard demands minimum isolation between combiner paths.

IM3 measured without two independent generators. The IM3 test demands two independent and uncorrelated interferer sources. A single generator with division does not fit, the two signals keep a phase coherence that alters the mixing product.

Made-up blocking frequencies. The tested frequencies are defined in the standard, not by the lab. A blocking test that freely sweeps the spectrum is non-compliant. The bench reproduces the exact list of frequencies defined by ETSI or 3GPP.

Reference without version. The report cites "EN 300 328" without version. Yet the edition (V2.2.2, V2.3.0, etc.) shifts thresholds and blocking frequencies. Without a version, the notified body rejects the file.

FR2 OTA tested in FR1 chamber. A sub-6 GHz chamber does not cover the mmWave range. Calibration is out of range, results are invalid. The OTA chamber must cover the exact range of the target band.

Pre-compliance treated as full-compliance. A pre-compliance in-house report is attached to a RED or FCC declaration file. The notified body or TCB rejects it: only a lab accredited ISO/IEC 17025 and recognised by the target scheme delivers an admissible report.

Self-interferer ignored. A combo Wi-Fi + BLE product is tested with a single radio active at a time. In real deployment, both radios are active, and self-interference degrades sensitivity. The realistic scenario is mandatory for carrier acceptance and for Bluetooth SIG qualification.

FCC KDB ignored. The FCC Part 15.247 file is submitted without checking KDB 558074. For certain product classes, the KDB demands additional RX tests, the absence of which triggers a TCB return.

Connection with other schemes

The RX tests covered here map to the following certification schemes:

RED article 3.2 (European Union), via RED tests and RED standards for the full ETSI mapping.
FCC Part 15 (United States), via FCC tests and the invocation of KDB 558074 for the RX scope.
PTCRB and GCF (cellular), via PTCRB tests and the 3GPP RF Conformance Test Plan for TS 36.521 and TS 38.521.
Carrier acceptance (AT&T NAFI, Verizon OPC, Deutsche Telekom IoT, China Mobile), which add RX tests under real network operating conditions.

The TX/RX parity in the ETSI and 3GPP frameworks requires treating both families with equal rigour. For the RX vocabulary (REFSENS, BER, BLER, ACS, blocking, IM3, spurious response, reciprocal mixing), see the spilma glossary.

Sources & references

ETSI EN 300 328, Wideband transmission systems 2.4 GHz , ETSI www.etsi.org/
ETSI EN 300 220-1, Short Range Devices sub-GHz , ETSI www.etsi.org/
ETSI EN 301 893, 5 GHz RLAN harmonised standard , ETSI www.etsi.org/
3GPP TS 38.521-1, NR UE conformance, FR1 standalone , 3GPP www.3gpp.org/dynareport/38521-1.htm
3GPP TS 36.521-1, LTE UE conformance, Part 1: RF , 3GPP www.3gpp.org/dynareport/36521-1.htm
FCC OET KDB 558074, RF Exposure and receiver guidance , FCC OET apps.fcc.gov/oetcf/kdb/