APH010 DW1000 Inter Channel Interference
APH010 DW1000 Inter Channel Interference
CHANNEL INTERFERENCE
Version 1.2
TABLE OF CONTENTS
1 INTRODUCTION ..................................................................................................................... 4
2 SUMMARY RESULTS AND CONCLUSIONS................................................................................ 5
2.1 INTRODUCTION ............................................................................................................................. 5
2.2 METHODOLOGY ............................................................................................................................ 5
2.3 TRANSMISSION / RECEPTION ON THE SAME CHANNEL.......................................................................... 5
2.4 TRANSMISSION / RECEPTION ON DIFFERENT CHANNELS ....................................................................... 5
2.5 CONCLUSIONS & GUIDELINES.......................................................................................................... 7
3 INTER CHANNEL INTERFERENCE IN THE DW1000 .................................................................... 9
3.1 OVERVIEW ................................................................................................................................... 9
3.2 RF SIGNAL INTERFERENCE BETWEEN CHANNELS .................................................................................. 9
3.3 RESULTING INTERFERENCE ............................................................................................................ 10
4 EFFECT OF USING DIFFERENT PRF AND PREAMBLE CODES ON THE SAME CHANNEL FOR
TRANSMIT & RECEIVE................................................................................................................. 15
4.1 INTRODUCTION ........................................................................................................................... 15
4.2 DIFFERENT VALUES OF PRF AND PREAMBLE CODE ............................................................................ 15
4.3 SAME PRF BUT DIFFERENT PREAMBLE CODES .................................................................................. 15
5 INTERFERENCE BETWEEN DIFFERENT CHANNELS WHEN USING DIFFERENT PRF AND PREAMBLE
CODES ....................................................................................................................................... 18
5.1 INTRODUCTION ........................................................................................................................... 18
5.2 DIFFERENT PRF AND PREAMBLE CODE SETTINGS .............................................................................. 18
5.3 SAME PRF BUT DIFFERENT PREAMBLE CODES .................................................................................. 18
6 CONCLUSIONS AND GUIDELINES FOR CHANNEL CONFIGURATION ........................................ 22
6.1 CONCLUSION .............................................................................................................................. 22
6.2 GUIDELINES TO MINIMIZE THIS INTERFERENCE .................................................................................. 22
7 MAJOR CHANGES ................................................................................................................ 25
8 DOCUMENT HISTORY .......................................................................................................... 26
9 FURTHER INFORMATION ..................................................................................................... 27
10 APPENDIX 1: TEST METHODOLOGY ................................................................................... 28
10.1 TEST SETUP ............................................................................................................................ 28
10.1.1 Initial Calibration of the test rig ...................................................................................... 28
10.1.2 Actual Interference testing .............................................................................................. 28
LIST OF TABLES
TABLE 1: SUMMARY OF INTER-CHANNEL INTERFERENCE USING THE SAME PRF AND PREAMBLE CODE. ....................... 6
TABLE 2: SUMMARY OF INTER-CHANNEL INTERFERENCE USING DIFFERENT PREAMBLE CODES AT 64MHZ PRF ............ 7
TABLE 3: EFFECT OF PRF / PREAMBLE CODE ON CHANNEL 1 ............................................................................ 16
TABLE 4: EFFECT OF PRF / PREAMBLE CODE ON CHANNEL 2 ............................................................................ 16
TABLE 5: EFFECT OF PRF / PREAMBLE CODE ON CHANNEL 3 ............................................................................ 16
TABLE 6: EFFECT OF PRF / PREAMBLE CODE ON CHANNEL 5 ............................................................................ 17
TABLE 7: SUMMARY OF INTER-CHANNEL INTERFERENCE USING THE SAME PRF / PREAMBLE CODE ........................... 18
TABLE 8: EFFECT OF PREAMBLE CODE ON INTERFERENCE FROM CHANNEL 1 TO CHANNEL 2 ................................... 19
LIST OF FIGURES
1 INTRODUCTION
The DW10000 can be configured to operate on one of 6 RF channels with centre frequencies from
3.5 GHz to 6.5 GHz. It uses the modulation scheme and messaging format as defined by the
IEEE802.15.4-2011 UWB standard shown in Figure 1 below.
Header and payload portions of the message are transmitted using very narrow pulses repeated at
the Pulse Repetition Frequency (PRF) which is slightly different for header and payload portions of
the message but nominally can be 16 or 64 MHz.
The preamble sequence portion of the message is transmitted as one of a number of selectable
codes. Restrictions are imposed on the selection of these codes depending on the channel being
used.
Preamble sequences are selected for use in the UWB PHY due to their perfect periodic
autocorrelation properties. The choice of codes that may be used for each channel is restricted to
ensure that codes with the lowest cross-correlation are used in the same UWB PHY channel. Thus,
different devices can communicate on the same physical channel each using different preamble
codes whilst interfering as little as possible.
This note examines how transmissions on one particular channel can cause interference with: -
• Receivers on the same channel configured for different values of PRF and / or preamble code
• Receivers on different channels configured for the same or different PRF and / or preamble
codes
Guidelines are presented for the selection of PRF and preamble codes to minimize such interference.
2.2 Methodology
The methodology used in testing for interference between a transmitter and a victim receiver was to
introduce and increase the level of attenuation between the transmitter and the victim receiver until
the victim receiver was unaffected by the transmissions. This level of attenuation was converted to an
equivalent distance assuming free space operation. This equivalent distance is denoted the “pickup
radius”.
Where the pickup radius proved to be unacceptably large (>2 m) further investigation was carried out
to determine whether particular combinations of preamble codes / PRF could be used to extend it and
with what effect.
With the same PRF configuration but different preamble codes selected then interference may occur
because while the preamble codes have a low cross-correlation they are not orthogonal.
Pickup Radius (m) for this channel at this PRF for worst case
combination of preamble codes
PRF CH1 CH2 CH3 CH5
16 MHz 72 80 76 57
64 MHz 45 48 45 32
These indicate that for the worst case configuration of preamble codes, at channel 1, 16 MHz PRF the
transmitter and victim receiver must be less than 72 m apart for interference to occur. This distance
reduces to 45 m for the 64 MHz PRF case.
1. The signal level received and down-converted by a receiver configured for a particular channel
because of transmissions on a different channel.
2. The degree to which an interfering signal will be demodulated by a receiver when that receiver is
configured for different demodulator settings such as PRF and preamble code
In terms of 1 above, the power spectral density of different channels overlaps with the receiver
bandwidth of adjoining channels to a greater or lesser extent depending on the separation between
the channels as shown, by way of example, in Figure 2 and Figure 3 below.
-30dBm
-40dBm
-50dBm
-60dBm
-70dBm
-80dBm
TX CH1
-90dBm
RX CH2
-30dBm
-40dBm
-50dBm
-60dBm
-70dBm
-80dBm
TX CH1
-90dBm
RX CH5
With respect to point 2 above, the table below shows the situation when the transmitting channel and
the victim channel are configured for the same PRF and preamble code.
Table 1: Summary of inter-channel interference using the same PRF and preamble code.
Legend Explanation
Desired Desired operation; e.g. transmit on channel 1, receive on channel 1
No interference
Interference will occur when range between transmitter and receiver is less than x.y
< x.y m
m
Pickup radius is large enough to warrant further investigation
Analysing the situations highlighted in red above and adjusting preamble codes to minimize the
pickup radius yields the following analysis.
Table 2: Summary of inter-channel interference using different preamble codes at 64MHz PRF
Legend Explanation
Desired Desired operation; e.g. transmit on channel 1, receive on channel 1
No interference
Interference will occur when range between Transmitter and Receiver is less than x.y
< x.y m
m
RECEIVE on
CH1 CH2 CH3 CH4 CH5 CH7
CH1 Desired <8m < 0.7 m <1m
TRANSMIT
This shows that the appropriate selection of preamble code can significantly reduce the pickup radius.
2. Minimize the extent to which the transmitted frame is demodulated / decoded by the victim
receiver.
• With different settings for PRF and Preamble code in the interfering transmitter and victim
receiver no unwanted data is received
• With the interfering transmitter and victim receiver configured for the same PRF but different
preamble codes, signals from the interfering channel may be received on the victim channel but
only when the distance between the interfering transmitter and the victim receiver is below a
certain limit as per Table 1Table 2.
• 64 MHz PRF is more resistant to inter-channel interference than 16 MHz PRF
From Table 2
• Table 13, with all channels configured for 64 MHz PRF, by choosing appropriate preamble codes
the inter-channel interference can be reduced to the point where the interfering transmitter must
be within approximately 1 m of the victim receiver for inter-channel interference to become a
problem.
• At 16MHz PRF this distance goes up to approximately 2.2 m
3. The signal level received and down-converted by a receiver configured for a particular channel
because of transmissions on a different channel. This phenomenon can be examined and
understood by looking at the transmitter power spectral density and understanding the frequency
response of the receiver.
4. The degree to which an interfering signal will be demodulated by a receiver when that receiver is
configured for different demodulator settings such as PRF and preamble code
These side-lobes are such that those on the right hand, higher frequency, side are stronger than
those on the left hand, or lower frequency, side. This means that lower channels interfere more
strongly with higher frequency ones than vice versa.
Note that it is possible to configure the DW1000 transmitter to generate side lobes of lower energy,
but such a configuration will consume more current. The transmitter configurations used for the
measurements presented in this document are the optimal settings as given in the DW1000 User
Manual.
The DW1000 wideband receiver is very sensitive and when it is configured to receive on a particular
channel it is also capable of receiving these low level side-lobes as a result of transmissions on a
different channel which will then be down-converted to baseband in the receive strip.
The PSD of the side-lobes is generally attenuated by 30 dB or more compared to the transmit PSD
main lobe.
The combination of this transmitter PSD with the receiver response for various channel frequencies is
shown in Figure 5 to Figure 20 below. These diagrams are for illustration purposes only and are
based on a zero-loss channel. In reality the transmit spectrum at the receiver will be at a lower level
due to attenuation in the channel. In these diagrams channels 4 and 7, the wideband channels, have
not been included
Figure 5 through Figure 8 show the expected response when both transmitter and receiver are
configured for the same channel (same centre frequency). As expected, the energy transmitted in-
band is received by the receiver and, assuming other parameters are consistent between transmitter
and receiver, the signal is demodulated correctly and the data is recovered.
Figure 9 through Figure 20 show the responses when the transmitter and receiver are configured for
different channels (different centre frequencies).
By way of illustration, Figure 9 shows the situation of a transmission on channel 1 (3.5 GHz) with a
receiver configured for Channel 2 (4 GHz). This illustrates the interference of the high frequency side
lobes in the channel 1 transmitter PSD with the channel 2 receiver.
Conversely, Figure 12 shows a transmission on channel 2 (4 GHz) and a receiver configured for
channel 1 (3.5 GHz). In this case, the low frequency side lobes of the channel 2 transmitter PSD
interfere with the channel 1 receiver but not to the same extent.
Obviously, the greater the separation between the channels the less the interference – this is the first
piece of useful information in terms of system configuration for multiple channel use; maximizing the
separation between the channels minimizes the inter-channel interference.
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH1
TX CH2
-90dBm
RX CH1 -90dBm
RX CH2
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH3 TX CH5
-90dBm -90dBm
RX CH3 RX CH5
1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 8GHz
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH1 TX CH1
-90dBm -90dBm
RX CH2 RX CH3
1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH2
TX CH1 -90dBm
-90dBm RX CH1
RX CH5
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH2 TX CH2
-90dBm -90dBm
RX CH3 RX CH5
1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH3 TX CH3
-90dBm -90dBm
RX CH1 RX CH2
1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH3 TX CH5
-90dBm -90dBm
RX CH5 RX CH2
1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 8GHz
-30dBm -30dBm
-40dBm -40dBm
-50dBm -50dBm
-60dBm -60dBm
-70dBm -70dBm
-80dBm -80dBm
TX CH5 TX CH5
-90dBm -90dBm
RX CH2 RX CH3
1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 8GHz 1GHz 2GHz 3GHz 4GHz 5GHz 6GHz 7GHz 8GHz
4.1 Introduction
In order to analyse the level of inter-channel interference suggested in the preceding plots it is first
useful to examine the effects of using different values of PRF and preamble code when the
transmitter and receiver are configured for the same channel.
The test methodology employed here is discussed in Appendix 1 but, in summary, a transmitter and
receiver are connected via a wired connection with in-line attenuators. These attenuators are
adjusted to simulate distance between the transmitter and receiver. For each configuration the
attenuation is adjusted until the interfering signal is no longer received. The attenuation setting is
converted to distance (knowing the attenuation of radio signals in free-space at the frequency of
transmission) and the resulting “interference radius” or “pickup radius” i.e. the distance beyond
which it can be guaranteed that the transmission will not interfere is recorded in the tables below.
Legend Explanation
Desired Desired operation; e.g. transmit on channel 1, receive on channel 1
xm Interference will occur when range between transmitter and receiver is less than x m
64 MHz
10 40 m 40 m 45 m
No Rx
11 40 m 40 m 45 m
12 40 m 40 m 40 m
16 MHz
1 72 m
No Rx
2 72 m
Preamble
9 10 11 12 3 4
code
9 43 m 43 m 43 m
TRANSMIT ON CHANNEL 2
64 MHz
10 43 m 48 m 48 m
No Rx
11 43 m 48 m 48 m
12 43 m 43 m 48 m
3 80 m
16 MHz
No Rx
4 80 m
Preamble
9 10 11 12 5 6
code
9 40 m 40 m 40 m
TRANSMIT ON CHANNEL 3
64 MHz
10 40 m 40 m 45 m
No Rx
11 40 m 40 m 40 m
12 40 m 40 m 40 m
5 76 m
16 MHz
No Rx
6 76 m
9 32 m 32 m 32 m
TRANSMIT ON CHANNEL 3
64 MHz
10 29 m 29 m 32 m
No Rx
11 32 m 32 m 32 m
12 32 m 32 m 32 m
3 57m
16 MHz
No Rx
4 57m
Table 7: Summary of inter-channel interference using the same PRF / preamble code
Legend Explanation
Desired Desired operation; e.g. transmit on channel 1, receive on channel 1
No interference
Interference will occur when range between transmitter and receiver is less than x.y
< x.y m
m
Pickup radius is large enough to warrant further investigation
For those cases where the pickup radius is seen to be greater than 2 m (in red in Table 7), the inter-
channel interference is tested further to determine what combinations of PRF and preamble codes
may be used to minimize it.
Preamble
9 10 11 12 3 4
code
9 56 m 8m 8m 8m
TRANSMIT ON CHANNEL 1
10 8m 56 m 8m 8m
64MHz
No Rx
11 8m 8m 56 m 8m
12 8m 8m 7m 56 m
1 14 m 14 m
16MHz
No Rx
2 16 m 16 m
TRANSMIT CH 1 / RECEIVE CH 3
Preamble
9 10 11 12 5 6
code
No Rx
11 0.7 m 0.7 m 3.5 m 0.7 m
1 2.2 m 2.2 m
16MHz
No Rx
2 2.2 m 2.2 m
9 60 m 9m 9m 9.7 m
TRANSMIT ON CHANNEL 2
64 MHz
10 9m 60 m 9m 9.7 m
No Rx
11 9m 9m 60 m 9.7 m
12 9m 9.7 m 9m 60 m
3 18 m 18 m
16 MHz
No Rx
4 18 m 20 m
9 85 m 11 m 11 m 11 m
TRANSMIT ON CHANNEL 2
64 MHz
10 11 m 85 m 10 m 11 m
No Rx
11 11 m 11 m 85 m 11 m
12 10 m 11 m 11 m 85 m
3 25 m 25 m
16 MHz
No Rx
4 22 m 22 m
9 40 m 7m 7m 7m
TRANSMIT ON CHANNEL 3
64 MHz
10 6m 40 m 6m 6m
No Rx
11 7m 7m 40 m 6m
12 7m 7m 6m 40 m
5 15 m 14 m
16 MHz
No Rx
6 14 m 14 m
• IC power consumption is increased because the receiver is processing unwanted signals and as
a result battery life is reduced.
• Interference from unwanted channels reduces the signal to noise ratio of the desired signal and
therefore reduces range.
• The level of interference may be such that the channel of interest can be jammed.
• There is a possibility that packets from a different physical channel could actually be received &
decoded correctly if the inter-channel interference is strong enough and the packet error rate
reaches a reasonable level.
• Keep channels well separated. The interference between channels 2 & 5 is far less than between
channels 1 & 2 or 2 & 3 purely because of the frequency spacing between them.
• It is possible to reduce the side lobes in the power spectrum of the transmitter by modifying the
transmitter configuration. Increasing the transmit amplifier gain while reducing that of the transmit
mixer will reduce the side lobes in the transmit spectrum PSD but at the expense of transmitter
power consumption.
• It is possible to modify the power spectrum of the transmitter by using a band-pass filter between
the DW1000 and the antenna centred on the channel centre-frequency of interest and with a
steep cut-off at the band edges. This minimizes the energy in the transmitted side-lobes and
reduces the problem at source. This does imply a BOM cost and solution footprint increase.
• Using an antenna optimized for a particular channel will help to reduce transmitted power in
adjacent channels (and reception of interfering signals from outside the channel of interest) but
this option may not always be possible depending on the desired geographical regions of
deployment.
If the reception of unwanted out-of band UWB interference cannot be avoided then using different
PRF and preamble codes can help: -
• With different settings for PRF and Preamble code in the interfering transmitter and victim
receiver no unwanted data is received
• With the interfering transmitter and victim receiver configured for the same PRF but different
preamble codes, signals from the interfering channel may be received on the victim channel but
only when the distance between the interfering transmitter and the victim receiver is below a
certain limit as per Table 13.
• 64 MHz PRF is more resistant to inter-channel interference than 16 MHz PRF
• From
• Table 13, with all channels configured for 64 MHz PRF, by choosing appropriate preamble codes
the inter-channel interference can be reduced to the point where the interfering transmitter must
be within approximately 1 m of the victim receiver for inter-channel interference to become a
problem.
• At 16MHz PRF this distance goes up to approximately 2.2 m
Table 13: Summary of inter-channel interference using different preamble codes at 64MHz PRF
Legend Explanation
Desired Desired operation; e.g. transmit on channel 1, receive on channel 1
No interference
Interference will occur when range between Transmitter and Receiver is less than x.y
< x.y m
m
RECEIVE on
CH1 CH2 CH3 CH4 CH5 CH7
CH1 Desired <8m < 0.7 m <1m
TRANSMIT
7 MAJOR CHANGES
V1.1
V1.2
8 DOCUMENT HISTORY
Table 14: Document History
9 FURTHER INFORMATION
Decawave develops semiconductors solutions, software, modules, reference designs - that
enable real-time, ultra-accurate, ultra-reliable local area micro-location services.
Decawave’s technology enables an entirely new class of easy to implement, highly secure,
intelligent location functionality and services for IoT and smart consumer products and
applications.
For further information on this or any other Decawave product, please refer to our website
www.decawave.com.
For each test configuration there is a certain amount of loss (cable, variable attenuator, 2 x 3 dB
attenuator) in the link which needs to be taken into account during the testing.
Table 15: Test equipment loss for various channel and PRF settings
PRF setting
Tx Channel 64 MHz 16 MHz
PTX = -14 dBm Rx level (dBm) Loss (dB) Rx level (dBm) Loss (dB)
Ch1 -26.4 -12.4 -25.53 -11.53
Ch2 -27.24 -13.24 -26.6 -12.6
Ch3 -24.57 -10.57 -26.12 -12.12
Ch5 -27.88 -13.88 -27.81 -13.81
Using the previously calibrated test rig, for each configuration of transmitter and receiver, a number of
packets are transmitted. DecaWave’s “DecaRanging” tool is used to indicate the number of
transmitted packets, received packets and various error counts.
Using these statistics we can determine the distance beyond which the inter-channel interference is
negligible or less than 1% of RSE when configuring different PRFs and Preamble codes. This is
denoted the “pickup radius” or “impact radius”.