IJE TRANSACTIONS A: Basics Vol. 34, No.

4, (April 2021) 825-831

International Journal of Engineering

Journal Homepage: www.ije.ir

Routing Protocols for IoT Applications based on Distributed Learning

R. S. Razaa,b, M. S. Al-Ani*b
a Technical College of Informatics, Sulaimani Polytechnic University, Sulaimani, Iraq
b College of Science and Technology, University of Human Development, Sulaymaniyah, Iraq

PAPER INFO A B S T R A C T

Paper history: The IPv6 routing protocol for lossy and low power networks (RPL) was introduced in March 2012 by
Received 29 January 2021 the Internet Engineering Task Force (IETF) as the standard routing protocol for the Internet of Things
Received in revised form 19 February 2021 (IoT). Since that time, it has had various applications in IoT. Despite meeting the IoT network necessities
Accepted 02 March 2021 by RPL, some unanswered issues have not been devised primarily for IoT usages. However, gathering a
large amount of data from these networks with videos and images typically leads to traffic congestion in
the network's central part. For providing a solution for this issue, the content-centric routing (CCR)-
Keywords:
based RPL is proposed in the present study, where the content specifies the routing pathways. It is
Contextual
Internet of Things possible to attain a larger data aggregation ratio by routing the relevant data to the middle relaying nodes
Protocol for the process. Thus, effective traffic is generated in the network. Subsequently, latency is significantly
Learning reduced. Moreover, energy use is decreased on wireless communication. More integration was conducted
Routing between IETF RPL protocol and CCR, using the MATLAB platform. Finally, according to simulated
and implemented results, the CCR-based RPL behavior based on the high packet transfer rates is
improved, and the numbers of dead nodes are reduced. High energy efficiency and low delay rates are
obtained in communication using the proposed routing method.
doi: 10.5829/ije.2021.34.04a.08

1. INTRODUCTION1 allowing large-scale connectivity among IoT policies.

However, the combination of the IoT ecosystem and
Recently, there has been a high interest in the distributed current wireless networks results in many issues,
computing in wireless networks, especially in the including self-organizing operation, coincidence with
emerging pattern of Internet of Things (IoT) human-kind systems, and limited communication sources
communications, where IoT instruments are supplied [7, 10]. Besides, the IoT devices are often machine-type
with independent processes, communication and storage ones, which have a significant difference from
capacities [1]. The important consequences include (i) a conventional human-type tools, such as smartphones
more economical approach with locally decreased data based on performance needs, memory, computation,
volume by in-network processing, (ii) just the processed traffic patterns, and energy constraints [11]. Additionally,
outcomes are routed rather than all raw data, directed the IoT tools require short packets, ultrareliable
across a costly (multi-hop) wireless network. Therefore, transmissions, and low latency. Thus, re-planning of the
bandwidth and energy are saved, the latency is decreased, current wireless networks is required for solving these
and the network life is prolonged in an IoT network with IoT challenges.
restricted resources [2]. A distributed learning approach with the Low Power
The IoT influences various application domains, Networks (RPL) protocol (CCR-based RPL protocol)
including smart grids [3, 4], home automation [5], drone- was presented by Azari and Cavdar [12] to solve this
based systems [6], healthcare systems [7], and industrial problem to collect data in multi-hop IoT networks
supervision [8, 9]. It is essential to use new wireless effectively as represented in Figure 1. Appropriate
technologies to support innovative IoT applications and allocation of restricted resources is achieved using the

*Corresponding Author Institutional Email: [email protected]

(M. S. Al-Ani)

Please cite this article as: R. S. Raza, M. S. Al-Ani, Routing Protocols for IoT Applications based on Distributed Learning, International Journal
of Engineering, Transactions A: Basics Vol. 34, No. 04, (2021) 825-831
 R. S. Raza and M. S. Al-Ani / IJE TRANSACTIONS A: Basics Vol. 34, No. 4, (April 2021) 825-831 826

it is required to have a changeable network, the overhead

must be confined to a minimum rate, and multi-hop
routing must be used. Thus, selecting the energy-efficient
routing protocols must be done cautiously based on
application necessities and system requirements. This
work attempts to deal with the network lifetime analysis
for the reactive routing protocols in heterogeneous and
homogeneous systems.
Aman et al. [15] investigated the RPL acceptance in
Figure 1. The architecture of the RPL routing domain the Cooja simulator by comparing the real networks’
findings. They showed higher energy consumption by
both the sink and client nodes in a real network compared
distributed method concerning the IoT tools' Quality of to the Cooja simulation calculation. The packet delivery
Service (QoS) needs [13]. These tools should have the rate (PDR) was found to be lower in actual scenarios and
capability to separately obtain their communication is considerably decreased by the network distance and
resources as supposing their regular communication with size. However, interference is not involved from external
the root node is not practical, given their severe resource wireless, signal reflection, and asymmetric links. This
constraints. Moreover, the root in a massive IoT cannot model is yet extensively employed in different works.
manage the communication resources in all tools A new uplink bandwidth assignment algorithm was
promptly. Thus, the distribution of resources is essential presented by Mardani et al. [9] for highly dense 4G
for all assignment occasions in the IoTs, considering the networks of the human-to-human (H2H)/ machine-to-
limited capacities of the IoT tools in terms of computation machine (M2M) services in a co-existing setting. The
and memory. Deployment of the IoT over the available algorithm utilized interval type-2 fuzzy logic for dealing
networks with limited communication resources would with system ambiguities. The intelligent type-2 fuzzy
be confined by a resource assignment framework that algorithm specifies the optimum bandwidth ratio for
meets the pre-described requirements. every M2M/ H2H service flow. To determine the correct
The current research focuses on two critical bandwidth ratio for every type of flow, the solution used
dimensions of IoT networks: energy efficiency and the M2M/H2H necessities, including the Human Type
transfer rates in a network having a high and heavy Communication (HTC) output, the power level of
dynamic load [14]. A new objective functions (OF) is Machine Type Communications (MTC) tools for non-
presented in this study, which periodically investigates real-time services, and the highest potential for real-time
the last states of node chains in the pathway to the root services. Assessment of the suggested system
while avoiding the passage of additional message performance was done based on throughput, bandwidth
requests in the network. Additionally, this work provides utilization, and delay.
a new routing metric to respond to the network’s A context awareness framework known as Awareness
dynamicity requirement. Besides, a new approach is Cognition (AC) was proposed by Kim and Yoon [16] that
presented to parent selection. focused on the Ambient Intelligence (AmI) system’s
middleware layer. Using this framework, the prediction
problem was solved by integrating multiple contexts for
2. REVIEW OF LITERATURE detecting personalized knowledge. Additionally, utilizing
knowledge-based cognition approaches for discovering
Various research works have been conducted on the future changes, a fusion contextual learning model was
adaptive routings in objective functions and IoT, generated for behavioral knowledge detection. Besides,
including energy-consumption, load balancing, and time modeling was provided for collecting data from various
delay. However, there are a rare number of studies sensors in the AmI system in a single meaningful context
conducted on distributed learning in routing. Some [13, 17].
studies related to context information-based routing are Two algorithms with two constraints were presented
presented in this section. by Huang et al. [18] for challenging the multicast routing
Mohamed et al. [5] studied the proactive routing problem for multimedia communication in the IoT.
protocols with the highest energy efficiency for a Multiple constraints are integrated into a complete metric
homogeneous system. The Wireless Sensor Network using the entropy method. Thus, the proposed algorithms
(WSN) energy efficiency results from three fundamental significantly decreased the complication of multi-
necessities: network overhead within network constrained multicast routing problems, and some
arrangement and re-arrangement, selecting a route for popular algorithms can be applied for problem-solving.
data transmission, and fault tolerance or network Additionally, the theoretical analysis was suggested on
adaptability. For guaranteeing a longer network lifetime, the approximation and complexity of the suggested
827 R. S. Raza and M. S. Al-Ani / IJE TRANSACTIONS A: Basics Vol. 34, No. 4, (April 2021) 825-831

algorithms, and wide simulations were carried out to future protocols have good performance for the BS
assess the algorithm's performance. According to outside the network.
analytical and experimental findings, one of the proposed
algorithms is better than a representative multi-
constrained multicast routing algorithm in terms of speed 3. STATEMENT OF THE PROBLEM
and accuracy. The results provide an in-depth perception
of the multicast routing algorithm design for multimedia RPL, which is primarily devised for low-power lossy
communications in IoT. networks, possesses numerous notable characteristics,
Rani et al. [19] proposed an improved solution for including self-healing mechanism, loop-freeness, low-
organizing objects to implement an energy-efficient and battery use, and fast topology construction. Nevertheless,
scalable IoT. Firstly, the framework was presented to it fails to address issues of a network with a high traffic
deploy the IoT with scalability features providing higher rate as it was chiefly devised for low traffic networks.
extensibility. Afterward, considering the framework, an When there is a high network traffic, RPL cannot control
optimization outline can support the deployment of an it well, making various difficulties for the network, such
IoT with energy efficiency. This optimization outline is as energy depletion, load imbalance, and high packet loss
confined by the loads on wireless links and energy rate. It would be more troublesome when a depleted node
expenses. Compared to conventional WSN outlines in is the only intermediate node for a network section close
terms of network lifetime, time, and scalability, various to the root. RPL problems are classified under heavy and
numerical tests confirm superiority of the proposed highly dynamic loads as follows:
outline. This work addresses the challenges of the way of 1- With the computation of the rank through two OFs
using heterogeneity advantages. Nevertheless, including Objective Function Zero (OF0) and
improvement of end-to-end delay, throughput minimum rank with hysteresis objective function
parameters, packet delivery ratios, and data compression (MRHOF) in normal RPL, many studies have been
approaches are proposed to achieve more effective green conducted for changing the RPL objective functions
IoT. based on related research works. Nevertheless, as it
An effective self-organization protocol called ETSP is known, the former parents of a node in the
was presented by Qui et al. [20] for sensor networks of sequence are not considered by the ordinary RPL
IoTs, saving energy and allowing a prolonged lifetime for OFs and other proposed OFs. A node could seem
a network by restricting a tree-based network. The nodes’ proper for serving as a parent; however, the node
weight, such as hop, residual energy, distance within the parent or another parent in a parent sequence could
nodes, and a number of child nodes were used to have low remaining power or small buffer space,
determine whether it can be a sink node. Therefore, the which causes inappropriate selection of the parent in
depth of the tree is improved by ETSP. The topology of the network with high traffic.
the network is drastically changed over the data 2- The route is created concerning the rank. When
transmission procedure. Because the sink nodes consume attempting to connect the network or alter its parent
energy faster than other nodes, each sink node is by a node, the network is selected with a lower rank
dynamically re-chosen. The simulation results showed value. Nevertheless, the parents' rank is obtained in
that with ETSP, reliable tree-based networks could be the primary phases of connecting the network.
generated, the energy consumption can be reduced, and Various difficulties could appear in the pathway to
the sensor network’s lifetime can be extended. the route followed by the rank computation. Thus,
Shen et al. [21] presented an energy-efficient the rank value in a high traffic dynamic network
centroid-based routing protocol (EECRP) for controlling cannot completely represent the last and real mode
the WSN-assisted IoT energy. To this end, they solved the of the candidate parents.
forming clusters problem considering the distance to the
energy centroid. An enhancement algorithm was
presented in the present study regarding the number of 4. METHODOLOGY
cluster head nodes and the number of dead nodes. The
simulation results indicated that with the base station (BS) As a node connects to a Destination Oriented Directed
deployment in the network, it would be possible to Acyclic Graph (DODAG) and if a DODAG Information
transfer a great amount of data by the EECRP with very Object (DIO) message is received by it, the node can
low energy dissipation. Besides, the EECRP has a longer process it via the following ways:
network lifetime than the GEEC, LEACH-C, and 1) Removing the DIO package for some RPL criteria.
LEACH. This protocol will be enhanced for future study 2) Processing the message for keeping its location in
by identifying the multi-hop pathway from cluster head the network
(CH) nodes to BS. For transferring data packets, a multi- 3) Improving its location by gaining a lower rank in
hop pathway is used by the CH nodes. It is hoped that the DODAG.
 R. S. Raza and M. S. Al-Ani / IJE TRANSACTIONS A: Basics Vol. 34, No. 4, (April 2021) 825-831 828

When a node's rank decreases, the node must eliminate calculated for the route that the parent provides. In cases
all parents with a rank lower than its new rank from its of malignancy of the parent and when the level of
parents' list. Thus, the development of a loop in the confidence for parent node calculation is lower, the
network is prevented. Following this phase, the nodes respective node chooses a parent with a higher confidence
each has a default route to the root and they would be able level among the candidate parents. Using this method,
to send their data packages to the root. each node can efficiently prevent malignant nodes. It is
According to the simulation results, the efficiency of possible to gain a higher data collection ratio by routing
the CCR-based RPL routing approach in data aggregation the intermediate relay node data in the processing order.
and recognition in IOT network is high. In this section, Thus, it is effective in the reduction of the network traffic
using the content-based RPL approach in the direction of rate. Consequently, the delay in the data transfer can be
data aggregation rate, we compare the number of live significantly reduced. Moreover, we can annihilate data
nodes, energy consumption, the network balance, and transfers following data collection, resulting in a decrease
correct data transfers with those in the conventional RPL in energy consumption in wireless communications and
approach. saving battery energy consumption.
Similar to the properties given in Table 1, the Three major units are included in this method:
simulation's completion and the node routings in the 1) Data collection unit: It gains information about the
network can be observed. number of data packages arriving at the root by
For increasing the routing efficiency in IOT network, periodically sending the root-inquiry packages.
we presented a content-based system in this work. In the 2) Confidence level calculation unit: This unit uses the
presented system, each node utilizes the root-inquiry information presented by the data collection unit and
packages for recognizing the parent nodes to become computes the confidence level of the route presented
aware of the number of packages arriving in the root. by the parent using the following equation:
Utilizing this awareness, the level of confidence is 𝑝𝑘 𝑠𝑒𝑛𝑡𝑖𝑗 −𝑝𝑘 𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑𝑖𝑗
𝑇(𝑝) = 1 −
𝑝𝑘 𝑠𝑒𝑛𝑡𝑖𝑗

𝑖𝑓 𝑝𝑘𝑟𝑒𝑐𝑖𝑒𝑣𝑒𝑑 = 𝑝𝑘𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 (1)

TABLE 1. Properties of the simulation employed in our work.
Parameter Value 𝑇(𝑐) = 𝑇(𝑐) − 0.01 𝑒𝑙𝑠𝑒
Network area 800𝑚 × 800𝑚 where T(p): Confidence level of each
number of nodes 50 parents (neighbors), Pk sendij: Total number
of packets sent to the root by specific node,
Velocity of Node 0
𝑝𝑘𝑑𝑒𝑙𝑖𝑣𝑒𝑟𝑒𝑑 : Total number of packets arrived
Transfer range 100 m to the root through each parent (neighbors)
Load size 512 Bytes 3) Parent selection unit: It chooses a parent with a
higher level of confidence among the candidate
Packet transfer rate 25 KB
parents as the selected parent.
Data aggregation rate 0.0011 μJ
Data reception energy consumption 8.22 μJ
5. EVALUATION
Data transfer energy consumption 9.72 μJ
Simulation numbers 20 To assess the protocol proposed in our work, the CCR-
based RPL was simulated using the MATLAB simulator,
commonly used for IoT. Then, the CCR-RPL was
compared with conventional RPL. MRHOF was adjusted
as OF of ordinary RPL, and the protocol factors were
tuned based on the CCR-based RPL factors (Table 1).
There are 800 nodes in our setting with 20 BS, which
were established in 50m × 50m. The BS serves as the root,
and the transmission range is 25m.
In this setting, duty cycling was disabled for reaching
a high load in the network, and the first-in-first-out
(FIFO) line was utilized with a volume of 80 packets. The
simulation setting and the simulation parameters are
represented in Figure 2 and Table 1. Besides, various
traffic rates and nodes are considered for assessing the
Figure 2. Routing the network topology protocol proposed in our work under different conditions.
829 R. S. Raza and M. S. Al-Ani / IJE TRANSACTIONS A: Basics Vol. 34, No. 4, (April 2021) 825-831

Firstly, a comparison was conducted based on the

queue loss ratio. Figure 3 indicates two protocols with
varying traffic loads (rising). According to the results, the
numbers of the live nodes are increased by the network's
CCR-based RPL. Nevertheless, the other important
property shown in Figure 3 indicates the worst case of the
queue loss ratio in the nodes in varying traffic loads. We
interpret it as the CCR-based RPL capability to make a
more uniform DODAG concerning the network load.
Following application of the simulation, the following
results were obtained:
Figures 3-8 indicate the higher proficiency of the
CCR-RPL approach than the ordinary RPL in increasing
system efficiency and reducing energy consumption.
Thus, it reduces the number of dead nodes and energy
Figure 5. Comparing the rates of the remaining energy
consumption in the system and increases the system
efficiency and power.

Figure 3. Comparing the number of the live nodes Figure 6. Comparing the balances dominating the system

Figure 4. Comparing the rates of transferred packages Figure7. The delay rates in package transfers
 R. S. Raza and M. S. Al-Ani / IJE TRANSACTIONS A: Basics Vol. 34, No. 4, (April 2021) 825-831 830

5. Mohamed, R.E., Saleh, A.I., Abdelrazzak, M. and Samra, A.S.,

"Survey on wireless sensor network applications and energy
efficient routing protocols", Wireless Personal Communications,
Vol. 101, No. 2, (2018), 1019-1055.
https://doi.org/10.1007/s11277-018-5747-9
6. Mozaffari, M., Saad, W., Bennis, M. and Debbah, M., "Unmanned
aerial vehicle with underlaid device-to-device communications:
Performance and tradeoffs", IEEE Transactions on Wireless
Communications, Vol. 15, No. 6, (2016), 3949-3963.
https://doi.org/10.1109/TWC.2016.2531652
7. Hwang, K. and Chen, M., "Big-data analytics for cloud, iot and
cognitive computing, John Wiley & Sons, (2017).
8. Naguib, A., Saad, W. and Shokair, M., "Remaining energy aware
ml-csma/tdma hybrid mac protocol for lte-m2m wireless
network", in 2019 International Conference on Innovative Trends
in Computer Engineering (ITCE), IEEE. 322-327.
9. Mardani, M.R., Mohebi, S. and Ghanbari, M., "Energy and
Figure 8. Comparing the rates of power consumed
latency-aware scheduling under channel uncertainties in lte
networks for massive iot", Wireless Personal Communications,
Vol. 103, No. 3, (2018), 2137-2154.
6. CONCLUSION https://doi.org/10.1007/s11277-018-5901-4
10. Park, T. and Saad, W., "Distributed learning for low latency
RPL routing protocol problems with heavy and dynamic machine type communication in a massive internet of things",
load were addressed in the present work, focusing on the IEEE Internet of Things Journal, Vol. 6, No. 3, (2019), 5562-
network lifetime and energy. It was observed that an 5576. https://doi.org/10.1109/JIOT.2019.2903832
ordinary RPL could not provide effective control over the 11. Dawy, Z., Saad, W., Ghosh, A., Andrews, J.G. and Yaacoub, E.,
"Toward massive machine type cellular communications", IEEE
heavy and dynamic loads. To provide a solution for this Wireless Communications, Vol. 24, No. 1, (2016), 120-128.
challenge, a load balancing and context-aware protocol https://doi.org/10.1109/MWC.2016.1500284WC
was proposed considering the parent-chain rank before 12. Azari, A. and Cavdar, C., "Self-organized low-power iot
selecting the chain’s ultimate parent as the node’s chosen networks: A distributed learning approach", in 2018 IEEE Global
to parent. Thus, the load was tried to be balanced in the Communications Conference (GLOBECOM), IEEE., 1-7.
13. Feizi, A., "Convolutional gating network for object tracking",
network. The present study took into account the residual
International Journal of Engineering, Vol. 32, No. 7, (2019),
queue and energy level of the candidate parents. 931-939. https://doi.org/10.5829/ije.2019.32.07a.05
Additionally, a proper parent's rush was prevented, 14. Hossein Motlagh, N., Mohammadrezaei, M., Hunt, J. and Zakeri,
making instability problems and a high control message B., "Internet of things (iot) and the energy sector", Energies, Vol.
rate for the network. Our proposed protocol was assessed 13, No. 2, (2020), 494. https://doi.org/10.3390/en13020494
in MATLAB in varying conditions. It was proved that the 15. Aman, M.S., Yelamarthi, K. and Abdelgawad, A., "A comparative
analysis of simulation and experimental results on rpl
performance of CCR-RPL is significantly better than performance", in 2017 IEEE 8th Annual Ubiquitous Computing,
RPL, while it does not impress a high overload for the Electronics and Mobile Communication Conference (UEMCON),
network. IEEE., 483-487.
16. Kim, S. and Yoon, Y.-I., "Ambient intelligence middleware
architecture based on awareness-cognition framework", Journal
7. REFERENCES Of Ambient Intelligence And Humanized Computing, Vol. 9,
No. 4, (2018), 1131-1139. https://doi.org/10.1007/s12652-017-
1. Xie, J., Chen, W., Dai, H., Liu, S. and Ai, W., "A distributed 0647-5
cooperative learning algorithm based on zero-gradient-sum 17. Firouzian, I., Zahedi, M. and Hassanpour, H., "Cycle time
strategy using radial basis function network", Neurocomputing, optimization of processes using an entropy-based learning for task
Vol. 323, (2019), 244-255. allocation", International Journal of Engineering, Transactions
https://doi.org/10.1016/j.neucom.2018.10.001 B: Applications, Vol. 32, No. 8, (2019), 1090-1100.
2. Debroy, S., Samanta, P., Bashir, A. and Chatterjee, M., "Speed- https://doi.org/10.5829/ije.2019.32.08b.05
iot: Spectrum aware energy efficient routing for device-to-device 18. Huang, J., Duan, Q., Zhao, Y., Zheng, Z. and Wang, W.,
iot communication", Future Generation Computer Systems, Vol. "Multicast routing for multimedia communications in the internet
93, (2019), 833-848. https://doi.org/10.1016/j.future.2018.01.002 of things", IEEE Internet of Things Journal, Vol. 4, No. 1,
3. Thivakaran, T. and Sakthivel, T., "Guard: An intrusion detection (2016), 215-224. https://doi.org/10.1109/JIOT.2016.2642643
framework for routing protocols in multi-hop wireless networks", 19. Rani, S., Talwar, R., Malhotra, J., Ahmed, S.H., Sarkar, M. and
Wireless Networks, Vol. 25, No. 2, (2019), 819-836. Song, H., "A novel scheme for an energy efficient internet of
https://doi.org/10.1007/s11276-017-1594-y things based on wireless sensor networks", Sensors, Vol. 15, No.
4. Nielsen, J.J., Madueño, G.C., Pratas, N.K., Sørensen, R.B., 11, (2015), 28603-28626. https://doi.org/10.3390/s151128603
Stefanovic, C. and Popovski, P., "What can wireless cellular 20. Qiu, T., Liu, X., Feng, L., Zhou, Y. and Zheng, K., "An efficient
technologies do about the upcoming smart metering traffic?", tree-based self-organizing protocol for internet of things", Ieee
IEEE Communications Magazine, Vol. 53, No. 9, (2015), 41- Access, Vol. 4, (2016), 3535-3546.
47. https://doi.org/10.1109/MCOM.2015.7263371 https://doi.org/10.1109/ACCESS.2016.2578298
‫‪831‬‬ ‫‪R. S. Raza and M. S. Al-Ani / IJE TRANSACTIONS A: Basics Vol. 34, No. 4, (April 2021) 825-831‬‬

‫‪21.‬‬ ‫‪Shen, J., Wang, A., Wang, C., Hung, P.C. and Lai, C.-F., "An‬‬ ‫‪in wsn-assisted iot", Ieee Access, Vol. 5, (2017), 18469-18479.‬‬
‫‪efficient centroid-based routing protocol for energy management‬‬ ‫‪https://doi.org/10.1109/ACCESS.2017.2749606‬‬

‫‪Persian Abstract‬‬
‫چکیده‬
‫در مارس ‪ 2012‬پروتکل مسیریابی ‪ IPv6‬برای شبکه های کم مصرف (‪ )RPL‬توسط ‪ Task Force‬مهندسی اینترنت به عنوان پروتکل استاندارد مسیریابی برای اینترنت اشیا‬
‫(‪ )IoT‬تأیید شد‪ .‬از آن زمان ‪ ،‬این برنامه کاربردهای مختلفی در اینترنت اشیا دارد‪ .‬علی رغم تأمین نیازهای شبکه اینترنت اشیا توسط ‪ ، RPL‬برخی از موضوعات بی پاسخ‬
‫برای کاربردهای اینترنت اشیا مطرح نشده است‪ .‬با این حال ‪ ،‬جمع آوری مقدار زیادی از داده ها از این شبکه ها با فیلم و تصاویر به طور معمول منجر به ازدحام ترافیک در‬
‫قسمت مرکزی شبکه می شود‪ .‬برای این موضوع‪ ،‬در این مقاله روش مسیریابی محتوا محور ‪ RPL‬مبتنی بر ‪ CCR‬ارایه شده است‪ .‬در این روش محتوا مسیرهای مسیریابی را‬
‫مشخص می کند‪ .‬با مسیریابی داده های مربوطه به گره های رله میانی برای فرآیند ‪ ،‬می توان به یک نسبت تجمع داده بزرگتر دست یافت‪ .‬بنابراین ‪ ،‬ترافیک موثر در شبکه ایجاد‬
‫می شود‪ .‬پس از آن ‪ ،‬تاخیر به طور قابل توجهی کاهش می یابد‪ .‬عالوه بر این ‪ ،‬استفاده از انرژی در ارتباطات بی سیم کاهش می یابد‪ .‬ادغام بیشتر بین پروتکل ‪ IETF RPL‬و‬
‫‪ CCR‬با استفاده از سیستم عامل ‪ MATLAB‬انجام شد‪ .‬سرانجام ‪ ،‬با توجه به نتایج شبیه سازی شده و پیاده سازی شده ‪ ،‬رفتار ‪ RPL‬مبتنی بر ‪ CCR‬بر اساس نرخ بالی‬
‫انتقال بسته بهتر است و تعداد گره های مرده کاهش می یابد‪ .‬بازده انرژی بال و تسریع در انتقال حاصل می شود‪.‬‬