Description
In this homework, we will use the OMNeT++ network simulator to evaluate the performance of the unslotted carrier sense medium access with collision avoidance (CSMA/CA)
scheme used by the well-known IEEE 802.15.4 standard discussed during class.
Background. The IEEE 802.15.4 standard addresses the physical and medium access
control (MAC) layers of the networking stack and is complemented by the ZigBee specifications, which cover higher layers (i.e., network and application layers). The standard defines
a Beacon-Enabled (BE) mode and a Non-Beacon-Enabled (NBE) mode. The BE mode relies on a periodic superframe bounded by beacons, which are special synchronization frames
generated by coordinator nodes. In the NBE mode, on the other hand, nodes do not use
beacons and use an unslotted CSMA/CA algorithm for frame transmissions.
Below, we provide a short description of the CSMA/CA algorithm used by sensor nodes
to transmit data to their coordinator node.
Algorithm. Upon receiving a data frame to be transmitted, the MAC layer at the sensor
node performs the following steps.
1. A set of state variables is initialized, namely the number of backoff stages carried out
for the on-going transmission (NB = 0) and the backoff exponent (BE = macMinBE).
2. A random backoff time is generated and used to initialize a timer. The backoff time is
obtained by multiplying an integer number uniformly distributed in [0, 2
BE−1
] by the
the duration of the backoff period (Dbp). As soon as the timer expires the algorithm
moves to step 3.
3. A Clear Channel Assessment (CCA) is performed to check the state of the wireless
medium.
(a) If the medium is free (i.e., the channel is perceived as free for TCCA time, the
frame is immediately transmitted.
(b) If the medium is busy, state variables are updated as follows: NB = NB+1 and
BE = min(BE+1,macMaxBE). If the number of backoff stages has exceeded
the maximum allowed value (i.e. NB > macMaxCSMABackoffs), the frame is
dropped. Otherwise, the algorithm falls back to step 2.
1
In this homework, we will focus on evaluating the performance of the unslotted CSMA/CA
scheme in perfect channel conditions.
First Question [50pt]
Implement a module named SensorNodeCSMACA that uses the unslotted CSMA/CA scheme
of the NBE IEEE 802.15.4 summarized in the previous section to access the physical channel
and transmit a data packet. In order to test your module, implement a SinkNodeCSMACA
module that is connected to a SensorNodeCSMACA node and collects its data. For the sake
of simplicity, you may assume that if the node finds the channel free at a particular time,
the CCA is passed (i.e., TCCA = 0 for the purpose of assessing whether the channel is
free). Furthermore, assume the channel as binary (i.e., busy/free). Your SensorNodeCSMACA
module will obtain the CSMA/CA parameters through the par(·) command passed through
the .INI file of your simulation.
Second Question [25pt]
Next, extend your simulation by implementing a network composed by a SinkNodeCSMACA
node connected to n SensorNodeCSMACA nodes (i.e., star network). We will study a periodic
traffic scenario where every node will try to transmit 1000 packets, where each packet is
transmitted every T = 5s. Sensors are assumed to be synchronized with each other (i.e.,
every node will attempt to transmit a packet at k · T, k ≥ 0). Implement the following
performance metrics:
• Average delivery ratio (DR): defined as the ratio between the number of packets that
are correctly received by the sink node and the total number of packets generated by
all nodes in the network. It characterizes the reliability of the CSMA/CA scheme.
• Average packet latency (L): defined as the average value of packet latency during the
reporting of an event. It characterizes the timeliness of the CSMA/CA scheme. The
packet latency during the reporting of an event is defined as the average delay of
successful packet transmissions occurred during the reporting of an event. You should
include the packet duration Dp in the computation of the packet latency.
You should generate the following graphs:
1. Average delivery ratio (expressed as a percentage value between 0% and 100%) as a
function of network size n ∈ {5, 10, 20, 30, 40, 50}.
2. Average packet latency (expressed in ms) as a function of the network size n ∈
{5, 10, 20, 30, 40, 50}.
3. Average delivery ratio (expressed as a percentage value between 0% and 100%) as a
function of macMaxCSMABackoffs ∈ {1, 2, 3, 4}. Assume n = 30, macMinBE = 3,
and macMaxBE = 4.
2
4. Average delivery ratio (expressed as a percentage value between 0% and 100%) as a
function of macMinBE ∈ {1, 2, 3, 4}. Assume n = 30, macMaxCSMABackoffs = 2,
and macMaxBE = 4.
To generate each point of your graphs, you should perform at least 10 independent
replicas, where each replica consists of at least 1000 packet attempts. Your graphs should
also show, for each simulation result, confidence intervals with 95% confidence level. (Hint:
Use a common parameter that you can change dynamically at simulation time to represent
the channel shared by the n nodes.)
Third Question [25pt]
Implement the following performance metric:
• Average energy consumption (E): defined as the average value of the energy consumed
during the reporting of the event. The energy consumed during the reporting of an
event is the average energy consumed by all nodes in the network during the reporting of
the event. It characterizes the energy efficiency of the CSMA/CA scheme. You should
only consider energy consumption during transmission (TX) and reception (RX) mode.
Therefore, consider (i) energy spent during CCA assessment (i.e., radio is in RX mode
during that time); (ii) energy spent during transmission (i.e., radio is in TX mode
during packet transmission.
Generate the following graph:
1. Average energy consumption (expressed in mJ) as a function of network size n ∈
{5, 10, 20, 30, 40, 50}.
Simulation Parameters
The following set of CSMA/CA parameter values should be used, unless specified otherwise:
macMinBE = 3, macMaxBE = 4, macMaxCSMABackoffs = 2. To run realistic simulations,
we will comply to the IEEE 802.15.4 specifications and use the radio parameters of the
Texas Instrument (TI) 2.4 GHz IEEE 802.15. 4/ZigBee-ready RF Transceiver CC2420 [1].
Therefore, assume as Dbp = 0.00032s, TCCA = Dbp/20·8 = 0.000128s. For energy consumption,
use PRX = 56.4mW and PT X = 49.5mW, and ignore power consumption in sleep mode.
Since the CC2420 radio has a bitrate of 250 Kbit/s, and the full IEEE 802.15.4 is long 133
bytes [2], set the packet duration to Dp = 4.256ms.
Important Remarks and Grading Policy
The primary target of this homework is to elicit your creativity in writing network simulations. Therefore, some of the specific implementation details have been omitted on purpose,
and left for you to decide as you are writing your own simulation code. Furthermore, as
explained in class, simulations depend on a lot of factors (i.e., choice of initial seeds, number
of runs, etc). Your homework will be evaluated by following these criteria, listed in order of
importance:
3
1. Functionality (i.e., does your simulation code compile and run?)
2. Correctness of implementation (i.e., is your implementation of the IEEE 802.15.4
CSMA/CA algorithm correct?)
3. Correctness of results (i.e., did you get reasonable results in terms of delivery ratio,
latency, and energy consumption, as compared to the research paper indicated by the
instructor? Do your graphs have confidence intervals? Are those intervals reasonable
in terms of span?)
4. Modularity of implementation (i.e, is your code understandable, modular, and clean?
Did you write enough comments in the code?)
For point 3 (correctness of results), we will use the results shown in Figure 1 as reference. Although a perfect match is not (and should not) be expected, your simulation results
should be close to the ones in Figure 1. In particular, we will consider point 3 fully accomplished if the results are within 10% of the results in Figure 1.
Submission Instructions
Please submit your assignment using Canvas. You should upload the following files: (i)
Three .ZIP files containing your simulation code (including every .INI and .NED files) for
each homework question; (ii) A .PDF file containing (a) your simulation results, including
confidence intervals, both in graphical and tabular form; and (b) a short description of your
simulation results and your interpretation of the results you obtained. Late submissions
will not be accepted.
References
[1] Texas Instruments, “CC2420: 2.4 GHz IEEE 802.15. 4/ZigBee-ready RF Transceiver,”
Available at http://www.ti.com/lit/ds/symlink/cc2420.pdf, 2006.
[2] G. Shi and K. Li, “Fundamentals of ZigBee and WiFi,” in Signal Interference in WiFi
and ZigBee Networks, pp. 9–27, Springer, 2017.
4
(a) Delivery ratio vs Number of Nodes (b) Average Latency vs Number of Nodes
(c) Delivery ratio vs macMaxCSMABackoffs (d) Delivery ratio vs macMinBE
(e) Average energy consumption vs Number of
nodes
Figure 1: Benchmark results for Homework I.
5