Demo Abstract: Realistic Simulation of Radio Interference in COOJA

Abstract

Radio interference drastically affects the reliability and robustness of wireless communications. As wireless sensor network protocols are frequently designed and tested in simula-tion environments first, it is important to have simulation tools that provide means to study the impact of radio interference. Radio propagation models available in simulation environ-ments are however often too simplistic, and can hardly capture the complexity of the real world. To increase the realism of simulations, we incorporate recorded interference traces into existing simulation models. We extend the COOJA simulator with the generation of realistic interference sources in the simulation environment. We add new features, such as an interference-aware propagation model and loading of interference traces, which can be captured and recorded through a mote-based application. In our interactive demo, we show the generation of interference patterns produced by devices operating in the 2.4 GHz band such as microwave ovens, Bluetooth, and Wi-Fi. We will monitor, capture, and record the ongoing interference at runtime, and load the recorded traces in our extended COOJA version. We then show how to use the captured patterns to simulate and study the impact of radio interference on sensornet communications and routing trees.

Full text

Demo Abstract: Realistic Simulation of
Radio Interference in COOJA
Carlo Alberto Boano and Kay R¨
omer
Universit¨
at zu L¨
ubeck
L¨
ubeck, Germany
{cboano, roemer}@iti.uni-luebeck.de
Fredrik ¨
Osterlind and Thiemo Voigt
Swedish Institute of Computer Science
Kista, Stockholm, Sweden
{fros, thiemo}@sics.se
Abstract—Radio interference drastically affects the reliability
and robustness of wireless communications. As wireless sensor
network protocols are frequently designed and tested in simula-
tion environments first, it is important to have simulation tools
that provide means to study the impact of radio interference.
Radio propagation models available in simulation environ-
ments are however often too simplistic, and can hardly capture
the complexity of the real world. To increase the realism of
simulations, we incorporate recorded interference traces into
existing simulation models.
We extend the COOJA simulator with the generation of
realistic interference sources in the simulation environment. We
add new features, such as an interference-aware propagation
model and loading of interference traces, which can be captured
and recorded through a mote-based application.
In our interactive demo, we show the generation of
interference patterns produced by devices operating in the 2.4
GHz band such as microwave ovens, Bluetooth, and Wi-Fi. We
will monitor, capture, and record the ongoing interference at
runtime, and load the recorded traces in our extended COOJA
version. We then show how to use the captured patterns to
simulate and study the impact of radio interference on sensornet
communications and routing trees.
I. INT RODUCTION AND MOTI VATION
Radio interference considerably affects the reliability and
robustness of wireless communications, hence representing
a major problem in wireless sensor networks. The strong
growth of the number of devices operating in the ISM bands
increases the congestion in the radio spectrum, leading to poor
performance, packet loss, and reduced energy-efficiency [1].
As wireless sensor networks also operate on these crowded
ISM bands, it is necessary to design and develop protocols that
are robust to radio interference. Sensornet protocols are fre-
quently designed and tested in simulation environments first,
and it is therefore important to provide simulation tools that
offer accurate means to study the impact of radio interference.
Modeling radio propagation and interference is complex due
to the large number of variables involved, ranging from the
device(s) operating concurrently on the frequency of interest,
their position, modulation, and transmission scheme, to the
characteristics of the environment and the presence of moving
objects or static obstacles. In certain scenarios with an exces-
sive number of such unknown parameters, e.g. in a crowded
shopping center or lively street, the creation of models that
accurately reflect reality is therefore almost impossible.
-100
-80
-60
-40
-20
0
0 20 40 60 80 100
RSSI Noise floor [dBm]
Time [ms]
Distance 1.0 m
Distance 7.5 m
Fig. 1. Interference recorded from a sensor mote scanning channel 23
in presence of an active Lunik 200 Microwave Oven. Ovens typically emit
frequencies with a periodic pattern, and for this particular model, the period
is approximately 20 ms.
Instead of attempting to create more precise and realistic
radio models, we augment existing simulation tools with the
playback of realistic interference traces. We use off-the-shelf
sensor motes to scan the radio channel and record the
interference patterns, and we then play back the recorded
traces directly in the simulation environment. Such traces
can be added on top of any existing radio model, improving
significantly the level of realism when simulating the impact of
radio interference on sensornet protocols and communications.
II. COOJA EXTENSIONS FOR REA LI ST IC
INT ER FE RE NC E SIM UL ATION
We enrich the COOJA simulator [2] with the generation of
realistic interference sources in the simulation environment.
We upgrade the Multi-path Ray-tracer Medium (MRM)
to correctly implement co-channel rejection according to the
results of Dutta et al. [3]. Co-channel rejection is a measure
of the capability of the receiver to demodulate a wanted
modulated signal without exceeding a given degradation due
to the presence of an unwanted modulated signal. Proper
handling of co-channel rejection enables us to simulate the
correct reception of a packet in presence of interference.
Unwanted signals are represented by (pre-)recorded inter-
ference traces in COOJA. Such traces are used to improve
the realism of sensornet simulations. Using, for example,

Fig. 2. Screenshot of COOJA with the proposed application.
pre-recorded interference traces that resemble the patterns
generated by typical appliances operating in the crowded 2.4
GHz ISM band, the user can arbitrarily place a number of Wi-
Fi and Bluetooth devices as well as microwave ovens inside
the simulation environment. We assume the signal propagation
can be modeled with the widely used log-normal model [4].
III. DEM O DES CR IP TI ON
In this demo, we monitor, capture, record, and play back
the ongoing interference at runtime.
We first show how to capture interference using a high
sampling rate. We use off-the-shelf sensor nodes and measure
the RSSI noise floor, i.e., the RSSI in absence of packet
transmissions in both time and frequency.
As we are interested in detecting also short transmissions
such as Wi-Fi beacons, we boost the CPU speed, optimize the
SPI operations, and compress the RSSI noise floor readings.
We achieve a sampling frequency of approximately 60 kHz
(3.5 kHz) when scanning one (all) 802.15.4 channels. Figure 1
shows a sample interference trace recorded from a sensor mote
scanning channel 23 in presence of an active microwave oven
in the neighborhood.
To collect the RSSI noise floor readings, we attach two
sensor motes to the laptop running COOJA. We use Max-
for MTM-CM5000MSP nodes, widely used sensor motes
equipped with the CC2420 radio transceiver. The two nodes
run Contiki [5]: the first node is used to scan the channel
of interest and record the interference traces at runtime; the
second node is used to give the user a snapshot of the ongoing
interference in the whole 2.4 GHz spectrum as in [6].
We also pre-record several interference traces and build an
object library of interfering devices available as new disturber
mote types, including different models of microwave ovens as
well as Bluetooth and Wi-Fi devices.
Finally, we create several simulation environments and
show the impact of realistic radio interference on sensornet
communications and routing trees. Figure 2 shows a screenshot
of the COOJA simulation.
IV. CONCLUSIONS
In this demo we show how to monitor, capture, and record
the ongoing interference in real time using COOJA. We then
use the captured patterns to simulate and study the impact of
realistic radio interference on sensornet communications and
routing trees.
ACK NOWLEDGMEN TS
This work has been supported by the European Commission
under the contract No. FP7-2007-2-224053 (CONET, the
Cooperating Objects Network of Excellence.
This research has been also partially financed by VIN-
NOVA, the Swedish Agency for Innovation Systems, and by
the Cluster of Excellence 306/1 ”Inflammation at Interfaces”
(Excellence Initiative, Germany, since 2006).
REF ER EN CE S
[1] Axel Sikora and Voicu F. Groza. Coexistence of IEEE 802.15.4 with
other systems in the 2.4 GHz-ISM-Band. In IEEE Instrumentation and
Measurement Technology, pages 1786–1791, Ottawa, Canada, May 2005.
[2] Fredrik ¨
Osterlind, Adam Dunkels, Joakim Eriksson, Niclas Finne, and
Thiemo Voigt. Cross-level Simulation in COOJA. In Proceedings of the
4th European Conference on Wireless Sensor Networks (EWSN), Delft,
The Netherlands, January 2007.
[3] Prabal Dutta, Stephen Dawson-Haggerty, Yin Chen, Chieh-Jan Mike
Liang, and Andreas Terzis. Design and Evaluation of a Versatile and
Efficient Receiver-Initiated Link Layer for Low-Power Wireless. In
Proceedings of the 8th Conference on Embedded Networked Sensor
Systems (SenSys), Zurich, Switzerland, November 2010.
[4] Dimitrios Lymberopoulos, Quentin Lindsey, and Andreas Savvides. An
Empirical Characterization of Radio Signal Strength Variability in 3-D
IEEE 802.15.4 Networks Using Monopole Antennas. In Proceedings
of the 3rd European Conference on Wireless Sensor Networks (EWSN),
pages 326–341, Zurich, Switzerland, February 2006.
[5] Adam Dunkels, Bj ¨
orn Gr¨
onvall, and Thiemo Voigt. Contiki - a
Lightweight and Flexible Operating System for Tiny Networked Sensors.
In 1st Workshop on Embedded Networked Sensors (EmNetS), Tampa,
Florida, USA, November 2004.
[6] Carlo Alberto Boano, Kay R ¨
omer, Zhitao He, Thiemo Voigt, Marco Zu-
niga, and Andreas Willig. Generation of Controllable Radio Interference
for Protocol Testing in Wireless Sensor Networks. In Proceedings of the
7th Conference on Embedded Networked Sensor Systems (SenSys), demo
session, pages 301–302, Berkeley, California, USA, November 2009.
ACM Press.