AN4356

Application note

How to solve antenna desense issue with ECMF

Introduction
Electronic systems feature high-speed links to provide high-definition video and high data rate storage and transfer. To carry this
kind of data, these systems implement multiple connectors with high-speed differential pair signals.
Furthermore, the antennas used by theses systems to communicate with their environment (for example a streaming box
communicating with a wireless interface) are now wide-band or smart antenna. They are more sensitive to electro-magnetic
interferences.
We explain in this document how differential lines can generate electromagnetic noise. This noise can affect the receiver
sensitivity. Then we describe how to maintain a correct noise-free reception of RF signals.

AN4356 - Rev 2 - January 2023 www.st.com

For further information contact your local STMicroelectronics sales office.
 AN4356
EMI and noise issues

1 EMI and noise issues

Inside consumer and industrial applications, such as industrial computer, medical equipment, game console,
tablet, or docking station, there are multiple subsets of the circuits, which communicate together or with other
electronic equipment’s. These communications use several wired links. Close to these electronic circuits, we often
find a WiFi or Bluetooth RF link with an antenna connected to the RF power amplifier, and to the receiver.

Figure 1. An electronic system with various high-speed links

Standards used in this kind of applications present high data rates. Here are some examples of speeds rates per
data lane:
• MIPI D-PHY up to 9 Gbps
• HDMI 2.1 up to 12 Gbps
• USB 2.0 up to 480 Mbps
• USB 3.2 up to 10 Gbps
• USB4 up to 20 Gbps
• MIPI M-PHY up to 11.66 Gbps
The spectral content of these signals can overlap Bluetooth, WiFi and WiFI 6E frequencies. It is between 2.4 GHz
and 7.1 GHz, which can reduce the sensitivity of the RF receiver.
Indeed, in the Figure 2 below, we can see electromagnetic emission (in red) from high-speed signals radiation
that creates harmonics. These harmonics are figured out in the frequency domain by spectral rays (in red). They
can be received by the antenna with the useful signal that we expect to receive (WiFi or BT in this case).
If the power of the useful signal (in blue) is lower than the high-speed signal radiation (in red), the RF antenna
receiver is eventually not able to detect the useful signal. This jamming of the signal or S/N ratio weakening is
called antenna de-sense or desensitization.

AN4356 - Rev 2 page 2/21

 AN4356
EMI and noise issues

Figure 2. High speed links radiated noise and antenna desensitization

These high data rates links use clocks and data signals with very fast transition in the range of hundreds
of picoseconds or lower. It is leading to strong radiation fields. This effect is amplified when more complex
subcircuits are implemented and together with flex connectors. The flex connectors behave like transmission
antenna and spread unwanted frequencies and noise (Figure 3).

Figure 3. Radiated and conducted emissions

The skewed signal that can occur between two lines of the differential pair link of a flexible PCB induce an
important source of radiated noise. The resulting common mode signal is the source of the radiated noise
(Figure 4).

AN4356 - Rev 2 page 3/21

 AN4356
EMI and noise issues

Figure 4. Effect of the skew on a differential pair link

The common mode noise can be generated by a line length difference and as shown in the test setup of Figure 5.
The common mode signal was introduced by a line pair with different length of the two conductors. This emulates
the case of a skewed differential pair.
This test setup provides high-speed “pseudo random binary sequence” (PRBS) differential signal, with the same
data rate and the same electrical features as a 10 Gbps USB 3.2 signal. We use an RF combiner to add the D+
and D- signals and to extract the common mode signal. This signal is then measured with a spectrum analyzer.

Figure 5. USB 3.2 at 10 Gbps with skew test setup

The Figure 6 and the Figure 7 shows the spectral content of the resulting common mode signal. We can observe
the impact of the signal skew. We show that high-speed differential pair signals like USB 3.2 at 10 Gbps can
generate noise in all WiFi and Bluetooth frequency bands.

AN4356 - Rev 2 page 4/21

 AN4356
EMI and noise issues

Figure 6. Spectral content of PRBS at 10 Gbps common mode noise within 2.4 GHz WiFi and Bluetooth
frequency bands

Figure 7. Spectral content of PRBS at 10 Gbps common mode noise within WiFi 5 and 6E frequency
bands

The noise amplitude, and its adverse effects on the equipment, depends on the PCB layout, shielding and the
proximity of the sensitive parts.
A second test has been performed to show the impact of the antenna desense on a real application. A home
gateway, which works at a 2.4 GHz and 5 GHz WiFi dual frequency was used. This box communicates with a
computer (number 1), which analyzes the WiFi signal received using a software. The software analyzes the power
of the received signal and the frequency channels used.
At the same time, data are exchanged between a “solid state drive” and a computer (number 2) communicating in
USB 3.2 at 10 Gbps through a USB Type-C® cable (Figure 8).

AN4356 - Rev 2 page 5/21

 AN4356
EMI and noise issues

Figure 8. Antenna desense demonstration test setup

The Figure 9 and the Figure 10 shows screenshots of the WiFi analyzer software for respectively 2.4 GHz and
5 GHz WiFi. The yellow curve gives the WiFi signal RSSI, showing the power of the received signal.
When the USB 3.2 transfer is inactive, we have a good WiFi signal reception. On the other hand, when we start
to exchange data between the laptop and the SSD we lose the WiFi connection. Then, the signal and its detection
reappear at once when the USB 10 Gbps data transfer is stopped. For this test, we used a standard shielded
USB Type-C® cable.
This behavior appears on both 2.4 GHz and 5 GHz WiFi.

Figure 9. WiFi 2.4 GHz signal RSSI for channel 1

Figure 10. WiFi 5 signal RSSI for channels 108 and 112

The conclusion of this test is that the USB 3.2 at 10 Gbps can generate radiated noise in the 2.4 GHz WiFi and
the 5 GHz WiFi frequency range. As a matter of fact, it decreases the sensitivity of the WiFi antenna receiver. This
phenomenon is called antenna desense or antenna desensitization.

AN4356 - Rev 2 page 6/21

 AN4356
Solution to avoid antenna desense

2 Solution to avoid antenna desense

The best method to prevent antenna desense is to insert a common-mode filter on the differential pair link.
This device includes two coupled inductances. They act as a subtracter for common-mode noise thanks to the
magnetic flux accumulation. It induces high-Z for the common-mode noise path. The differential signal is not
affected (Figure 11).

Figure 11. Common mode filter basics

The common-mode filter is efficient on all the common-mode noises induced on the link (see Figure 12):
• Conduction noise
• Radiated noise
• Noise induced by signal skew.

Figure 12. Common Mode Filter against various noise sources

The key points on the selection criteria for a common-mode filter are:
• High common-mode rejection (SCC21) in the targeted frequency range to reject the unwanted common-
mode noise.
• High differential bandwidth (SDD21) to pass the useful high data-rate signals without distortion.
• Low DC resistance to ensure lossless signal and impedance matching.

AN4356 - Rev 2 page 7/21

 AN4356
ECMF Family

2.1 ECMF Family

The STMicroelectronics ECMF device range combines a common-mode filter with ESD protection in a single
device.
A wide range of products is available. All are dedicated to high-speed links.
For example, Figure 13 shows the pinout and the functional schematic of the ECMF2-40A100N6 and the
ECMF4-40A100N10 common-mode filters.

Figure 13. ECMF2-40A100N6 and ECMF4-40A100N10 pinout and functional schematic

Common mode rejection (SCC21) and differential frequency response (SDD21) are shown in Figure 14.
This device provides a high differential bandwidth, which is compatible with high-speed standards like USB 4,
USB 3.2 Gen 2, MIPI M or D-PHY, HDMI 2.1 and so on. It also integrates an ESD protection. This product also
provides a good common-mode rejection between 2.4 GHz and 7 GHz, which are inside the Bluetooth, WiFi, and
WiFi 6E frequencies channels.

Figure 14. ECMF4-40A100N10 common mode rejection versus frequency and differential frequency
response

In the Figure 15, the same test setup as the Figure 5. USB 3.2 at 10 Gbps with skew test setup is performed, but
we add ECMF4-40A100N10 to filter the common mode noise generated by the skew.

AN4356 - Rev 2 page 8/21

 AN4356
ECMF Family

Figure 15. USB 3.2 at 10 Gbps with skew test setup with ECMF4-40A100N10

The Figure 16 and the Figure 17 shows the spectral content of the resulting common-mode signal without any
filter (in blue) and with ECMF4-40A100N10 (in orange). We have a good attenuation of around -15 dB on the
whole WiFi and Bluetooth frequency bands. This is in line with the specification of this product shown in the
Figure 14. With this attenuation, we can allow the antenna to receive the desired signal and to secure the
communication.

Figure 16. Common-mode signal spectrum with and without ECMF4-40A100N10 on WiFi 2.4 GHz and
Bluetooth frequency bands

AN4356 - Rev 2 page 9/21

 AN4356
ECMF Family

Figure 17. Common-mode signal spectrum with and without ECMF4-40A100N10 on WiFi 5 and WiFi 6E
frequency bands

We repeated the test shown in the Figure 5, but we added the STEVAL-OET005VD with ECMF4-40A100N10 to
filter the common mode noise generated by USB 3.2 at 10 Gbps link.

Figure 18. Antenna desense reduction test setup with ECMF4-40A100N10 common-mode filtering

The Figure 19 and the Figure 20 summarize the results obtained in the 2.4 and 5 GHz WiFi bands with the
addition of the ECMF. We can see the benefit of this solution to keep a good reception of the WiFi signal.
For comparison, the Figure 9 and the Figure 10 snapshots were taken before common mode filtering.

AN4356 - Rev 2 page 10/21

 AN4356
ECMF Family

Figure 19. WiFi 2.4 GHz signal RSSI for channel 1 with ECMF4-40A100N10

Figure 20. WiFi 5 signal RSSI for channels 108 and 112 with ECMF4-40A100N10

With the ECMF, cable radiations effects are attenuated, removing the antenna desense effect of the USB 3.2 on
WiFi bands.
These trials have been performed on the USB 3.2 interface, but the same issues could be encountered on other
links such as MIPI, display Port, HDMI…
STMicroelectronics also provides automotive-grade ECMF devices for various automotive applications.
Depending on the useful signals used in the application and the frequencies to be rejected, a proper selection of
the ECMF device could help to solve these kinds of problems.
Indeed, as a reminder, the SDD21 parameter is the high differential bandwidth to pass the useful high data-rate
signals without distortion. The SCC21 parameter is the high common-mode rejection in the targeted frequency
range to reject the unwanted common-mode noise. These parameters, in addition to those for ESD protection,
must be chosen in relation to the intended application.
The two first tests performed before have shown the efficiency of a common-mode filter to suppress unwanted
common-mode noise that can be responsible for the desensitization of the antenna.

AN4356 - Rev 2 page 11/21

 AN4356
ECMF transparency for differential signals

2.2 ECMF transparency for differential signals

A second important point is the transparency of the ECMF in the application, meaning the filters should not affect
the useful high-speed signal transmitted between transceivers. To verify the transparency, we carried out mixed
mode S-parameters measurement and eye-diagrams measurements.
To see the transparency of our filter in the application, we measured the speed of the data transmission between
the PC and the 10 Gbps SSD with and without the ECMF4-40A100N10. This measurement was done using a
data-rate measurement software (Figure 21). The software measures the read and write speeds of the SSD for
sequential and random operations.
As we have similar values for both read and write operations, we can conclude that our filter does not affect the
transmission speed of the useful signal.

Figure 21. Test setup for USB 3.2 data speed transmission measurement, with and without ECMF

Many high-speed digital standards like HDMI 2.1, USB 3.2, or USB 4 provide eye diagram mask templates to
pass the standards.
An eye diagram is produced by repetitively sampling a digital signal on an oscilloscope’s vertical axis, while
triggering the horizontal sweep with the data rate.
In the Figure 22, eye-diagrams measurements on a USB3.2 standard are performed. A PRBS pattern generator
is used with de-emphasis and preshoot functions to simulate the real digital signal of the standard being tested.
The oscilloscope integrates a USB3.2 type-C® reference cable insertion losses, CTLE, and DFE that are signal
processing algorithms defined in the standard.

Figure 22. Eye-diagrams test setup for USB 3.2 standard

AN4356 - Rev 2 page 12/21

 AN4356
ECMF transparency for differential signals

The test is performed in two steps:

• the first one (Figure 25) is measured on a perfect calibration through (it means without filter),
• the second one (Figure 26) is performed with ECMF4-40A100N10.
As we can see, there is a very low eye diagram distortion with the addition of the filter, and the template of the
standard is respected.

Figure 23. USB4 20.0 Gbps eye diagram without Figure 24. USB4 20.0 Gbps eye diagram with
ECMFx-40A100Nx (with preset 0, reference cable 0.8m, ECMFx-40A100Nx (with preset 0, reference cable 0.8m,
equalizer with ADC = 0dB and DFE) equalizer with ADC = 0dB and DFE)

Figure 25. USB3.2 Gen 2 10.0 Gbps eye diagram without Figure 26. USB3.2 Gen 2 10.0 Gbps eye diagram with
ECMFx-40A100Nx (with type C connector, reference ECMFx-40A100Nx (with type C connector, reference
cable, equalizer with ADC = 5 dB and DFE) cable, equalizer with ADC = 5 dB and DFE)

AN4356 - Rev 2 page 13/21

 AN4356
ECMF transparency for differential signals

Figure 27. HDMI2.1 12 Gbps eye diagram without Figure 28. HDMI2.1 12 Gbps eye diagram with
ECMFx-40A100Nx (with worst cable model (WCM3), ECMFx-40A100Nx (with worst cable model (WCM3),
EQ with 8 dB CTLE and one-tap DFE) EQ with 8 dB CTLE and one-tap DFE)

Figure 29. USB 2.0 High Speed 480 Mbps eye diagram, Figure 30. USB 2.0 High Speed 480 Mbps eye diagram,
template 1, without ECMFx-40A100Nx template 1, with ECMFx-40A100Nx

Figure 31. MIPI M-PHY–Gear 4 at 11.66 Gbps eye diagram Figure 32. MIPI M-PHY–Gear 4 at 11.66 Gbps eye diagram
without ECMFx-40A100Nx with ECMFx-40A100Nx

AN4356 - Rev 2 page 14/21

 AN4356
ECMF layout recommendations

2.3 ECMF layout recommendations

The ECMF4-40A100N10, structure is symmetrical: D+ or D- can be swapped.
The most important recommendation is to connect the device to the external connector on the side where ESD
protection diodes are implemented.
The best position for the ECMFs is as close as possible to the connectors for the external links, and as close
as possible to the flex connector for the internal ones. For long flex or cables, one ECMF at each end is more
efficient.
Finally, ECMF connection to the ground needs to be as short as possible, with multiple vias to minimize parasitic
inductance.

Figure 33. Layout example for HDMI TMDS type A connector

AN4356 - Rev 2 page 15/21

 AN4356
Conclusion

3 Conclusion

Loosing sensitivity due to EMI is usual: cables, PCB, or flexes can radiate a lot of noises. Maintaining a low
noise level during reception is a key point on various applications to ensure their qualification and to keep good
reliability results.
Common-mode filters are the suitable device to guard against this issue. But it must be chosen correctly as
regards to the various application constraints such as attenuation level in the right frequency bands, frequency
bandwidth, being in common mode or in differential mode.
Another benefit of ECMF technology is for highly disturbed places like, for example, industrial environments.
Indeed, in that kind of environments, we can have strong magnetic fields, which can create saturation of
ferromagnetic materials. With ECMF, there is no saturation phenomenon because there is no magnetic material.
Inductors coupling is done without magnetic materials.
STMicroelectronics presents a wide range of devices dedicated for each application (including automotive grade
solutions), that provide a good fit to customer needs and offer the integration of function not allowed by using
discrete devices. It also takes less PCB surface area and improves assembly cost.

AN4356 - Rev 2 page 16/21

 AN4356

Revision history
Table 1. Document revision history

Date Revision Changes

03-Oct-2013 1 Initial release.

09-Jan-2023 2 Updated for current products. Minors text changes.

AN4356 - Rev 2 page 17/21

 AN4356
Contents

Contents
1 EMI and noise issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Solution to avoid antenna desense. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 ECMF Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 ECMF transparency for differential signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 ECMF layout recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16
Revision history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

AN4356 - Rev 2 page 18/21

 AN4356
List of tables

List of tables
Table 1. Document revision history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

AN4356 - Rev 2 page 19/21

 AN4356
List of figures

List of figures
Figure 1. An electronic system with various high-speed links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Figure 2. High speed links radiated noise and antenna desensitization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Figure 3. Radiated and conducted emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Figure 4. Effect of the skew on a differential pair link . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Figure 5. USB 3.2 at 10 Gbps with skew test setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Figure 6. Spectral content of PRBS at 10 Gbps common mode noise within 2.4 GHz WiFi and Bluetooth frequency bands 5
Figure 7. Spectral content of PRBS at 10 Gbps common mode noise within WiFi 5 and 6E frequency bands . . . . . . . . . . 5
Figure 8. Antenna desense demonstration test setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 9. WiFi 2.4 GHz signal RSSI for channel 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 10. WiFi 5 signal RSSI for channels 108 and 112 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Figure 11. Common mode filter basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Figure 12. Common Mode Filter against various noise sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Figure 13. ECMF2-40A100N6 and ECMF4-40A100N10 pinout and functional schematic. . . . . . . . . . . . . . . . . . . . . . . . . 8
Figure 14. ECMF4-40A100N10 common mode rejection versus frequency and differential frequency response . . . . . . . . . 8
Figure 15. USB 3.2 at 10 Gbps with skew test setup with ECMF4-40A100N10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Figure 16. Common-mode signal spectrum with and without ECMF4-40A100N10 on WiFi 2.4 GHz and Bluetooth frequency
bands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Figure 17. Common-mode signal spectrum with and without ECMF4-40A100N10 on WiFi 5 and WiFi 6E frequency bands 10
Figure 18. Antenna desense reduction test setup with ECMF4-40A100N10 common-mode filtering . . . . . . . . . . . . . . . . 10
Figure 19. WiFi 2.4 GHz signal RSSI for channel 1 with ECMF4-40A100N10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Figure 20. WiFi 5 signal RSSI for channels 108 and 112 with ECMF4-40A100N10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Figure 21. Test setup for USB 3.2 data speed transmission measurement, with and without ECMF. . . . . . . . . . . . . . . . . 12
Figure 22. Eye-diagrams test setup for USB 3.2 standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Figure 23. USB4 20.0 Gbps eye diagram without ECMFx-40A100Nx (with preset 0, reference cable 0.8m, equalizer with
ADC = 0dB and DFE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 24. USB4 20.0 Gbps eye diagram with ECMFx-40A100Nx (with preset 0, reference cable 0.8m, equalizer with ADC =
0dB and DFE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 25. USB3.2 Gen 2 10.0 Gbps eye diagram without ECMFx-40A100Nx (with type C connector, reference cable,
equalizer with ADC = 5 dB and DFE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 26. USB3.2 Gen 2 10.0 Gbps eye diagram with ECMFx-40A100Nx (with type C connector, reference cable, equalizer
with ADC = 5 dB and DFE). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 27. HDMI2.1 12 Gbps eye diagram without ECMFx-40A100Nx (with worst cable model (WCM3), EQ with 8 dB CTLE
and one-tap DFE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 28. HDMI2.1 12 Gbps eye diagram with ECMFx-40A100Nx (with worst cable model (WCM3), EQ with 8 dB CTLE and
one-tap DFE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 29. USB 2.0 High Speed 480 Mbps eye diagram, template 1, without ECMFx-40A100Nx . . . . . . . . . . . . . . . . . . 14
Figure 30. USB 2.0 High Speed 480 Mbps eye diagram, template 1, with ECMFx-40A100Nx. . . . . . . . . . . . . . . . . . . . . 14
Figure 31. MIPI M-PHY–Gear 4 at 11.66 Gbps eye diagram without ECMFx-40A100Nx . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 32. MIPI M-PHY–Gear 4 at 11.66 Gbps eye diagram with ECMFx-40A100Nx. . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 33. Layout example for HDMI TMDS type A connector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

AN4356 - Rev 2 page 20/21

 AN4356

IMPORTANT NOTICE – READ CAREFULLY

STMicroelectronics NV and its subsidiaries (“ST”) reserve the right to make changes, corrections, enhancements, modifications, and improvements to ST
products and/or to this document at any time without notice. Purchasers should obtain the latest relevant information on ST products before placing orders. ST
products are sold pursuant to ST’s terms and conditions of sale in place at the time of order acknowledgment.
Purchasers are solely responsible for the choice, selection, and use of ST products and ST assumes no liability for application assistance or the design of
purchasers’ products.
No license, express or implied, to any intellectual property right is granted by ST herein.
Resale of ST products with provisions different from the information set forth herein shall void any warranty granted by ST for such product.
ST and the ST logo are trademarks of ST. For additional information about ST trademarks, refer to www.st.com/trademarks. All other product or service names
are the property of their respective owners.
Information in this document supersedes and replaces information previously supplied in any prior versions of this document.
© 2023 STMicroelectronics – All rights reserved

AN4356 - Rev 2 page 21/21