Available via license: CC BY 4.0
Content may be subject to copyright.
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.DOI
Integration of Steerable Smart Antennas
to IETF 6TiSCH Protocol for High
Reliability Wireless IoT Networks
SERCAN KULCU1, SEDAT GORMUS1, AND YICHAO JIN2
1Department of Computer Engineering, Karadeniz Technical University, Trabzon, 61080 Turkey. (e-mail: {sedatgormus, sercan.kulcu}@ceng.ktu.edu.tr.)
2Bristol Research and Innovation Lab., Toshiba Europe Ltd., Bristol, BS1 4ND U.K. (e-mail: yichao.jin@toshiba-bril.com)
Corresponding author: Sercan Kulcu (e-mail: sercan.kulcu@ceng.ktu.edu.tr)
“This work is supported by the Scientific and Technological Research Council of Turkey (TÜB˙
ITAK) Research Fund under Grant
118E289.”
ABSTRACT Steerable directional antennas are increasingly utilised to improve the overall performance of
the traditional wireless sensor networks. Steerable directional antenna based networking solutions increase
the network capacity by providing a longer range of transmission and reduced interference as compared
to networking solutions with omni-directional antennas. However, the use of smart antennas requires
complex algorithms and such algorithms may not be easily leveraged in low power Internet of Things
(IoT) networks. This study presents mechanisms for integrating low complexity smart antenna solutions
into IETF 6TiSCH protocol with the aim of creating scalable and reliable industrial IoT networks. The
solution defines extensions to MAC layer and scheduling mechanisms of IETF 6TiSCH protocol to enable
its seamless integration with low complexity steerable smart antennas. The results of this study show that
smart antenna enabled 6TiSCH protocol stack outperforms the legacy 6TiSCH stack in terms of data delivery
performance especially in high density scenarios.
INDEX TERMS 6TiSCH, Network Formation, Smart Antennas, Wireless Sensor Networks.
I. INTRODUCTION
IOT networks are envisaged to be employed in a wide
range of applications such as habitat monitoring, disaster
relief, smart metering, asset tracking, etc. Such wireless IoT
networks are spread over a large area and they are expected
to operate for years using limited energy resources. Such net-
works are generally composed of low power, low cost devices
and characterised by intermittent connectivity. Therefore,
IoT applications require computationally-light, and energy-
efficient protocols that can run on such constrained devices
[1].
In various scenarios, the data generated by the IoT appli-
cations are delivered to a central node over a multi-hop wire-
less mesh network for analysis and decision making. This
requires the nodes in the network make simultaneous data
transfer. However, such data transmissions create in-network
interference and lead to packet drops. This is especially true
when the network density is high and available bandwidth
resources are limited which is the case for most of the low
power IoT networks. There are protocols and mechanisms
to address in-network interference issues, but such solutions
have limited spatial re-use capabilities which may limit the
network throughput [2]. On the other hand, packet drops
due to interference can be mitigated by utilising directional
antenna systems instead of omni-directional communications
[3].
IEEE 802.15.4 standard is the accepted technology for
low-power sensor networks [4] which shares the same unli-
censed band with other technologies such as WiFi. Therefore,
the interference caused by these technologies decrease the
performance of IEEE 802.15.4 networks [5]. Electronically
steerable parasitic array radiator (ESPAR) antennas may
provide a solution for mitigating interference related perfor-
mance degradation in IEEE 802.15.4 based sensor networks
[6]. ESPAR antenna can steer the radiation pattern electron-
ically on the fly towards the desired direction [7]. However,
using dynamically steerable directional antennas in 802.15.4
based WSNs, increases the complexity of the protocol stack
design. Furthermore, the wireless channel conditions are
time-varying requiring additional adaptive mechanisms to
take advantage of directional communication [8].
The IEEE 802.15.4e Time-slotted Channel Hopping
VOLUME 4, 2016 1
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
(TSCH) MAC layer is adopted for addressing the high re-
liability requirements of industrial applications. TSCH pro-
vides multi-channel and synchronized communication to en-
able high reliability one-hop communication. IETF Working
Group 6TiSCH builds a mesh networking solution on top of
the TSCH MAC layer to create Industrial IoT networks [9].
6TiSCH networks build a schedule for the nodes to follow.
Nodes decide their activities by following this schedule,
that enables them to go into low energy modes during the
unscheduled instances [10].
The configurable directional communication opens up a
wealth of opportunities. Particularly, it is of interest to
integrate such low complexity antenna solutions to Time
Synchronized Medium Access Control (MAC) mechanism
where the antenna direction can be configured as a chan-
nel access parameter. In this work, we introduce Steerable
Antenna Agnostic 6TiSCH Solution (SAA6) which inte-
grates a low complexity steerable antenna solution to IETF
6TiSCH protocol. However, the use of steerable antennas
is not limited to 6TiSCH networks. It can be used in other
solutions like GALLOP protocol [11] that has very low and
deterministic latency, very high reliability and low-overhead
signaling mechanism for large networks by providing multi-
path communication.
To exploit the benefits of directional antennas, 6TiSCH
protocol stack must be redesigned. The redesigned protocol
requires the direction information to a destination/source
node while transmitting/receiving to improve communica-
tion reliability. Voigt et al. showed that the Received Signal
Strength Indicator (RSSI) can be used to accurately estimate
the best transmit/receive direction [12].
The proposal in this paper has the following novel features:
•A novel autonomous steerable antenna aware schedul-
ing mechanism for a faster neighbor discovery.
•RSSI based optimal transmit direction finding algorithm
to suppress interference within the network.
•A method to reduce shared cell collisions by utilising di-
rectional communication which has a significant impact
on 6TiSCH network scalability.
This paper presents the relevant background work in smart
antenna systems, and scheduling algorithms in Section II.
Section III introduces the developed antenna model. The
proposed solution and analysis are presented in Section IV.
Performance evaluation and test results are given in Section
V. Finally, Section VI outlines the concluding remarks and
possible future directions.
II. RELATED WORK
A. SMART ANTENNA SYSTEMS
Smart antennas can be classified as switched beam, and adap-
tive array antennas. A switched beam antenna has several
fixed beam patterns which can behave as an omni-directional
antenna to broadcast a frame [13]. On the other hand, adap-
tive array antennas can steer the beam in any desired direction
and cancel interfering signals from undesired directions. But,
they can not transmit omni-directionally. Therefore, such
antenna solutions require new mechanisms to integrate them
into 6TiSCH protocol stack.
ESPAR antenna is one of the most popular and easily
integrated structure providing beam steering capability. The
ESPAR antenna has a single radiating active element sur-
rounded by a series of passive elements connected to variable
reactants. It is possible to change the direction of the main
antenna beam by changing the values of the passive elements
electronically [6]. ESPAR antennas can be used with WSN
nodes as shown in [14]. ESPAR antennas can achieve a 360◦
beam scan with different discrete angular steps (30◦[15],
45◦[6], 60◦[14], [16]). Such antennas have different gains
depending on their designs (7.5 dBi [16], 8 dBi [14], 9 dBi
[15]).
B. 6TISCH PROTOCOL
6TiSCH Working Group (WG) proposes a set of specifica-
tions to provide IPv6 compliant networking solution over
the TSCH mode [9]. Two steps must be completed for an
operational 6TiSCH network: As the first step, a globally
synchronized network is built by exchanging Enhanced Bea-
con (EB) frames. To be able to communicate in a scheduled
time slot, nodes must be tightly synchronized. TSCH network
coordinator1broadcasts EB frames to advertise the presence
of the network. Nodes follow a scan process to associate
to the network. After receiving an EB frame, node learns
the minimal schedule and gets synchronized to the network.
Following the synchronisation, the nodes in the network are
configured as a directed tree rooted at the sink by exchanging
DODAG Information Object (DIO) frames as defined by the
RPL routing protocol [17]. After this step the node has the
necessary information to interact with the rest of the network.
Here, it is assumed that the join process does not require a
subsequent authentication step.
TSCH splits the time into time slots, and multiple time
slots form a slotframe which repeats over time. A slotframe
is represented as a matrix where each cell being defined by
a pair of time slot and channel offset. In a 6TiSCH network
each node is responsible for creating a schedule that defines
their interaction with the neighboring nodes. During each
time slot a node may perform the following actions; transmit,
listen, or sleep. All communications are orchestrated by this
schedule that contains two types of cells: (1) Shared cell al-
lows a slotted-aloha with contention. For example, Hard cells
in 6TiSCH are often shared cell type for signaling purpose.
Once a node joins the network, it communicates in shared cell
to start a negotiation with its neighbors. (2) Dedicated cell
allows interference-free and collision-free communication to
provide high reliability. These resources are allocated by a
mechanism in a centralized, distributed or autonomous way
[18].
1RPL Root node takes up the network coordinator role in 6TiSCH
networks [17].
2VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
In WSNs, nodes are deployed without any prior informa-
tion about the neighbors and the network topology. There-
fore, neighbor discovery process is the first step of self-
organization. During this process, each node identifies all
of the nodes which it can communicate with directly [19].
To achieve this goal, each node broadcasts periodic neigh-
bor discovery messages. Neighbors should be discovered as
quickly as possible to create an accurate picture of the local
topology from the nodes point of view.
Neighbor discovery process is trivial when omni-
directional antennas are used, since a broadcast transmission
can ideally reach all of the neighbors at once. However, use
of directional communication makes the neighbor discovery
more challenging [19], [20] because of the following reasons:
(1) Each node should estimate when and where to point its
antenna to discover each potential neighbor, (2) The limited
coverage of the directional beam requires the steps to be
repeated until the whole azimuth is covered. To be able
to discover a neighbor node, the receiving node must be
listening at the right time, on the right channel and in the right
direction with the transmitter node. This requires additional
mechanisms to minimize the synchronization time of the
6TiSCH networks.
C. 6TISCH SCHEDULING MECHANISMS
The IEEE 802.15.4e-TSCH standard does not define specific
scheduling policies for the transmission of the control frames.
The 6TiSCH WG addresses such scheduling challenges by
relying on a “minimal profile” standardized in the RFC8180
[21]. This profile defines a common schedule for all the
nodes during the network formation. 6TiSCH handles the
scheduling functionality via two components: the 6TiSCH
Operation Sublayer (6top) Protocol (6P) and the Scheduling
Function (SF). 6top Protocol (6P) is a negotiation protocol
that allows neighbor nodes to allocate cells between each
other [22]. Scheduling Function (SF) is a separate entity
and supports different cell allocation policies for different
network scenarios. Furthermore, 6TiSCH WG standardizes
the Minimal Scheduling Function (MSF) for interoperability
purposes [9].
Naturally, the scheduling of the control messages plays a
key role in the formation phase. The majority of the research
efforts focus on the resource allocation in order to guarantee
delivery of the messages after the initial formation phase
[18], [23]. Most of these solutions assume that all of the
nodes are part of the network and have discovered the optimal
routes to the sink. The algorithms related to EB scheduling
are generally shared cell based solutions and they assume
that nodes have only omni-directional antennas [24]. 6TiSCH
minimal configuration is the standard strategy for the alloca-
tion of cells for the control frame transmission [21]. In cases
where one shared cell is not enough especially for dense
or large networks, extra cells can be allocated as proposed
in [25]. Also, the frequency of the control messages can be
optimized taking the number of neighbors into account [26],
[27].
0°
90°
180°
270° −15 dBi
−10 dBi
−5 dBi
0 dBi
5 dBi
(a) Radiation pattern
0°
90°
180°
270°
80 m
160 m
OMNI TR
OMNI INT.
SAA6 TR
SAA6 INT.
(b) Transmission/Interference range
FIGURE 1. Simulated radiation pattern, and transmit/interference ranges
according to the active direction of the antenna.
Orchestra [28] and ALICE [29] are autonomous schedul-
ing mechanisms, which are not designed with a focus on the
network formation. These mechanisms build a schedule con-
sisting of a set of slotframes dedicated for MAC, routing and
application layers separately. While Orchestra run on a sin-
gle channel, ALICE uses three different channel offsets for
three slotframes. These mechanisms are not designed from a
frequency diversity perspective. However, these mechanisms
can be utilised for allocating cells between individual pairs
of nodes using nodes’ identifiers as proposed in this paper.
III. STEERABLE ANTENNA MODEL
In this study, a new antenna model is created to emulate the
use of steerable antennas in 6TiSCH networks. The radiation
pattern of the steerable directional antenna is given in Fig.
1(a). The maximum gain of the antenna is assumed to be 5
dBi according to [6], [16], [30]. Gain of the receiver antenna
is calculated according to (1).
G(θ)[dB] = 10(1 + cos(θ)) −15.(1)
where θis the angular difference between the radiation
direction of the transmitter and receiver antennas. 10 and
15 values are chosen considering radiation pattern which is
given in Fig. 1(a). Simulated antenna increases the trans-
mission range around 30 m at 0◦by providing 5 dB gain,
and has around 40 m shorter range at 180◦as shown in Fig.
1(b). Transmission range is calculated according to (2) where
antenna gain is added to the standard path loss formula.
VOLUME 4, 2016 3
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
RSS =PT X +PP L −10Klog10(d12
d0
) + G(θ12).(2)
where PT X is the transmission power, PPL is the path loss
at the reference distance d0,Kis the path loss exponent,
d12 is the distance between nodes 1 and 2, G(θ12)is the
directional gain of the receiver antenna as given in (1).
The receiver sensitivity level used in the evaluation of the
proposed mechanism is chosen according to the TI exp5438
platform which comes with a CC24202radio which has as
sensitivity level of -95 dBm. Based on [31], the reference
distance (d0) is set to 1m, the reference path loss is set to
-52 dBm, and the path loss exponent is set to 2.5.
IV. IMPROVING NETWORK FORMATION
In this section, methods for improving network formation are
introduced.
A. NEIGHBOR DIRECTION ESTIMATION
A node, which has a steerable directional antenna, must turn
its antenna in the right direction prior to communication.
Therefore, nodes should know in which direction they shall
point their antennas to reach their neighbors. In this section,
an RSSI based neighbor direction estimation mechanism is
outlined. The best direction for each neighbor is kept in
a list, and this list is updated considering the RSSI values
retrieved from the physical layer for every successful frame
exchange. Each entry of the list contains MAC ID of the
neighbor node, the current best antenna direction and highest
RSSI value among the frames received from that neighbor.
This list is updated for every successful frame received from
the neighbor. When a node receives a frame, it extracts the
transmitter id from the frame which is uses to search the
list, and compare current RSSI value with the recorded RSSI
value to update the best direction. Neighbor list is stored at
the MAC layer by all of the nodes in the network. The details
of the algorithm are given in Algorithm 1 which is called only
after receiving a packet.
During the network discovery, nodes set their antennas
toward a random direction at the start of each slotframe rather
than following a steering pattern or configuration. Therefore,
it is not possible to define an upper bound on how long the
discovery stage will take. Furthermore, nodes may choose
non-optimal parents for a certain period of time until the
network reaches a stable stage. The direction information is
updated for every successful frame exchange between neigh-
boring nodes. And the largest received RSSI value represents
the best direction to communicate with a neighbor. When
a new packet is received, the last RSSI value is compared
with the recorded largest RSSI value for the neighbor and the
direction information is updated.
This mechanism does not require node to transmit/receive
probing frames. The control frames of the 6TiSCH network
2CC2420 radio [online]. Website http://www.ti.com/product/cc2420 [ac-
cessed May 2021]
Algorithm 1 Update Neighbor List
Input: packet keeps the last taken packet
Data: list_neighbors keeps the updated neighbor list
1: id ←extract_id(packet)
2: n←get_neighbor_from_list(id)
3: if n== null then
4: n←create_a_new_neighbor()
5: n.id ←id
6: n.dir ←current_direction
7: n.rssi ←current_rssi
8: insert_to_the_neighbor_list(n)
9: else if current_rssi > n.rssi then
10: n.dir ←current_direction
11: n.rssi ←current_rssi
(EB, RPL, DATA) are used to find the best direction. If a node
receives fewer frames from one of its neighbors, it updates
the list less frequently for that neighbor. Due to performance
and resource constraints, the number of neighbors that nodes
can have is limited to 8.
During the synchronisation phase, the antenna direction
is changed randomly for each slotframe. Due to the narrow
antenna coverage area, neighbor discovery process takes
longer as compared to the omnidirectional antenna solution.
On the other hand, the synchronisation time can be shortened
by allocating autonomous cells through which multicast mes-
sages are duplicated to provide a better opportunity for the
near by nodes to get synchronised to the network.
The ability to dynamically steer the antenna and transmit
in the desired direction can improve the communication
performance of the node as long as the radiation direction
is chosen properly. The selected antenna direction toward
a particular neighbor may not be stable over time due to
variations in the wireless communication channel. Therefore,
adaptive protocols are required to dynamically change the
best radiation direction according to the channel conditions.
Algorithm 2 presents a mechanism that updates the best
radiation direction dynamically for the destination node. To
achieve this goal;
•Transmitter node transmits the frame by setting the
antenna direction toward the best direction of the in-
tended destination. Please note that the direction of the
destination is kept in the neighbor table which is needed
by the unicast traffic.
•The receiver node sets its antenna direction to the best
direction but with an angular margin (i.e., from -15◦to
+15◦)
•The receiver node updates its neighbor table, if the new
direction provides a better RSSI value as compared to
the current direction as given in Algorithm 1.
•Antenna direction is set to a random direction in shared
cell since multicast frames are transmitted without the
knowledge of the receiver.
Data packets generally consist of unicast messages trans-
4VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
mitted from the child node towards its parent. However, mul-
ticast messages should be received by all of the neighbors of
the transmitting node. Therefore, multicast messages provide
an opportunity for the transmitting nodes to discover the
direction of their neighbors via the allocated autonomous
cells. As a result, this repetitive process enables nodes to
adjust their radiation direction to increase the their radio link
qualities.
The RSSI value changes with the relative directions of the
transmitter and receiver antenna pairs. Fig. 3 shows the RSSI
change over time for four pairs of nodes where the topology
of the network for the selected pairs of nodes is given in Fig.
2. The two reference lines in the figure present (a) if omni-
directional antenna is used, (b) if directional antenna is used,
but they point directly at each other (best case). Initial RSSI
values for the communicating pairs of the nodes are greater
than -95 dBm, since the receiver sensitivity value of the radio
is set to this value. However, initial RSSI value differ for each
pair due to both their relative positions in the network and
their initial antenna orientations. The nodes in the simulated
scenario are located at 40 m equal distance to each other in a
grid formation. Therefore, RSSI values converges to -82 dBm
according to RSS equation in (2) indicating that the optimal
antenna orientations have been established after a certain
discovery time period. As it can be seen from the Fig. 3, the
node pairs improve their observed RSSI values over time by
adjusting their antenna directions using the Algorithm 2.
Algorithm 2 Set Antenna Direction
Input: asn current absolute slot number
Data: list_cells keeps the updated scheduled cell list
1: cell, id ←get_next_scheduled_cell(asn)
2: if cell.type == T R ANS M IT then
3: dir ←get_best_direction(id)
4: set_antenna_direction(dir)
5: else if cell.type == REC EIV E then
6: dir ←get_best_direction(id)
7: margin ←get_random_margin(-15, 15)
8: set_antenna_direction(dir + margin)
9: else if cell.type == S H ARED then
10: dir ←get_random_direction()
11: set_antenna_direction(dir)
Nodes must receive frames successfully to be able to mea-
sure the RSSI value with their neighbors. However, they can
not send frames before getting synchronized to the network
except the root node. The nodes within the transmission
range of a synchronized node join the network after receiving
EB frames. EB frames are sent periodically via shared cells
toward a random direction. In order to optimize the antenna
direction, a certain number of frame exchanges must be
carried out to estimate the best direction to a neighbor. These
constraints result in higher elapsed time to reach the best
transmission direction (i.e., max RSSI value), and due to the
randomness involved in the process, node pairs reach the op-
timal radiation direction at different instances as presented in
A
B
CD
E F
G
H
FIGURE 2. Network topology used during the Cooja based performance
evaluations.
−95
−90
−85
−80
0 10 20 30 40 50 60 70
RSSI (dBm)
Elapsed time (min)
Node A − Node B
Node C − Node D
Node E − Node F
Node G − Node H
Omni−directional
Steerable (best case)
FIGURE 3. RSSI v.s. Elapsed time. (According to topology given in Fig. 2)
Fig. 3. Smaller elapsed time can be achieved by transmitting
EB and RPL frames more frequently, which will cause the
nodes to consume more energy. Of course, this process may
be optimised further. But, this is not studied in this paper and
will form part of our future work.
B. UTILIZING DIRECTION INFORMATION IN SHARED
CELLS
Three conditions must be satisfied by the two nodes equipped
with steerable directional antennas for a successful commu-
nication, where X shows the transmitter and Y shows the
receiver node.
•Node X must transmit in the direction of Node Y,
•Node Y must listen in the direction of Node X,
•No other node must transmit in the direction of Node Y
on the same frequency channel.
These conditions can be satisfied by building a proper
schedule. In the case of omni-directional antenna, the receive
probability of a neighboring node within the range of the
transmitting node can be assumed to be 100% (P(R) = 1)
VOLUME 4, 2016 5
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
provided that there is no interference in the communica-
tion medium. For the steerable antenna case, each node
must cover full 360◦area to find a feasible neighbor to
communicate with. However, the nodes must select random
directions to broadcast beacons during the discovery phase.
In this case, the EB reception probability of a node becomes
P(R) = α
2π
β
2π, where αand βare the beamwidths of the
transmitter and receiver antennas, respectively.
Following the discovery phase, nodes at the edge of the
network will have neighbors towards the direction of the
network coordinator (i.e. center of the network). Such nodes
are named as “leaf nodes” and they do not need to scan the
whole 360◦. In this case, the receive probability of an RPL
parent node can be estimated as P(R) = α
2π
n
n+1 for the
leaf nodes, where nis the number of neighbors. Of course,
the leaf node may need to point its antenna toward a random
direction occasionally to let new nodes to join the network.
(n+ 1) is used to factor this in to the analysis.
C. TRANSMITTING BROADCAST MESSAGES VIA
DEDICATED CELLS
Broadcast messages such as EB, DIO, and DIS are transmit-
ted over the shared cells in 6TiSCH protocol. Furthermore,
these cells are also used for the initial 6P negotiations [9].
Therefore, shared cells are indispensable resources for the
network formation, responsible for the transmission of im-
portant control messages.
Integration of a steerable directional antenna to 6TiSCH
protocol poses new challenges since a node can not reach
all of its neighbors in the shared cell due to the directional
coverage of the antenna as given in Fig. 1(a). To overcome
this challenge, one solution is to increase the number of
shared cells by 2π
α
2π
β. In this case, each neighbor is covered
with a shared cell dedicated to the direction of the neigh-
bor. Predictably, this solution makes inefficient use of the
bandwidth and energy resources. Another solution to this
challenge is to allocate additional dedicated cells between
neighbors instead of shared cells. Dedicated cells can be ne-
gotiated between the nodes and they provide efficient use of
resources since they are expected to be collision free. When a
node transmits a broadcast frame over the dedicated cell, the
broadcast message is kept in the MAC buffer until the node
sends the broadcast frame towards all of its neighbors via the
dedicated cells.
D. AUTONOMOUS SCHEDULING OF BROADCAST
MESSAGES
Two slotframes with different lengths which are mutually
prime are created for the proposed solution. One of the
slotframes is used for data traffic, and the other one is used
for the control messages, which is shown in detail in Fig. 4.
A simple topology consisting of 4 nodes is given in Fig.
4(a). The slotframe created by the distributed scheduling
algorithm for this topology is shown in Fig. 4(b). Blue
color indicates the cells scheduled for transmission, and
green color indicates the cells scheduled for reception. Since
these dedicated cells are scheduled for the pairs of nodes,
they create a reliable and interference free communication
link between these nodes. The red colored cells represent
the shared cells, and provide contention based access to
the medium. Since the directional antenna cannot transmit
omnidirectionally, a second slotframe is needed to be able
to transmit multicast messages to all neighboring nodes. The
slotframe generated autonomously is given in Fig. 4(c).
Autonomously scheduled slotframe is used for control
messages and does not require negotiations with the neigh-
boring nodes. Instead, it relies on (3) to select a timeslot and
a channel offset to transmit towards a neighboring node. The
slotframe used for data traffic is scheduled by the Scheduling
Function Zero (SF0) as introduced in the IETF 6TiSCH
standard.
slotT X (n) = mod(hash(IDn, I Do), Lauto)
slotRX (n) = mod(hash(IDo, I Dn), Lauto)
channel(n) = mod(hash(I Do, I Dn), Lhop)
(3)
where IDnis the neighbor’s id, IDois the own id, Lauto
is the autonomous slotframe length and Lhop is the length of
the channel hopping sequence.
When a node is associated with the network, it allocates the
necessary cells by looking at its neighbor table. Here, each
node autonomously allocates one Tx and one Rx cell for each
of its neighbors. This process is summarized in Algorithm
3. Moving the shared cell traffic to dedicated cells in this
manner provides a new approach as compared to literature
[9], [18], [21], [23].
Algorithm 3 Autonomous Scheduling
Input: packet keeps last taken packet
Data: list_neighbors keeps updated neighbor list
1: id ←extract_id(packet)
2: n←get_neighbor_from_list(id)
3: if n== null then
4: sl_tx ←HASH_func1(id, my_id) based on (3)
5: sl_rx ←HASH_func1(my_id, id) based on (3)
6: ch ←HASH_func2(id, my_id) based on (3)
7: if is_cell_scheduled(sl_tx, ch)then
8: unschedule_cell(sl_tx, ch)
9: if is_cell_scheduled(sl_rx, ch)then
10: unschedule_cell(sl_rx, ch)
11: else
12: if is_cell_scheduled(sl_tx, ch) == F ALSE then
13: schedule_cell(sl_tx, ch)
14: if is_cell_scheduled(sl_rx, ch) == F ALS E then
15: schedule_cell(sl_rx, ch)
E. RESOURCE USAGE ANALYSIS
A section of the network topology used in the test scenario
is given in Fig. 5. As seen in the figure, the same network
has fewer number of hops when nodes are equipped with
6VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
1
43
2
(a) Topology
ASN
1
2
3
4
5
6
7
8
9
10
…
Slot ID
1
2
3
4
5
6
7
1
2
3
…
Node 1
S
Rx
S
…
Node 2
S
Rx
Rx
Tx
S
Rx
…
Node 3
S
Tx
S
Tx
…
Node 4
S
Tx
S
…
7 slots slotframe
Time
(b) Slotframe for data traffic (SF0)
ASN
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
…
Slot ID
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1
2
3
…
Node 1
Rx
Tx
Rx
…
Node 2
Rx
Tx
Tx
Rx
Rx
Tx
Rx
Tx
…
Node 3
Tx
Rx
Tx
…
Node 4
Tx
Rx
…
19 slots slotframe
Time
(c) Slotframe for control messages (Autonomous)
FIGURE 4. Illustration of the slotframes generated by SF0 and autonomously, in a 4-node network topology.
1.hop 2.hop 3.hop 4.hop 1.hop 2.hop 3.hop 4.hop
Omni-directional antenna case Steerable antenna case
5.hop
6.hop
7.hop
FIGURE 5. A section of the 81 node network topology used in the
performance evaluations of the proposed solution. The black node is the RPL
Root located at the center of the network. White and grey colored nodes
represent different hop levels.
steerable antennas. Establishing connectivity with the RPL
Root node with fewer hops implies that fewer cells are
needed for communication leading to a better utilisation of
the bandwidth resources. Furthermore, directional antennas
have better interference suppression properties as compared
to omni-directional antennas meaning that fewer frequency
channels can be used in the network.
In the evaluation scenario, each node produces data pack-
ets periodically and transmits them to the RPL Root. So,
each node requires to allocate a number of Tx cells for these
transmissions as given in (4).
C=d(DtLs)/Ite(4)
where Dtis the time slot duration, Lsis the data slotframe
length (slotframe which is used for the transmission of data
packets), Itis the transmission interval (indicates the data
generation period), and Cis the cost which shows the number
of required cells for each node to handle the generated data
traffic. The total number of Tx cells required by all of the
nodes at any hop level is given in (5).
Ch(x)=Nh(x)C
Ch(x−1) =Ch(x)+Nh(x−1)C(5)
where Ch(x)is the total number of the required Tx cells
for the nodes at hop level x,Nh(x)is the number of nodes at
hop level x. Therefore, the total number of Tx cells allocated
for the whole network can be found as given in (6)3.
Cs=
x
X
i=1
Ch(i)(6)
The total number of Tx and Rx cells should be proportional
since each node is required to act as a relay in the network.
In addition to standard shared cells, SAA6 solution makes
use of autonomously allocated cells for broadcast messages
to speed up the synchronisation process. The number of extra
cells allocated by autonomous scheduling function per node
is given as (7),
Ca=2NnLs
La
(7)
where Nnis the number of 1-hop neighbors, Lsis the
slotframe length which is used for the data traffic, and La
3Please note that the topology requires a converge-cast traffic pattern
having a negative impact on the nodes closer to the RPL Root.
VOLUME 4, 2016 7
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
is the slotframe length which is used by the autonomous
scheduling function. For each neighbor, 1 TX and 1 RX
cells are allocated to enable a faster network synchronisation
by transmitting copies of the control frames via these extra
cells. Since the autonomous slotframe length is larger than
that of the data slotframe length, we divide Lsby Lawhen
calculating the cost of allocating these extra autonomous
cells.
The cost introduced by the extra cells in relation to the cells
allocated by the standard 6TiSCH scheduler can be calculated
using (8). The number of cells required for the steerable
antenna solution can be calculated as the sum of Csand Ca.
However, omni-directional antenna solution does not require
the Cacells.
Ct=Csif omni
Cs+Caif steerable (8)
As shown in the example scenario presented in Fig. 5, each
node in the network except leaf nodes has a total number
of 8 neighbors. For each neighbor, 1 Tx and 1 Rx slots are
allocated in the autonomous slotframe. Therefore, each node
needs additional 16 slots for SAA6 mechanism to make the
discovery process quicker. Of course, the cost of introducing
extra cells to the network to speed up the synchronisation pro-
cess will vary depending on the topology of the network. The
topology in this section is used as an performance indicator
for the real life networks. In the evaluation tests, autonomous
slotframe length is taken as 397 and data slotframe length is
taken as 51. There is not a standard value for the length of
a slotframe. It is chosen according to network topology and
traffic needs to provide minimum network latency and energy
consumption. When multiple slotframes are used together,
their lengths need to be prime among themselves so that they
do not overlap often. In case they overlap, autonomous slots
have higher priority in the proposed solution.
Since the steerable antenna can reach the root node with
fewer number of hops due to higher antenna gains toward the
receiving node, a system utilizing steerable antennas needs
fewer cells compared to the omnidirectional antenna solu-
tion. In this case, the required number of cells for the network
formations is reduced about 25% as indicated by (5) and (6).
However, the steerable antenna solution presented here needs
a fixed number of additional autonomous cells regardless
of the transmission interval to speed up the synchronisation
process as shown in (8).
When the transmission interval is greater than 750 ms,
each node must allocate minimum one cell due to the ceil-
ing function in (4). As the transmission interval value gets
smaller, the cost of autonomous cells decreases, since Ca
values is proportional to the network size only. When the
transmission interval is chosen as 1 s, the required number of
cells for the steerable antenna solution is about 8% more than
that of the solution utilising omnidirectional antennas. But,
for the presented topology, when the transmission interval is
set to 0.5 s, the number of the cells required by the proposed
solution is 6% less than that of omnidirectional solution due
TABLE 1. Cost Analysis
ItComni
t(8) Csteer
t(8) Csteer
t/Comni
t
1 296 240 + 80 = 320 1.08
0.5 592 480 + 80 = 560 0.94
to the gains achieved by the reduced hop numbers in the
network as presented in Table 1. In summary, the proposed
solution provides an efficient solution for networks with a
heavy network traffic.
V. PERFORMANCE EVALUATION
Contiki OS4is used to evaluate the performance of the
SAA6 mechanism. Contiki OS supports IETF 6TiSCH pro-
tocol stack, and works integrated with a network simulator
called COOJA which simulates the radio medium and allows
the emulation of firmwares produced by several toolchains.
COOJA simulator supports omni-directional antennas by de-
fault [31]. Therefore, the total received signal strength (RSS)
value calculation in COOJA has been adapted using the path
loss formula given in (2) to create a deterministic path loss
model.
A modified COOJA simulator with steerable support is
used for the performance evaluation of the proposed solution
whose performance is analyzed for three distinct scenarios;
(1) Nodes are equipped with only an omni-directional an-
tenna (base scenario), (2) Nodes are equipped with a steer-
able directional antenna and transmit broadcast messages in
a random direction, (3) Nodes are equipped with a steerable
directional antenna and transmit broadcast messages over au-
tonomously allocated cells as unicast by setting the antenna
in the best direction (SAA6 mechanism). The performance
results are presented in terms of synchronization latency,
shared cell collision rate, packet delivery ratio, and energy
consumption metrics.
A. NETWORK TOPOLOGY AND ASSUMPTIONS
In this section, the network topology and the assumptions
used in the test scenarios are introduced.
•Each node is assumed to be static, and equipped with
only one radio having a fixed transmit power (0 dBm).
At any time, a node can either be only in transmit or
listen mode, but not both.
•Each node is equipped with a steerable directional an-
tenna, that can scan 360◦by pointing the antenna at
different directions.
•Root node is equipped with three switchable directional
antennas [13], where each of them can scan 120◦area.
All of the switchable antennas can be activated to emu-
late an omni-directional antenna behaviour.
•The parameters used in the tests are given in Table 2.
4Contiki: The Open Source OS for the Internet of Things [online]. Website
http://www.contiki-os.org [accessed May 2021]
8VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
5
10
25 49 81
Elapsed time (minutes)
Node count
omni
steerable
SAA6
(a) Synchronization latency
0
20
40
60
80
100
20 30 40
Collision rate (%)
Distance between nodes (m)
Nch=16 (omni)
Nch= 8 (omni)
Nch= 4 (omni)
Nch=16 (SAA6)
Nch= 8 (SAA6)
Nch= 4 (SAA6)
(b) Collision rate
20
40
60
80
100
1 2 3 4 5
PDR (%)
Transmission Interval (s)
omni
SAA6
(c) Transmission Interval v.s. PDR
40
60
80
100
1 4 16
PDR (%)
Number of Available Channels
omni
SAA6
(d) Frequency Channels v.s. PDR
0.5
0.75
1
70 80 90 100
Energy Consumption (mWh)
Receive Probability (%)
omni
SAA6
(e) Energy consumption
FIGURE 6. Results for 81 Node Network
B. SYNCHRONIZATION LATENCY TESTS
Synchronization time is the elapsed time from the trans-
mission of the first EB frame from the root node until the
synchronization of the entire network. At this time, all nodes
would have received at least one EB frame, and get synchro-
nized to its time source neighbor by learning the minimal
schedule.
Fig. 6(a) shows the variation of the synchronization time
for different antenna configurations with growing network
size when the number of available channels is set to 4. As
expected, the required time to achieve network synchroniza-
tion increases with the network size, but at different rates for
each configuration. Omni and SAA6 mechanisms present a
lower synchronization time where SAA6 achieves a faster
network synchronisation as compared to the pure steerable
antenna solution at the expense of slightly increased energy
consumption due to autonomous cells. The standard solution
makes use of an omni-directional antennas. On the other
hand, the steerable antenna solution can not achieve the same
synchronization performance due to its narrow coverage area
in the neighbor discovery phase. The proposed SAA6 solu-
tion achieves a better performance as compared to the stan-
dard steerable antenna solution since it transmits the control
messages multiple times over the allocated autonomous cells.
In this case, SAA6 shows similar performance with the
solution utilizing omni-directional antennas. The increasing
VOLUME 4, 2016 9
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
TABLE 2. Test Parameters
Parameter Value
Slotframe length 51
Autonomous Slotframe length 397
Timeslot duration 15 ms
Shared cell count (omni) 3
Shared cell count (steerable) 1
Transmit range (omni) 50 m
Interference range (omni) 100 m
Transmit range (steerable) from 15 to 80 m
Min EB period 16 s
Max EB period 50 s
Min DIO interval 4.096 s
Max DIO interval 1048.576 s
Inter-node distance 40 m
Network dimension 320 m x 320 m
Test duration 2 hours
number of neighbors creates an opportunity for the nodes
equipped with omni-directional antennas to receive broadcast
messages. Unfortunately, steerable antenna solutions do not
scale as well as networks utilising omni-directional antennas
due to scanning latency attached to directional antennas. The
simulation results clearly indicate that the SAA6 scheme per-
forms better than the pure steerable antenna solution which
does not use Algorithm 1 and Algorithm 2, and improvement
is more prominent for the larger networks.
C. COLLISION RATE TESTS
The distance parameter indicates the distance of a node with
its 1-hop neighbor. Therefore, shorter the distance between
nodes, the denser the network. As shown in Fig. 6(b), dense
network has higher collision rate. Steerable antennas de-
crease the collision probability more than 20% in the shared
cell without the need for an extra method such as adjusting
EB interval dynamically. The collisions have a negative im-
pact on the synchronization times. The results indicate that
it is possible to significantly improve the shared cell success
probability by employing directional antennas.
Since all synchronized nodes use the same frequency
channel in the shared cell, the number of channels does not
have a considerable impact on the collision rate. However,
the contention in the shared cell increases in dense networks
resulting in an increased collision rate. From the analytical
and simulation results, it is observed that smart antenna
solutions reduce the collisions in the shared cells of the
6TiSCH networks significantly especially in scenarios with
high network density.
D. PACKET DELIVERY RATIO TESTS
Fig. 6(c) shows the Packet Delivery Ratio (PDR) results for
different transmission intervals when a total of 4 frequency
channels are utilized. Lower PDR values are observed for
shorter transmission intervals due to the lack of resources
according to (4). SAA6 provides up to 30% higher PDR for
the evaluated scenarios, since it utilizes spatial separation
provided by steerable antennas. Fig. 6(d) shows the PDR
for different number of available frequency channels when
a fixed transmission interval of 3 s is used. When the number
of available frequency channels is limited, it is possible that
the scheduler allocates cells with the same channel offsets.
In this case, the simultaneous transmissions interfere each
other more often. On the other hand, having a smart steerable
antenna solution creates a third dimension that cell resources
can be scheduled in allowing a higher throughput in the
network.
E. ENERGY CONSUMPTION TESTS
Fig. 6(e) presents the average energy consumed at each node.
The receive probability in the figure presents the probability
of a node receiving a frame destined to itself and this value
changes according to the distance between the source and
destination nodes. At the maximum distance, where the node
is withing the communication range, the receive probability
is at its minimum. On the other hand, if the source and
destination nodes are next to each other, the receive proba-
bility becomes 100%. As can be seen in the Fig. 6(e), energy
consumption increases as the receive probability decreases.
The lower receive probability results in increased number of
dropped frames which causes the nodes to consume more
energy because of the retransmission of the frames. SAA6
provides up to 20% energy saving due to increased commu-
nication range which improves the receive probability.
VI. CONCLUSION AND FUTURE WORK
In this paper, adaptation of an electronically steerable an-
tenna to 6TiSCH protocol is investigated by considering
performance metrics such as synchronization time, PDR, and
energy consumption. The 6TiSCH protocol is modified to
accomodate the steerable antenna solution. The proposed
Steerable Antenna Agnostic 6TiSCH Solution (SAA6) is ex-
tensively evaluated and the results verify that SAA6 outper-
forms omni-directional antenna based solution in all aspects
especially in terms of reliability, and network throughput.
It is proved that 6TiSCH protocol has the potential to
support directional and smart antennas. This study provides
useful insights for the future industrial IoT applications
utilizing directional communications. As part of the future
work, simulation results will be validated on real hardware
under real-life conditions. Determining the best direction will
be improved by utilising machine learning algorithms. Also,
integration of multi-path communication with the steerable
antenna solutions promises to be a way forward to improve
network latency which will be investigated as a future re-
search topic.
REFERENCES
[1] D. Minoli, K. Sohraby, and B. Occhiogrosso, “Iot considerations, require-
ments, and architectures for smart buildings-energy optimization and next-
generation building management systems,” IEEE Internet of Things J.,
vol. 4, no. 1, pp. 269–283, Feb. 2017.
[2] T. Okazawa, J. Ma, N. Komuro, Y. Choi, Z. Li, T. Pei, and H. Sekiya,
“An opportunistic directional mac protocol with pulse/tone exchange in
wireless ad-hoc networks,” Wirel. Pers. Commun., vol. 111, no. 2, pp.
1187–1205, Mar. 2020.
10 VOLUME 4, 2016
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
[3] A. Varshney, T. Voigt, and L. Mottola, “Using directional transmissions
and receptions to reduce contention in wireless sensor networks,” in
Real-World Wireless Sensor Networks - Proc. of the 5th Int. Workshop,
REALWSN 2013, Como, Italy, Sept. 19-20, 2013, ser. Lecture Notes in
Electrical Engineering, K. Langendoen, W. Hu, F. Ferrari, M. Zimmerling,
and L. Mottola, Eds., vol. 281. Springer, 2013, pp. 205–213. [Online].
Available: https://doi.org/10.1007/978-3-319-03071-5_21
[4] M. R. Palattella, N. Accettura, X. Vilajosana, T. Watteyne, L. A. Grieco,
G. Boggia, and M. Dohler, “Standardized protocol stack for the internet of
(important) things,” IEEE Commun. Surv. Tutor., vol. 15, no. 3, pp. 1389–
1406, 2013.
[5] R. Musaloiu-E. and A. Terzis, “Minimising the effect of wifi interference
in 802.15.4 wireless sensor networks,” Int. J. Distrib. Sens. Netw., vol. 3,
no. 1, pp. 43–54, Dec. 2008.
[6] M. Burtowy, M. Rzymowski, and L. Kulas, “Low-profile espar antenna
for rss-based doa estimation in iot applications,” IEEE Access, vol. 7, pp.
17 403–17 411, 2019.
[7] H. Liu, S. Gao, and T. Loh, “Electrically small and low cost smart antenna
for wireless communication,” IEEE Trans. Antennas Propag., vol. 60,
no. 3, pp. 1540–1549, Mar. 2012.
[8] X. Liu, A. Sheth, M. Kaminsky, K. Papagiannaki, S. Seshan, and
P. Steenkiste, “Dirc: Increasing indoor wireless capacity using directional
antennas,” SIGCOMM Comput. Commun. Rev., vol. 39, no. 4, p. 171–182,
Aug. 2009. [Online]. Available: https://doi.org/10.1145/1594977.1592589
[9] X. Vilajosana, T. Watteyne, T. Chang, M. Vuˇ
cini´
c, S. Duquennoy, and
P. Thubert, “Ietf 6tisch: A tutorial,” IEEE Commun. Surv. Tutor., vol. 22,
no. 1, pp. 595–615, 2020.
[10] S. Duquennoy, A. Elsts, B. A. Nahas, and G. Oikonomo, “Tsch and 6tisch
for contiki: Challenges, design and evaluation,” in 13th Int. Conf. on
DCOSS, June 2017, pp. 11–18.
[11] A. Aijaz, A. Stanoev, and U. Raza, “Gallop: Toward high-performance
connectivity for closing control loops over multi-hop wireless networks,”
in Proc. of the 27th Int. Conf. on RTNS, ser. RTNS ’19. New York, NY,
USA: Association for Computing Machinery, 2019, p. 176–186. [Online].
Available: https://doi.org/10.1145/3356401.3356413
[12] T. Voigt, L. Mottola, and K. Hewage, “Understanding link dynamics
in wireless sensor networks with dynamically steerable directional
antennas,” in Wireless Sensor Networks - 10th European Conf., EWSN
2013, Ghent, Belgium, February 13-15, 2013. Proceedings, ser. Lecture
Notes in Computer Science, P. Demeester, I. Moerman, and A. Terzis,
Eds., vol. 7772. Springer, 2013, pp. 115–130. [Online]. Available:
https://doi.org/10.1007/978-3-642-36672-7_8
[13] S. Gormus and S. Kulcu, “Enabling space time division multiple access in
ietf 6tisch protocol,” Turk. J. Elec. Eng. & Comp. Sci., vol. 27, no. 6, pp.
2779–2791, 2019.
[14] A. Kausar, H. Mehrpouyan, M. Sellathurai, R. Qian, and S. Kausar,
“Energy efficient switched parasitic array antenna for 5g networks and iot,”
in 2016 LAPC, Nov. 2016, pp. 1–5.
[15] M. Rzymowski, P. Woznica, and L. Kulas, “Single-anchor indoor localiza-
tion using espar antenna,” IEEE Antennas Wireless Propag. Lett., vol. 15,
pp. 1183–1186, 2016.
[16] M. Rzymowski and L. Kulas, “Influence of espar antenna radiation pat-
terns shape on ppcc-based doa estimation accuracy,” in 22nd Int. Conf.
MIKON, May 2018, pp. 69–72.
[17] R. Alexander, A. Brandt, J. Vasseur, J. Hui, K. Pister, P. Thubert, P. Levis,
R. Struik, R. Kelsey, and T. Winter, “RPL: IPv6 Routing Protocol for
Low-Power and Lossy Networks,” RFC 6550, Mar. 2012. [Online].
Available: https://rfc-editor.org/rfc/rfc6550.txt
[18] D. D. Guglielmo, S. Brienza, and G. Anastasi, “Ieee 802.15.4e: A survey,”
Comput. Commun., vol. 88, pp. 1 – 24, 2016.
[19] L. Chen, Y. Li, and A. V. Vasilakos, “On oblivious neighbor discovery
in distributed wireless networks with directional antennas: Theoretical
foundation and algorithm design,” IEEE/ACM Trans. Netw., vol. 25, no. 4,
pp. 1982–1993, 2017.
[20] B. El Khamlichi, D. H. N. Nguyen, J. El Abbadi, N. W. Rowe, and
S. Kumar, “Collision-aware neighbor discovery with directional antennas,”
in 2018 Int. Conf. on ICCNC, Mar. 2018, pp. 220–225.
[21] X. Vilajosana, K. Pister, and T. Watteyne, “Minimal IPv6 over the TSCH
Mode of IEEE 802.15.4e (6TiSCH) Configuration,” RFC 8180, May
2017. [Online]. Available: https://rfc-editor.org/rfc/rfc8180.txt
[22] Q. Wang, X. Vilajosana, and T. Watteyne, “6TiSCH Operation Sublayer
(6top) Protocol (6P),” RFC 8480, Nov. 2018. [Online]. Available:
https://rfc-editor.org/rfc/rfc8480.txt
[23] R. T. Hermeto, A. Gallais, and F. Theoleyre, “Scheduling for ieee802.15.4-
tsch and slow channel hopping mac in low power industrial wireless
networks: A survey,” Comput. Commun., vol. 114, pp. 84 – 105, 2017.
[24] J. Vera-Pérez, D. Todolí-Ferrandis, S. Santonja-Climent, J. Silvestre-
Blanes, and V. Sempere-Payá, “A joining procedure and synchronization
for tsch-rpl wireless sensor networks,” Sensors, vol. 18, no. 10, 2018.
[25] C. Vallati, S. Brienza, G. Anastasi, and S. K. Das, “Improving network
formation in 6tisch networks,” IEEE Trans. Mobile Comput., vol. 18, no. 1,
pp. 98–110, Jan. 2019.
[26] D. De Guglielmo, A. Seghetti, G. Anastasi, and M. Conti, “A performance
analysis of the network formation process in ieee 802.15.4e tsch wireless
sensor/actuator networks,” in 2014 IEEE ISCC, June 2014, pp. 1–6.
[27] M. Vuˇ
cini´
c, T. Watteyne, and X. Vilajosana, “Broadcasting strategies in
6tisch networks,” Internet Technol. Lett., vol. 1, no. 1, p. e15, 2018.
[28] S. Duquennoy, B. Al Nahas, O. Landsiedel, and T. Watteyne,
“Orchestra: Robust mesh networks through autonomously scheduled
tsch,” in Proc. of the 13th ACM Conf. on Embedded Networked
Sensor Systems, ser. SenSys ’15. New York, NY, USA: Association
for Computing Machinery, 2015, p. 337–350. [Online]. Available:
https://doi.org/10.1145/2809695.2809714
[29] S. Kim, H. Kim, and C. Kim, “Alice: Autonomous link-based cell schedul-
ing for tsch,” in 18th ACM/IEEE Int. Conf. on IPSN, Apr. 2019, pp. 121–
132.
[30] H. Liu, S. Gao, and T. Loh, “Electrically small and low cost smart antenna
for wireless communication,” IEEE Trans. Antennas Propag., vol. 60,
no. 3, pp. 1540–1549, Mar. 2012.
[31] J. Schandy, S. Olofsson, L. Steinfeld, and T. Voigt, “Improving sensor
network convergecast performance with directional antennas.” in Proc. of
the 2019 Int. Conf. on EWSN, 2019, pp. 13–22.
SERCAN KULCU received the B.Sc. degree in
computer engineering and the M.S. degree in
information systems from Hacettepe University,
Ankara, Turkey, in 2006 and 2010, respectively.
He is now a Ph.D. student at Karadeniz Tech-
nical University, Trabzon, Turkey. From 2005 to
2015, he worked as a Research and Development
Engineer at SDT company. Since 2015, he has
been a Lecturer with the Computer Technologies
Department, Giresun University, Giresun, Turkey.
His research interests include the embedded systems, Internet of Things
(IoT), and the IETF 6TiSCH protocol.
SEDAT GORMUS received the B.Sc. degree in
computer science from Karadeniz Technical Uni-
versity, Trabzon, Turkey, in 1999, and the Ph.D.
degree in wireless networking from the Univer-
sity of Bristol, U.K., in 2008. Consequently, he
joined the Telecommunications Research Labo-
ratory, Toshiba Research Europe Ltd., in 2009,
where he was the Leader of the Cross Layer Team
until 2014. He was a Research and Development
Engineer in several Turkish and British technology
companies. Since 2014, he has been with the Computer Hardware Chair,
Computer Engineering Department, Karadeniz Technical University, where
he was an Assistant Professor, became an Associate Professor in 2020. His
current research activities include the Internet of Things (IoT), IoT edge
computing, high reliability IoT communications, IoT security, and the IETF
6TiSCH protocol. In 2004, he received the Ph.D. Scholarship as part of the
OSIRIS Project at the University of Bristol.
VOLUME 4, 2016 11
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3125144, IEEE Access
S. Kulcu, S. Gormus, and Y. Jin: Integration of Steerable Smart Antennas to IETF 6TiSCH Protocol for High Reliability Wireless IoT Networks
YICHAO JIN is a wireless expert for industrial
IoT networks and systems with more than 30
publications and 15+ patents. He is currently lead-
ing the industrial wireless network programme in
Toshiba Bristol Research and Innovation Labora-
tory which focuses on R&D activities for Reli-
able wireless technologies for Industry 4.0. Large
IoT networks with millions of devices for smart
city or smart metering applications as well as
collaborative robotic technologies for drones and
AGVs. He is an IET Royal Chartered Engineer and have over 12 years
of industrial experiences in design and develop reliable industrial wireless
communication protocols where it can’t afford to fail.
12 VOLUME 4, 2016
