Implementation growing problem of scarcity of free spectrum and

Implementation of Relay Hopper Model for
Reliable Communication of IoT Devices in LTE
Environment through D2D Link
Anish Pradhan, Soumi Basu, Sreetama Sarkar, Saptarshi Mitra and Sanjay Dhar Roy
Department of Electronics and Communication Engineering
National Institute of Technology, Durgapur
West Bengal, India
Abstract—Internet of things (IoT) is an emerging technology
that can bring about a revolution in our day-to-day lives. Deviceto-device
(D2D) communication is an energy and spectral efficient
solution to the growing problem of scarcity of free spectrum and
overloading of base stations in cellular networks. The concept
of using D2D communication in IoT networks has already been
proposed. But the reliability of the links established by reusing
the resource blocks allocated to licensed cellular users has not
been investigated earlier. In this paper, our objective is to provide
connectivity between IoT devices and the associated gateway
using D2D communication ensuring that links established are
reliable, and at the same time improve performance by providing
connectivity to the maximum number of IoT devices. We propose
a three step approach to achieve this. In the first step, the IoT
devices satisfying the Quality of Service (QoS) constraint are
selected and matched with appropriate reuse candidates, that is,
the cellular user equipments (CUEs) by an optimum resource
allocation scheme. Next, link reliability of these links are computed
and weak links are discarded. Finally, the disconnected IoT
devices are rerouted to the IoT gateway via IoT devices possessing
strong links following a relay hopper model. Simulation results
indicate significant improvement in the network performance
metrics, namely, access rate and sum throughput of IoT devices.
Index Terms—D2D, IoT, QoS, Resource Allocation, Linkreliability,
Relay Hopper Model
Internet of things (IoT) is a revolutionary concept where
‘things’ that have the ability to sense and communicate
collaborate on a network to make important decisions. In a
cellular IoT network, there are a number of IoT devices with
sensors attached to them. The data collected through these
devices from the environment are sent to the LTE Evolved
Node B (eNodeB or eNB) via an IoT Gateway (IoT-GW)1.
Device to Device (D2D) communication provides a number
of advantages such as higher data rate 2, reduced energy
consumption 34, spectrum efficiency 5 reduced latency
6 etc over conventional methods like WiFi, bluetooth etc. The
concept of using D2D communication to provide connectivity
between IoT devices and the IoT gateway has been proposed
in 5.
We have adopted the concept of offloading Machine to
Machine (M2M) Communication 7. In this method, a device
lying in poor cell coverage area can establish connection to
another device via a third device (also called offloader or relay)
which lies inside the coverage area of the cell in a strategic
location between two devices intending to connect.
We propose a novel three step approach to provide reliable
communication between the IoT devices and their associated
gateway. In the first step, the IoT devices that satisfy the
Quality of Service (QoS) requirement are selected and these
IoT devices are assigned the resources of licensed CUEs
which they can reuse. We use maximum bipartite matching
as the resource allocation scheme so as to provide service to
the maximum number of IoT devices present in the system.
The reliability of the links thus established is investigated in
the next step and the weak links are discarded. Finally, the
disconnected IoT devices are mapped with the ones having
strong links such that the former acts as a hopper and the
latter as relay to establish connectivity between IoT devices
with weak links and the IoT gateway. So, the contributions of
this paper can be summarized as: (a) proposition of a resource
allocation scheme that considers not only QoS requirement
but also reliability of the route 8 between IoT devices and
IoT-GW, (b) extension of effective IoT coverage area by
introducing relays, (c) improved network performance with
increase in access rate, sum throughput and average link
reliability of IoT devices.
The remaining sections are organized as follows: The system
model is discussed in Chapter II. The problem formulation
and theoretical analysis of the problem is given in Chapter
III. Chapter IV posits the Simulation Results and Discussion.
Finally, the paper has been concluded in Chapter V.
The proposed system model is based on reliable reuse of
uplink resources of CUEs in LTE environment to enable the
IoT devices to communicate through the IoT gateway. In the
simulation environment 5, a cell with several CUEs and an
eNodeB at the center is considered. A cluster of IoT devices
with IoT gateway at its center is introduced in that environment
to facilitate the communication between the IoTs. The IoT
gateway acts as the data condenser of all the IoTs and it
sends the data to the eNodeB by using the available radio
resources from the CUEs. The whole bandwidth is divided
equally among the CUEs such that they are allocated only one
Resource Block (RB) to avoid interference among themselves.
Let there be N number of CUE in the cell and they are denoted
Cn|n = 1, 2, 3, …, N
and K number of IoT devices which are represented by
Ik|k = 1, 2, 3, …, K
We have used a distance based exponential path-loss model5
Fig. 1: System Model
described as
Prx = P0 · (dtx,rx)
· Ptx (1)
where Prx is the received power, Ptx is the transmitted power,
P0 is the path-loss constant, dtx,rx is the distance between the
transmitter and the receiver and ‘a’ is the path-loss coefficient.
Following 5 we can see, links can be established between
IoT device and IoT-GW reusing PRBs from CUEs but the
author did not consider the reliability of links. Here we
have calculated the reliability of the established links with
the method described in 8 and tried to improve them by
proposing a relay hopper based system model. We have taken
a three step approach towards this solution:
A. Selection of IoT devices based on QoS requirements5:
Due to the constraint that an average fixed value of received
power is maintained at eNodeB, the transmission power of n
CUE becomes
PCT x(n) = PRx,eNB
P0 · (dCn,eNB)?a
where n varies from 1 to N, PRx,eNB is the received power
at eNodeB, dCn,eNB = distance between n
th CUE device and
eNodeB. Interference at eNodeB, InteNB(k) is caused by IoT
devices trying to use the same radio resource of CUE. To fulfill
QoS constraint of CUE, the SIR condition of CUE is
P0 · (dCn,eNB)
· PCT x(n)
P0 · (dIk,eNB)?a · PIT x(k)
? SIRCn (3)
PIT x(k) ?
PRx,eNB · (dIk,eNB)
P0 · SIRCn
where PIT x(k) is the power transmitted by k
th IoT device
that is using the RB of n
th CUE, dIk,eNB is the distance
between k
th IoT device and eNodeB and SIRCn is the target
SIR condition for CUEs. As the maximum transmission power
of the IoT devices are fixed (PImax), the value of PIT x(k)
will be
PIT x(k) = (PRx,eNB·(dIk,eNB)
if dIk,GW ? Dn
PImax else
where Dn is the minimum distance between k
th IoT device
and eNodeB after which the transmit power of the k
th IoT
device reaches its maximum limit.
Dn =

P0 · PImax · SIRCn
PRx,eNB 1/a
Similarly, interference at IoT-Gw, IntGW (n) is caused by
CUEs whose radio resources are being used by IoT devices.
This is given by,
PRx,GW (k)
IntGW (n)
P0 · (dIk,GW )
· PIT x(k)
P0 · (dCn,GW )?a · PCT x(n)
? SIRIk (7)
where PRx,GW (k) is received power at IoT-GW, from k
IoT device. dIk,GW and dCn,GW are distances from IoT-GW
to k
th IoT Device and n
th CUE device respectively. So,
PIT x(k) ?
PRx,eNB · SIRIk · (dIk,GW )

RatioIk !a
RatioCn =
RatioIk =
?(k) = PRx,eNB · SIRIk · (dIk,GW )
? = SIRIk · SIRCn (10)
If dIk,GW ? Dn, then from equations (5) and (8) we get,

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

RatioIk !a
? ? (11)

RatioIk !a
? ?(k) (12)
Equations (11) and (12) are the necessary conditions to determine
which IoT device can use the PRB of which reuse
candidates and this information is stored. Then matching of
CUE and IoT devices can be done with an optimum resource
allocation scheme. In our case, we have used maximum
bipartite matching so that maximum number of CUE-IoT
device pairing is possible that will contribute in improved
access rate of the system.
B. Elimination of IoT devices with weak links:
After initial pairing of CUE and IoT devices, we calculate
the reliability of the links between paired IoT devices and
IoT-GW 8 by the following equation.
RelIk = exp
(dIk,GW )

SINRGW (k) = PIT x(k)
NT + Interference from respective CUEs
and NT = Thermal Noise. This reliability is defined as the
probability of successful transmission over the given route
8. We have assumed transmit power of each IoT devices
as their maximum transmit power (PImax) considering the
complexity of calculations and its almost negligible impact
on the results. After calculation of the reliabilities of those
direct links from IoT devices to IoT-GW, we discard the
weaker links by thresholding them against a minimum fixed
probability value (Relth). Now, we have a number of IoT
devices which have direct strong links to the IoT-GW and
a handful of remaining IoT devices which satisfy the QoS
constraints and were paired with CUEs by bipartite matching
but can not provide reliable transmission of data as per the
requirement of real time environment. Discarding weak links,
on one hand, abates wastage of resources, but on the other
hand, access rate potentially decreases. We can further increase
access rate by introducing a relay-hopper system as discussed
in the next section.
C. Rerouting the eliminated IoT devices by choosing relayhopper
With the help of a link threshold value based on distance between
connected and disconnected IoT devices with ability to
access PRB, a new link can be established between them where
the previously disconnected IoT device named as hopper can
connect to IoT-GW through one of the previously connected
IoT devices named as relay with reliable link strength. So the
new link reliability between a probable hopper candidate k3
and IoT-GW through a probable relay candidate k2 can be
calculated as,
RelIk3,Ik2 = exp
(dIk2,GW )
The information about all the remaining and discarded IoT
devices is stored. If the total link reliability of any of this
relay-hopper pair exceeds a threshold value (Relth), then they
become eligible for being a relay or a hopper. After that, we
used bipartite matching to yield maximum number of relayhopper
pairs. This proposed system can increase access rate
as more reliable links are established. The trade off done here
is link sharing between two IoT devices, which do not impose
serious threat as the IoT devices do not require an entire PRB
for their communication as found from data.
We simulated the proposed model described in section
II in MATLAB. Access rate, average link reliability, sum
throughput of IoT devices are used as the performance metrics
to evaluate the system. Access rate is the ratio of number
of allocated IoT devices to the total number of IoT devices
whereas sum throughput is the aggregate of throughputs of all
the allocated IoT devices and average link reliability is average
value of reliability of all the existing links between connected
IoT devices and IoT-GW.
Average link reliability =
RelIk (15)
Sum Throughput =

1 +
PRx,GW (k)
NT + IntGW (n)
where N1 = Number of reuse CUE candidates, K1 = Number
of resource allocated IoT devices. As IoT devices transmit data
usually at a small amount for some limited slot of time 9,
the effect of adding a hopper on calculating the throughput
of that relay IoT device is not considered here. Our proposed
scheme is compared with the scheme discussed here 5.
Fig. 2: Access Rate vs number of IoT devices before and after
introducing Relay-Hopper mechanism where dIoTD,GW =
150m, dGW,eNB = 500m
From Fig.2 and Fig.3, we can see the improvement of
Access rate and sum throughput with the relay-hopper model.
It is observed that the improvement increases with increase
in number of IoT devices. Fig.4 shows that average link
reliability is improved significantly after implementing relayhopper
Fig. 3: Sum throughput vs number of IoT devices before
and after introducing Relay-Hopper mechanism where
dIoTD,GW = 150m, dGW,eNB = 500m
Fig. 4: Average link reliability vs number of IoT devices
before and after introducing Relay-Hopper mechanism where
dIoTD,GW = 150m, dGW,eNB = 500m
In this paper, we have studied a new method of interconnecting
IoT devices using a two-hop scheme on device
to device communication technology in LTE environment
using the spectrum more efficiently. We have calculated the
link strength between IoT devices and Gateway to tackle the
problem of bad links by implementing Relay-Hopper pairs.
In device to device communication of IoT devices, the link
reliability parameter has not been considered before but that
is of paramount importance in real time data traffic. The
increment of sum throughput of IoT devices after introducing
Relay-Hopper pairs has also been illustrated. In future, this
work can be extended by considering more than two hops
and finding out the optimum solution by examining the trade
off between number of hops and performance improvement.
Considering mobility of the devices and incorporating approTABLE
I: Simulation Parameters 5
Parameters Value
Uplink System Bandwidth 100 MHz
Number of RBs 500
Bandwidth of each RB 180 KHz
Number of CUEs(N) 500
Number of IoTDs(K) 1-100% of CUEs
Cell Radius 1000 m
SIRCn 25 dB
SIRIk 20 dB
Path loss exponent, a 4
P0 10?2
Noise Power Spectral Density -174 dBm/Hz
PImax 23 dBm
Link reliability threshold, Relth 0.5
priate power control scheme may also be introduced for better
system efficiency.
1 A. Orsino, I. Farris, L. Militano, G. Araniti, S. Andreev, I. Gudkova,
Y. Koucheryavy, and A. Iera, “Exploiting d2d communications at the
network edge for mission-critical iot applications,” in European Wireless
2017; 23th European Wireless Conference; Proceedings of. VDE, 2017,
pp. 1–6.
2 D. Feng, L. Lu, Y. Yuan-Wu, G. Li, S. Li, and G. Feng, “Deviceto-device
communications in cellular networks,” IEEE Communications
Magazine, vol. 52, no. 4, pp. 49–55, apr 2014. Online. Available:
3 B. Wang, L. Chen, X. Chen, X. Zhang, and D. Yang, “Resource
allocation optimization for device-to-device communication underlaying
cellular networks,” in 2011 IEEE 73rd Vehicular Technology Conference
(VTC Spring). IEEE, may 2011. Online. Available:
4 A. Osseiran, K. Doppler, C. Ribeiro, M. Xiao, M. Skoglund, and
J. Manssour, “Advances in device-to-device communications and network
coding for imt-advanced,” 01 2009.
5 M. M. Elmesalawy, “D2d communications for enabling internet of things
underlaying lte cellular networks,” Journal of Wireless Networking and
Communications, vol. 6, no. 1, pp. 1–9, 2016.
6 A. Asadi, Q. Wang, and V. Mancuso, “A survey on device-to-device
communication in cellular networks,” IEEE Communications Surveys
Tutorials, vol. 16, no. 4, pp. 1801–1819, Fourthquarter 2014.
7 B. Panigrahi, H. K. Rath, R. Ramamohan, and A. Simha, “Energy
and spectral efficient direct machine-to-machine (m2m) communication
for cellular internet of things (IoT) networks,” in 2016 International
Conference on Internet of Things and Applications (IOTA). IEEE, jan
2016. Online. Available:
8 A. Khandani, J. Abounadi, E. Modiano, and L. Zheng, “Reliability and
route diversity in wireless networks,” IEEE Transactions on Wireless
Communications, vol. 7, no. 12, pp. 4772–4776, dec 2008. Online.
9 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 communications surveys & tutorials, vol. 15,
no. 3, pp. 1389–1406, 2013.