J.

Electrical Systems 20-7s (2024):482-494

Animesh Giri1* Advancing Healthcare Networks:

Dr Annapurna2 Optimizing Multi-Objective RPL for
Diverse Traffic in Low-Power, Lossy
Environments

Abstract: RPL: Powering modern healthcare IoT Networks. The Routing Protocol for Low Power and Lossy Networks(RPL) is a protocol specifically
designed for routing in networks characterized by low power and lossy connections built on IPv6. It is specifically designed for the growing use of
Instantaneous IoT (Internet of Things) applications. These networks cater to the specialized requirements of Low Power (resource-constrained) and
Lossy Networks(unstable networks), also known as LLNs, where routing data smoothly and prioritizing traffic types are the significant challenges.
Studies have shown that standard RPL, which makes use of a set of routing rules, struggles to provide the level of performance that many IoT applications
in modern-day healthcare demand. This limits RPL’s potential in scenarios where we have a mix of data traffic, which is very common in healthcare
settings. To handle the diverse data traffic found in healthcare settings, our method involves distributing several instances of RPL settings throughout
the network. The main objective of this strategy is to improve the effectiveness of important and critical healthcare applications.
We have developed and thoroughly tested our system using Contiki, which is an open-source IoT operating system. We have three instances in our
MultiInstance RPL setup, each of which is intended to handle different kinds of network traffic and provide varying Quality of Service (QoS) measures.
Different sets of Objective Function (OF) rules are used by each instance to decide which route is optimal for the data it processes. We have made use
of OF0 (Objective Function 0) and MRHOF (Minimum Rank with Hysteresis Objective Function). We have rigorously tested our solution. Our focus
has been on metrics like delay, average energy consumption and Packet Delivery Ratio (PDR). Compared to standard RPL, our approach improved
packet delivery rates and significantly reduced delays for high priority data packets. This research shows how to better address the complex data traffic
needs of real-time healthcare IoT networks.

Keywords: Internet of Things, RPL, Multiple Instances, Heterogeneous Traffic, LLNs, RDC and MAC Protocols

1 Preface
Over the past few years, the rapid advancements in IoT have sparked significant interest across various research domains,
particularly IoT communication. Notably, protocols such as Bluetooth, ZigBee, and Wi-Fi, which boast low energy
capabilities, have emerged as critical players. However, integrating LLNs utilizing 6LoWPAN (IPv6 over Low-Power
Wireless Personal Area Networks) presents unique challenges, including limitations on energy consumption, storage
capabilities, and computational resources for embedded sensors. These connections often exhibit unreliability,
characterized by significant packet loss and slow transmission rates [1][2].
The progress and expansion in the IoT domain face numerous obstacles, with routing standing out as one of the most
difficult challenges. IoT encounters routing complexities influenced by energy consumption and interference when
operating within a wireless environment.
In the healthcare sector, IoT technologies are instrumental in continuously monitoring and assessing patient data. Given
this data’s critical and time-sensitive nature, the need for dependable communication systems becomes paramount. These
systems are crucial in ensuring continuous monitoring and prompt patient care provision. Therefore, healthcare networks
require highly dependable routing protocols, given their constrained memory and energy resources [3][4].
Routing in LLNs often relies on RPL, a protocol specifically designed for such networks and standardized by the IETF in
RFC 6550. The challenges associated with LLNs, including memory, energy, computational constraints, and unreliable
connectivity paths, are effectively addressed by RPL. This protocol is capable of handling both Point-to-Multipoint
(P2MP) and Multipoint-to-Point (MP2P) communication traffic, and it can scale to networks comprising thousands of
nodes [5][6]. Unlike many Wireless Sensor Network (WSN) protocols like LEACH, CTP, and PEGASIS, which struggle
with accommodating multiple applications, RPL excels in this aspect. Therefore, RPL emerges as the optimal choice for
the project, instilling confidence in its suitability and effectiveness.

2 About the Routing Protocol - RPL

This routing protocol is an active distance vector-based routing protocol for networks that use 6LoWPAN technology. It
breaks down the network topology into several easily comprehensible tree structures referred to as DODAG, which stands
for Destination Oriented Directed Acyclic Graph. Every DODAG represents an instance of RPL; All the DODAGs are
equipped with a specific optimization criterion aimed at refining the network’s topology based on metrics such as shortest
routes, network link reliability, and number of hops [7][8]. The focal point of each DODAG is the root or sink node.
In an RPL network, the root node, often called the border router (LBR), is the bridge to external networks. It serves as the
main link to external networks. Positioned at the edge of the RPL network, the LBR facilitates communication between
two distinct networks. The LLN typically connects to the external domain, comprising the IPv6 network or the Internet,
through the DODAG root. A multi-hop mechanism engages other nodes as parent nodes when direct communication with
the LBR is unattainable.

1*
Research Scholar, Department of Computer Science and Engineering, PESIT - Bangalore South Campus, Bengaluru, 560100, Karnataka, India
Visvesvaraya Technological University, Belagavi, 590018, Karnataka, India. Email: [email protected]
2
Professor, Department of Computer Science and Engineering, PESIT Bangalore South Campus, Bengaluru, 560100, Karnataka, India. Email:
[email protected]

482
 J. Electrical Systems 20-7s (2024):482-494

The effectiveness of information transmission in the IoT network depends on the crucial operation of parent selection,
which RPL manages.
Control messages are crucial packets essential in managing network stability by establishing and updating routing
information. The trickle timer algorithm, coupled with these messages, ensures stability in the network.
RPL control messages encompass several categories, including:
DODAG Information Solicitation (DIS): New nodes do this, which initiates connection and participation within the
network.
DODAG Destination Advertisement Object (DAO): These packets are optional and assist in establishing downward routes
[8].
DODAG Information Object (DIO): This facilitates establishing, maintaining, and discovering DODAGs.
DAO-Acknowledgment control messages.
The RPL network orchestrates node arrangement to enable routing paths from all nodes down to the root.
Routing within RPL manifests in two primary modes:
• Upward Routing: This mode involves the transmission of traffic through nodes, where packets are forwarded directly to
their preferred parent nodes, eventually reaching the network’s root.
• Downward Routing: This process can be carried out in two ways: by maintaining a routing table at each node (also
known as a solid mode) or by handling it centrally at the root node (also known as a non-storing mode). Downward traffic
is predominantly employed for overseeing network control, retrieving data, and enabling end-to-end communication
objectives [9].
Routing strategies within RPL are tailored using objective functions, which determine how nodes select preferred parents
and calculate their ranks. Standard objective functions such as MRHOF and OF0 are utilized for routing in RPL. MRHOF
typically relies on the expected transmission count (ETX) measure, and this concept deals with the reliability of packet
transmissions required to reach the destination without errors. On the other hand, OF0 adjusts rank increments based on
Hop Count [10][11].
Default OFs in RPL are tailored to consider a single metric. This may lead to compromised routing performance,
particularly in IoT setups with diverse quality of service requirements within a unified network [58]. Furthermore, these
OFs, which were initially designed for networks with minimal data flow, encounter significant challenges in larger
network configurations. Despite RPL’s flexibility in accommodating multiple metrics for parent selection, predetermined
criteria for metric combinations remain unspecified [12].

2.1 Establishment of DODAG in RPL: The Foundation of Network Structure:

The creation of DODAGs in RPL initiates when the LBR broadcasts DIO control packets to nearby nodes. These packets
contain vital details, including node ranks and distances, including those of the root node, which are essential for
establishing the DODAG structure [12].
After receiving a DIO packet, neighboring nodes evaluate their proximity to the root. The next hop is set to be the root,
and the packet is routed to nearby nodes. The individual receiving node uses the packet’s information and identifies its
next hop towards the root. This sequence continues until all nodes are incorporated into the DODAG [13].
RPL’s operating approach focuses on creating channels directly connecting to the root. A node has to transmit a Destination
Advertisement Object (DAO) to its immediate superior node to allow downstream traffic while retaining access to parent
nodes. The Destination Advertisement Object (DAO) details node prefixes in its subDODAG. Upon reaching the root,
these prefixes undergo aggregation. Additionally, RPL nodes can dispatch DODAG Information Solicitation (DIS)
communications to neighboring nodes, soliciting DIO communications. Figure 1 illustrates the process of DODAG
creation in RPL.

2.2 Instances in RPL

One or more DODAGs, which function as a logical arrangement of these structures, are included in an RPL instance.
Every DODAG in an instance has its own RPLInstance ID. To verify DODAG coherence, unique DODAG IDs are
assigned to each DODAG in a model. Within RPL, these identifiers are shared via the IPv6 protocol
[14].
A single instance of a DAG can have more than one RPL node participating in it. Each RPL entity is associated with a
unique Objective Function (OF), enabling the creation and classification of different DODAGs within the same topology.

2.3 RPL Multi-Instance

In real-world scenarios, different applications might generate various types of network traffic, each having particular
Quality of Service (QoS) requirements. To address this,

483
 J. Electrical Systems 20-7s (2024):482-494

Fig. 1 Formation of DODAG in RPL

RPL enables the generation of numerous instances within a single network, each featuring a unique objective function.
This flexibility enables a single physical network to support diverse routing metrics and topology. Within a multi-instance
RPL network, every node oversees multiple routing tables tailored to different objective functions. To route data
effectively in a multi-instance RPL network, a leaf node must do two things when it receives a packet. It should decide
which routing instance best fits that type of data and use the relevant routing table within that instance to figure out where
to send the packet next [12,15]. Figure 2 demonstrates this concept within the architecture of a multi-hop LLN, where the
LBR connects the 6LoWPAN network to the broader internet. It should be noted that RPL does not support multiple
instances by default. Modifications to the default protocol and routing methods are made to accommodate various use
cases.

3 Related Works
In a 2019 publication ”Enhancing IoT-based Systems through Multiple RPL Instances,” Al-Abdi, et al. [14] delve into the
advantages of supporting numerous RPL instances. Utilizing simulations via the Cooja Simulator, they establish a baseline
of enhanced performance, focusing on Packet Delivery Ratio (PDR) across varying Received Signal Strength Indicator
(RX) values and traffic types using multiple RPL instances. Their study underscores the need for more than a single RPL
instance to handle critical data traffic effectively.
In 2017, Nassar et al. [15], in their paper ”Towards Efficient Quality of Service with Multi-instance RPL for Smart Grids,”
introduced OFQS, an objective function designed to adapt to different traffic classes and deliver Quality of Service (QoS)
distinctions customized to the diverse needs of smart grid applications. OFQS employs

Fig. 2 Multi-hop Low-Power and Lossy Network (LLN) architecture)

484
 J. Electrical Systems 20-7s (2024):482-494

a multi-criteria metric that considers factors such as available energy in battery nodes, delay, and network quality, resulting
in an extended network lifetime, higher packet delivery ratio, and reduced latency.
Mardini et al. [17], in 2021, established multiple RPL instances grouped by traffic class. Their approach assigned critical
data to the first instance and low-priority and periodic data to the second, resulting in superior average PDR across all RX
values compared to single-instance RPL. Notably, critical data traffic experienced marked improvement and significantly
reduced average latency.
In 2020, Bhandari et al. [18] proposed various OFs to ensure network-level quality of service differentiation. Their
approach involved dividing the network into several RPL instances, each accommodating different traffic types with
distinct OFs. Additionally, they introduced a novel framework, parent node selection, employing multi-attribute decision-
making criteria, significantly reducing packet loss, delays, and reliability issues while ensuring minimum overhead and
network stability, unlike default RPL.
Bouzebiba and Lehsani [19], in their 2020 paper ”FreeBW-RPL: A Novel Objective Function for Multimedia IoT,”
introduced FreeBW. This function selects routing paths based on the required bandwidth for quality of service routing.
FreeBW-RPL improves multimedia applications by dynamically selecting the most suitable forwarding candidate based
on available free bandwidth information, hence decreasing congestion and improving performance measurements like
end-to-end delay, Packet Delivery Ratio, throughput and energy consumption compared to conventional RPL.
Talking about 2019, Brandon Foubert [21] proposed a technique to enhance redundancy for border routers using a virtual
DODAG. This approach enables multiple border routers to participate in a single DODAG, thereby synchronizing their
parameters and allowing congested border routers to offload and distribute traffic to neighboring ones. This method
improves the overall end-to-end and Media Access Control (MAC) Delivery Ratio by mitigating link-layer errors,
surpassing the performance of RPL alone.

4 Proposed Approach
This research proposes a novel network architecture incorporating three distinct channels within the overall structure.
They are known as Instance-1, Instance-2, and Instance-3 channels. Each provides a particular kind of information.
Consider a data transmission system that is a multi-lane highway. Instance-1 serves as the express lane, giving priority to
urgent communications. This lane layout aims to reduce hold-ups and guarantee the quickest delivery of vital information.
Conversely, instance 2 is the standard channel for critical information that is only sometimes time-sensitive. This lane
ensures timely delivery while preserving network resources by finding a compromise between efficiency and speed. As
the high-occupancy vehicle lane, instance-3 prioritizes efficiency for infrequent changes that use little bandwidth and least
critical and non-time-sensitive data.
To evaluate the efficiency of this three-channel approach, we carried out a comparative analysis. Using Instance-1,
Instance-2 and Instance 3, our system was compared against a standard RPL network that employs two separate channels.
In the standard network, one channel prioritizes speed for all data traffic, regardless of its criticality. This prioritization is
achieved through a metric known as Expected Transmission Count. The other channel, like Instance-3, focuses on efficient
data transmission for routine updates using Hop Count as a metric. This comparison allows us to evaluate the benefits of
our three-channel design in optimizing network performance for diverse data types.

4.1 Experimental Setup

Contiki OS lacks native support for multiple RPL instances; however, we have created a solution that allows this feature.
We concentrated on making data flow smoothly upwards in the network. To give critical data that extra push, we combined
NullRDCMAC with ContikiMAC. We also assigned specific roles (called objective functions) to different instances:
MRHOF would take care of the super important and urgent data packet, while OF0 would handle regular traffic and
routine updates. Figures 3 and 4 illustrate our experimental setups to compare single-entity and multi-entity scenarios. For
the first scenario, we established a baseline using Contiki OS’s default protocol settings, including ContikiMAC and
CSMA, representing a standard network configuration. We introduced a more sophisticated approach in the second
scenario by employing multiple instances to prioritize critical traffic. To achieve this, we customized the NullRDC
protocol, tailoring it to the specific needs of our experiment.

4.2 Simulation methods and Experiments

To test how well our proposed approach works, we did several simulations using the Cooja simulator within a Contiki 3.0
OS environment. Each simulation involved a

485
 J. Electrical Systems 20-7s (2024):482-494

Fig. 3 Nodes dispersed across a 300 * 300 meter expanse

network topology consisting of 41 Zolertia motes strategically positioned across a 300m by 300m area (refer to Figure 3).
Three designated sinks acted as data collection points.
We segregated the network into the following zones:
• ”ICU” (16 motes): Designated for critical data generation (Instance 1).
• ”Special Ward” (12 motes): Designated for important, but less time-sensitive data (Instance 2).
• ”Lab” (10 motes): Designated for routine updates (Instance 3).
Motes generated UDP data packets at frequencies determined by their assigned zone (Table 1). We configured radio
transmission and interference ranges to be 50m and 100m, respectively. Reception success ratios varied from 20% to
100%, while the transmission success ratio remained fixed at 100%.

Table 1 Recommended transmission intervals and proposed packet payload sizes.

Motes Count Objective Function Size of Payload Transmission Interval Type of Traffic RPL Instances
16 MHROF (ETX) 16 Bytes 15 Seconds Critical Instance 1
12 MHROF (ETX) 32 Bytes 30 Seconds Non-Critical Instance 2
10 OF0 (HC) 48 Bytes 180 Seconds Periodic Instance 3

Fig. 4 Visualization depicting the arrangement of sensors in a hospital setting.

5 Experimentation Results and Interpretation

We tested how well our modified RPL network works under various conditions. We will look at a basic network setup
(single instance) and our special version with multiple channels (multiple instances) to see which handles things better.
To measure and compare how well the network is doing, we will track the following metrics:
486
 J. Electrical Systems 20-7s (2024):482-494

Packet Delivery Ratio (PDR): The proportion of messages successfully reaching their intended destination in comparison
to the number of messages transmitted. Think of it like our mail delivery success rate.
Average End-to-End Latency: The time it takes for messages to traverse the entire network
Handling Different Data Types: We will specifically test how our multiple-instance RPL handles a mix of data (urgent vs.
routine). It varies according to the size of the packet and packet send interval. This is known as heterogeneous data.
Formulas:
We’ll use this equation to calculate PDR. It indicates the overall health of the network [9]: [Equation 1]

Table 2 Simulation environment for networks and their corresponding parameters

Simulation Parameter Values
Type of Operating System Contiki 3.0 / Ubuntu 21.04
MAC Layer CSMA with Collision Avoidance
Radio-Duty Cycle NullMAC and ContikiMAC
Physical Layer IEEE 802.15.4 (Channel 26), CC2420 2.4
GHz
Mote Device Model Z1 Zolertia
Network Layer IPv6, ContikiRPL
Adaption Layer 6LoWPAN
Number of Sink Node 3
Number of Sender Nodes 38
Reception Ratio Varying from 30%, 50%, 70%, 85%, and
100%
Transmission Ratio Fixed at 100 %
Interference range for every 100 m
Instance
Transmission Range for 50 m
every Instance
Simulation duration 2700 Seconds
Deployment Coverage Area 300 X 300 m
Wireless Channel Loss of Distance in the Unit Disk Graph
Medium (UDGM)
Objective Function (OF) MHROF – ETX, OF0 – Hop Count

Table 3 RPL instances set up with diverse RDC and MAC protocol configurations.
RDC Protocol MAC Protocol Type of Data Traffic Objective Function RPL Instances
ContikiMAC/NullMAC CSMA Critical MHROF (ETX) Instance 1
ContikiMAC CSMA Non-Critical MHROF (ETX) Instance 2
ContikiMAC CSMA Periodic OF0 (HC) Instance 3

Table 4 Proposed method for facilitating multiple RPL instances.

Reliability Maximum allowed Delay Type of Data Traffic Instance Classification
90% to 100% 5 seconds Critical Instance 1
90% to 100% 5 seconds Non-Critical Instance 2
No strict reliability constraints Periodic 1 to 5 minutes Instance 3

100 (Eq.1)

End-to-End Latency: This is the overall duration for a communication to move from its origin to its ultimate destination.
We analyze the average latency to evaluate The network’s overall performance with respect to speed and time taken to
reach the destination.
Energy Consumption: This is how much energy a node uses when talking to other parts of the network, sending either
regular messages or control messages. Since these nodes have limited power sources, keeping track of energy use is
important[10].
5.1 PDR
Our results show that the MRHOF function’s network instance delivers more messages successfully (higher Packet
Delivery Ratio, or PDR) than the one using OF0. We tested this at a low packet reception ratio (30%) to make things
challenging and facilitate real-world-like circumstances. To really see how well MRHOF and OF0 handle harsh network
487
 J. Electrical Systems 20-7s (2024):482-494

conditions, we tested them with different packet reception ratios. MRHOF held firm, delivering messages consistently –
it is the reliable choice if the network is expected to have weak and unreliable connection. OF0 was less predictable, with
its PDR sometimes dropping as the packet reception ratio changed. Figure 5 shows our findings. As we can observe from
the graph, MRHOF consistently outperforms OF0 in terms of PDR across different conditions.
The comparison between MRHOF and OF0 regarding PDR is effectively depicted in Figure 5, highlighting MRHOF’s
advantages across diverse conditions.

Fig. 5 Packet Delivery Ratio (PDR) for a single instance with varying Objective Functions such as
OF0 and MRHOF

5.1.1 Scenario 1: Routing Protocol for Low-Power and Lossy Networks instances established with typical default
Medium Access Control (MAC) and Radio Duty Cycle (RDC) Protocol.
We commence by employing RPL instances pre-configured with typical RDC and MAC protocols. MAC protocols, such
as TDMA and CSMA, act as the network traffic controllers, governing the communication between devices. Precisely, in
wireless networks, the CSMA/CA (Collision Avoidance) protocol consists of a set of regulations aimed at averting data
collisions when multiple devices endeavor to access the network concurrently.
On the other hand, RDC is like a power-saving feature. It tells the devices when to sleep and when to be active to receive
messages. The traditional RDC usually follows a simple on-and-off pattern to keep the devices from draining their
batteries.
When using the default setup, the parameters controlling these protocols come with preset values. These parameters are
the fine-tuning knobs that govern how the protocols work and how the devices in our network interact. The defaults are
chosen based on the type of wireless network and how much data you expect to be flowing through it.
In short, using the standard MAC and RDC protocols with RPL is a good starting point for LLNs. However, for specific
situations, we need custom MAC and RDC protocols to fine-tune the network and get the best performance for your
application. Healthcare is one such case.
Our experiments showed that with a single instance of RPL, once the ReceptionRatio (RX) dropped below 70%, PDR
suffered, especially for critical and non-critical traffic – it fell below 50% due to the weak connection. Our proposed
Multiple Instance RPL maintained a PDR of over 90% for both critical and non-critical traffic, even in the presence of a
weaker signal.

Fig. 6Scenario 1 Packet Delivery Ratio for Single Instance and Proposed Multiple Instance

488
 J. Electrical Systems 20-7s (2024):482-494

Using the Hop Count metric for periodic traffic with the Multiple Instance approach also maintained high PDRs even with
low RX ratios. Figure 6 shows the average PDR across various data types (critical, non-critical, periodic) for both single
and Multiple Instance RPL. Our setup shows that PDR in the Multiple Instance setup is higher for critical and periodic
traffic.
Overall, our Multiple-instance method consistently provided a superior PDR for both critical and non-critical data
transmission compared to the single-instance method, particularly under weak signal conditions. Also, the Hop Count
metric within our Multiple Instance setup delivered better PDR for periodic traffic, even with a low RX ratio. This
demonstrates that using Multiple Instances and choosing the proper objective functions can significantly boost LLN
reliability and efficiency.

5.1.2 Scenario 2: RPL Instances configured with NullRDC and MAC Protocol
NullRDC and MAC protocols offer alternatives to the usual MAC and RDC protocols for RPL instances. The NULLRDC
driver is a simple RDC driver in Contiki that does not implement any duty cycling or radio power management. It
effectively disables any duty cycling, making the radio continuously available for communication. The MAC protocol, in
turn, handles the organization of the wireless channel, ensuring devices do not collide when trying to communicate. Pairing
NullRDC and MAC together creates a more robust and reliable interface for the transmission of data packets.

Fig. 7 Packet Delivery Ratio (PDR) for a Singular Instance and Proposed Multiple Instances utilizing NullMAC
Setting up RPL with NullRDC and MAC allows us to measure the network’s performance. We can look at metrics
like:
Packet Delivery Ratio (PDR): How many packets reach their destination compared to how many were sent.
Average End-to-End Latency: The overall duration for a packet to traverse the network.
Throughput: The overall amount of data the network can handle.
In scenario 2, Multiple Instances for Critical Traffic are set up with the NullRDC protocol, keeping the radio active for
transmitting Critical Traffic. This may result in increased energy consumption. Figure 7 compares the PDR of a Single
Instance and Proposed Multiple Instances with NullMAC. Notably, the Packet Delivery Ratio (PDR) indicates the
proportion of received packets compared to those sent.
Our simulation findings indicate that in a single instance configuration when the reception ratio (RX) falls below 70%,
there is a notable decrease in the overall Packet Delivery Ratio (PDR) due to the diminished connection quality. This
decline impacts both Critical and Non-critical traffic, resulting in PDR values dropping below 50%. However,
implementing our proposed Multiple Instance RPL utilizing NullRDC effectively maintains a higher Packet Delivery
Ratio for both critical and non-critical traffic, even with lower RX values. This underscores the advantage of the Multiple
Instance approach in terms of PDR compared to the basic Single Instance model. While maintaining radio activity for
Critical Traffic via NullRDC enhances reliability, it naturally increases energy consumption.

489
 J. Electrical Systems 20-7s (2024):482-494

Fig. 8 Packet Delivery Ratio (PDR) for Proposed Multi-Instance – Scenarios 1 and 2

Our proposed Multiple Instance approach has broader applications beyond the scenarios tested here. Figure 8 showcases
the PDR comparison between Scenarios 1 and 2, demonstrating that Multiple Instances with NullRDC consistently
outperform a Single Instance setup. It delivers higher PDR across all traffic types, regardless of signal strength (RX).
These results emphasize the effectiveness of our Multiple-instance approach with NullRDC for routing Critical Traffic. It
guarantees higher PDR and minimizes packet loss and delays. Although there is a trade-off in energy consumption, the
gains in reliability are significant. Overall, the findings highlight the paramount importance of MAC and RDC protocol
choices in maximizing routing efficiency, mainly when dealing with critical data.

5.2 Average Latency

We evaluated our proposed multiple instances method for handling different traffic types (critical, non-critical, and
periodic) across various signal strengths (30% to 100% reception). It proved to be much better than the traditional single-
instance RPL approach. We also found that MRHOF has significantly lower latency than OF0 when dealing with standard
single-instance network traffic.
The multiple instances approach showcased an average latency of nearly 0.1% of that observed in single-instance RPL.
This signifies a considerable reduction in the latency as compared to the current method. Table 5 outlines the average
latency comparison between the proposed and existing systems.
This method improves the system’s overall performance by lowering end-to-end delay significantly. Because of how well
packets are routed over the network, latency has decreased. The MRHOF objective function takes into account both
network reliability and hop count. This makes such results feasible.
Large-scale IoT networks can benefit from the many instances of scalability of methods. This allows it to handle a variety
of traffic types and scenarios. It enables routing techniques to be tailored for particular traffic kinds, thus maximizing the
use of network resources. As a result, the multiple instances method shows promise for improving the performance of IoT
networks by outperforming single-instance RPL in terms of average latency and packet delivery ratio for all traffic
categories. Successful Reception Single Instance Proposed Multiple Type of Data Traffic

Ratio OF0 MRHOF Instances

100% 33.6 30.4 0.0273 Overall Traffic
85% 37.1 32.7 0.0281 Overall Traffic
70% 39.8 34.5 0.0314 Overall Traffic
50% 41.5 37.8 0.0348 Overall Traffic
30% 43.1 40.1 0.0375 Overall Traffic

5.3 Average Energy Consumption

Figures 9, 10, and 11 depict how much more energy efficient the single instance approach is when using MRHOF than
when using OF0. Additionally, in both Scenario 1 and 2, the multiple-instance setup consistently uses less energy on
average than the single-instance approach – this holds regardless of signal reception strength.
The enhanced energy efficiency of the MRHOF-based single-instance approach stems from its consideration of
communication link quality, a factor overlooked by OF0. Consequently, MRHOF can establish more robust
communication paths with

490
 J. Electrical Systems 20-7s (2024):482-494

Average power consumption for a single instance with varying objective functions such as OF0 and MRHOF.

Fig. 9Average energy usage for a single instance and proposed multiple instances – Scenario 1.

fewer retransmissions, particularly at lower reception ratios, substantially reducing energy consumption. This observation
aligns with prior research highlighting MRHOF’s efficacy in enhancing energy efficiency in low-power wireless networks.
The multiple-instance approach boosts network energy efficiency by providing several routes for data to travel
simultaneously. This spreads out the workload, so every path gets manageable. This makes the network more reliable and
less likely to fail—which is extremely important for applications where downtime is not an option.

Fig. 10 Average energy consumption for a singular instance and the proposal of multiple instances – Scenario 2.

491
 J. Electrical Systems 20-7s (2024):482-494

To sum up, Figures 9, 10, and 11 demonstrate that the proposed multiple-instance approach using MRHOF is significantly
more energy-efficient than the traditional single-instance approach using OF0. This advantage is especially noticeable
when signal quality is weaker.

6 Conclusion
Our research offers a new solution to the complex problem of managing diverse types of data in healthcare environments
that use LLNs. We propose refining the existing RPL protocol to better handle different types of traffic.
We conducted a thorough analysis comparing two routing methods – MRHOF and OF0. MRHOF consistently
outperformed, underscoring the importance of considering link quality when making routing decisions. Additionally, we
tested NullRDC MAC against the traditional ContikiMAC for handling critical ICU data. NullRDC MAC proved faster
and more reliable but at the cost of increased energy consumption.
Our methodical methodology blends multiple strategies to address the difficulties presented by LLN healthcare scenarios.
Together with the Null-RDC MAC, we strategically use both MRHOF and OF0 to handle various traffic types, resulting
in an effective healthcare system that caters to the unique needs of the networks.
This study introduces a novel strategy for efficiently managing the complexities of mixed traffic in healthcare LLNs. The
results show significant performance improvements, particularly for critical data. This signifies a promising step forward
and lays the groundwork for future research in the field.

References
[1] Kushalnagar, Nandakishore, Gabriel Montenegro, and Christian Schumacher. ”IPv6 over low-power wireless personal
area networks (6LoWPANs): overview, assumptions, problem statement, and goals.” (2007).
[2] Clausen, Thomas, Ulrich Herberg, and Matthias Philipp. ”A critical evaluation of the IPv6 routing protocol for low power
and lossy networks (RPL).” In 2011 IEEE 7th International Conference on Wireless and Mobile Computing, Networking
and Communications (WiMob), pp. 365-372. IEEE, 2011.
[3] Bouhafs, Faycal, Michael Mackay, and Madjid Merabti. ”Links to the future: Communication requirements and challenges
in the smart grid.” IEEE Power and Energy Magazine 10, no. 1 (2011): 24-32.
[4] Pister, Kris, Pascal Thubert, Sicco Dwars, and Tom Phinney. Industrial routing requirements in low-power and lossy
networks. No. rfc5673. 2009.
[5] Cam-Winget, Nancy, J. Hui, and D. Popa. Applicability statement for the routing protocol for low-power and lossy
networks (RPL) in advanced metering infrastructure (AMI) networks. No. rfc8036. 2017.
[6] J. Martocci, P. De. Mil, N. Riou and W. Vermeylen. Building automation routing requirements in low-power and lossy
networks. No. rfc5867. 2010.
[7] Winter, Tim, et al. RPL: IPv6 routing protocol for low-power and lossy networks. No. rfc6550. 2012.
[8] Zhang, Tao, and Xianfeng Li. ”Evaluating and Analyzing the Performance of RPL in Contiki.” In Proceedings of the first
international workshop on Mobile sensing, computing and communication, pp. 19-24. 2014.
[9] O. Gnawali and P. Levis, 2012. The minimum rank with hysteresis objective function (No. rfc6719).
[10] Pradeska, N., Najib, W. and Kusumawardani, S.S., 2016, October. Performance analysis of objective function MRHOF
and OF0 in routing protocol RPL IPV6 over low power wireless personal area networks (6LoWPAN). In 2016 8th
international conference on information technology and electrical engineering (ICITEE) (pp. 1-6). IEEE.
[11] Rajalingham, G., Gao, Y., Ho, Q.D. and Le-Ngoc, T., 2014, September. Quality of service differentiation for smart grid
neighbor area networks through multiple RPL instances. In Proceedings of the 10th ACM symposium on QoS and security
for wireless and mobile networks (pp. 17-24).
[12] H. Ali, A Performance Evaluation of RPL in Contiki- A Cooja Simulation based study, Master Thesis, Swedish Institute
of Computer Science (SICS, Stockholm sweden), October, 2012
[13] Monowar, M.M. and Basheri, M., 2020. On providing differentiated service exploiting multi-instance RPL for industrial
low-power and lossy networks. Wireless Communications and Mobile Computing, 2020.
[14] Al-Abdi, A., Mardini, W., Aljawarneh, S. and Mohammed, T., 2019, December. Using of multiple RPL instances for
enhancing the performance of IoT-based systems. In Proceedings of the Second International Conference on Data Science,
E-Learning and Information Systems (pp. 1-5).
[15] Nassar, J., Gouvy, N. and Mitton, N., 2017, November. Towards multi-instances QoS efficient RPL for smart grids. In
Proceedings of the 14th ACM Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, & Ubiquitous
Networks (pp. 85-92).
[16] Banh, M., Mac, H., Nguyen, N., Phung, K.H., Thanh, N.H. and Steenhaut, K., 2015, October. Performance evaluation of
multiple RPL routing tree instances for Internet of Things applications. In 2015 international conference on advanced
technologies for communications (ATC) (pp. 206-211). IEEE.
[17] Mardini, W., Aljawarneh, S. and Al-Abdi, A., 2021. Using multiple RPL instances to enhance the performance of new 6G
and Internet of Everything (6G/IoE)based healthcare monitoring systems. Mobile Networks and Applications, 26(3),
pp.952-968.
[18] Bhandari, K.S., Ra, I.H. and Cho, G., 2020. Multi-topology based QoSdifferentiation in RPL for internet of things
applications. IEEE Access, 8, pp.96686-96705.

492
 J. Electrical Systems 20-7s (2024):482-494

[19] Bouzebiba, H. and Lehsaini, M., 2020. Freebw-rpl: A new rpl protocol objective function for internet of multimedia things.
Wireless Personal Communications, 112(2), pp.1003-1023.
[20] Long, N.T., Uwase, M.P., Tiberghien, J. and Steenhaut, K., 2013, October. QoSaware cross-layer mechanism for multiple
instances RPL. In 2013 International Conference on Advanced Technologies for Communications (ATC 2013) (pp. 4449).
IEEE.
[21] Brandon Foubert. Cooperation between multiple RPL networks. Networking and Internet Architecture [cs.NI]. 2018.
ffhal-02307955ff
[22] Draves, R., Padhye, J. and Zill, B., 2004, September. Routing in multi-radio, multi-hop wireless mesh networks. In
Proceedings of the 10th annual international conference on Mobile computing and networking (pp. 114-128).
[23] Gaddour, O., Koubˆaa, A., Baccour, N. and Abid, M., 2014, May. OF-FL: QoSaware fuzzy logic objective function for
the RPL routing protocol. In 2014 12th International symposium on modeling and optimization in mobile, ad hoc, and
wireless networks (WiOpt) (pp. 365-372). IEEE.
[24] Kamgueu, P.O., Nataf, E. and Djotio, T.N., 2015, October. On design and deployment of fuzzy-based metric for routing
in low-power and lossy networks. In 2015 IEEE 40th Local Computer Networks Conference Workshops (LCN
Workshops) (pp. 789-795). IEEE.
[25] Kim, H.S., Paek, J. and Bahk, S., 2015, June. QU-RPL: Queue utilization based RPL for load balancing in large scale
industrial applications. In 2015 12th Annual IEEE International Conference on Sensing, Communication, and Networking
(SECON) (pp. 265-273). IEEE.
[26] Vasseur, J.P., Kim, M., Pister, K., Dejean, N. and Barthel, D., 2012. Routing metrics used for path calculation in low-
power and lossy networks (No. rfc6551).
[27] Yang, S., Baek, Y., Kim, J., Cho, K. and Han, K., 2009, February. A routing metric for load balance in wireless mesh
networks. In 2009 11th International Conference on Advanced Communication Technology (Vol. 3, pp. 1560-1565). IEEE.
[28] N. BUI, A. Castellani, P. Casari, M. Rossi, L.Vangelista, M. Zorzi, “Smart grids using wireless sensors and actuators
networks,” Chapter in book “Smart Grid Communications and Networking,” Cambridge University Press, UK, June 2012.
[29] Gubbi, J., Buyya, R., Marusic, S. and Palaniswami, M., 2013. Internet of Things (IoT): A vision, architectural elements,
and future directions. Future generation computer systems, 29(7), pp.1645-1660.
[30] Al-Turjman, F., Ever, E. and Zahmatkesh, H., 2018. Small cells in the forthcoming 5G/IoT: Traffic modelling and
deployment overview. IEEE Communications Surveys & Tutorials, 21(1), pp.28-65.
[31] Chowdhury, M.Z., Shahjalal, M., Ahmed, S. and Jang, Y.M., 2020. 6G wireless communication systems: Applications,
requirements, technologies, challenges, and research directions. IEEE Open Journal of the Communications Society, 1,
pp.957-975.
[32] Taghizadeh, S., Bobarshad, H. and Elbiaze, H., 2018. CLRPL: context-aware and load balancing RPL for IoT networks
under heavy and highly dynamic load. IEEE access, 6, pp.23277-23291.
[33] Al Ameen, M., Liu, J. and Kwak, K., 2012. Security and privacy issues in wireless sensor networks for healthcare
applications. Journal of medical systems, 36(1), pp.93-101.
[34] Agustin, J.P.C., Jacinto, J.H., Limjoco, W.J.R. and Pedrasa, J.R.I., 2017, November. IPv6 routing protocol for low-power
and lossy networks implementation in network simulator—3. In TENCON 2017-2017 IEEE Region 10 Conference (pp.
3129-3134). IEEE.
[35] Mardini, W., Aljawarneh, S., Al-Abdi, A. and Taamneh, H., 2018, March. Performance evaluation of RPL objective
functions for different sending intervals. In 2018 6th international symposium on digital forensic and security (ISDFS)
(pp. 1-6). IEEE.
[36] Borges, L.M., Velez, F.J. and Lebres, A.S., 2014. Survey on the characterization and classification of wireless sensor
network applications. IEEE Communications Surveys & Tutorials, 16(4), pp.1860-1890.
[37] Al-Fuqaha, A., Guizani, M., Mohammadi, M., Aledhari, M. and Ayyash, M., 2015. Internet of things: A survey on enabling
technologies, protocols, and applications. IEEE communications surveys & tutorials, 17(4), pp.2347-2376.
[38] Ghaleb, B., Al-Dubai, A.Y., Ekonomou, E., Alsarhan, A., Nasser, Y., Mackenzie, L.M. and Boukerche, A., 2018. A survey
of limitations and enhancements of the ipv6 routing protocol for low-power and lossy networks: A focus on core
operations. IEEE Communications Surveys & Tutorials, 21(2), pp.1607-1635.
[39] Suljanovic, N., Borovina, D., Zajc, M., Smajic, J. and Mujcic, A., 2014, May. Requirements for communication
infrastructure in smart grids. In 2014 IEEE International Energy Conference (ENERGYCON) (pp. 1492-1499). IEEE.
[40] Kim, H.S., Kim, H., Paek, J. and Bahk, S., 2016. Load balancing under heavy traffic in RPL routing protocol for low
power and lossy networks. IEEE Transactions on Mobile Computing, 16(4), pp.964-979.
[41] Karkazis, P., Trakadas, P., Leligou, H.C., Sarakis, L., Papaefstathiou, I. and Zahariadis, T., 2013. Evaluating routing metric
composition approaches for QoS differentiation in low power and lossy networks. Wireless networks, 19(6), pp.1269-
1284.
[42] Hassan, A., Alshomrani, S., Altalhi, A. and Ahsan, S., 2016. Improved routing metrics for energy constrained
interconnected devices in low-power and lossy networks. Journal of communications and networks, 18(3), pp.327-332.
[43] Chen, Y., Chanet, J.P., Hou, K.M., Shi, H. and De Sousa, G., 2015. A scalable context-aware objective function (SCAOF)
of routing protocol for agricultural low-power and lossy networks (RPAL). Sensors, 15(8), pp.19507-19540.
[44] Lamaazi, H. and Benamar, N., 2020. A comprehensive survey on enhancements and limitations of the RPL protocol: A
focus on the objective function. Ad Hoc Networks, 96, p.102001.
493
 J. Electrical Systems 20-7s (2024):482-494

[45] Alishahi, M., Yaghmaee Moghaddam, M.H. and Pourreza, H.R., 2018. Multi-class routing protocol using virtualization
and SDN-enabled architecture for smart grid. Peer-to-Peer Networking and Applications, 11(3), pp.380-396.
[46] Lamaazi, H. and Benamar, N., 2018. OF-EC: A novel energy consumption aware objective function for RPL based on
fuzzy logic. Journal of Network and Computer Applications, 117, pp.42-58.
[47] Zhao, M., Ho, I.W.H. and Chong, P.H.J., 2016. An energy-efficient region-based RPL routing protocol for low-power and
lossy networks. IEEE Internet of Things Journal, 3(6), pp.1319-1333.
[48] Wang, Z., Zhang, L., Zheng, Z. and Wang, J., 2018. Energy balancing RPL protocol with multipath for wireless sensor
networks. Peer-to-Peer Networking and Applications, 11(5), pp.1085-1100.
[49] De Couto, D.S., Aguayo, D., Bicket, J. and Morris, R., 2003, September. A highthroughput path metric for multi-hop
wireless routing. In Proceedings of the 9th annual international conference on Mobile computing and Networking (pp.
134-146).
[50] Wang, Y.M. and Luo, Y., 2010. Integration of correlations with standard deviations for determining attribute weights in
multiple attribute decision making. Mathematical and Computer Modelling, 51(1-2), pp.1-12.
[51] Warneke, B.A. and Pister, K.S., 2002, September. MEMS for distributed wireless sensor networks. In 9th international
conference on electronics, circuits and systems (Vol. 1, pp. 291-294). IEEE.
[52] Karkazis, P., Leligou, H.C., Sarakis, L., Zahariadis, T., Trakadas, P., Velivassaki, T.H. and Capsalis, C., 2012, July. Design
of primary and composite routing metrics for RPL-compliant wireless sensor networks. In 2012 international conference
on telecommunications and multimedia (TEMU) (pp. 13-18). IEEE.
[53] Gonizzi, P., Monica, R. and Ferrari, G., 2013, July. Design and evaluation of a delay-efficient RPL routing metric. In 2013
9th International Wireless Communications and Mobile Computing Conference (IWCMC) (pp. 1573-1577). IEEE.
[54] Dunkels, A., Gronvall, B. and Voigt, T., 2004, November. Contiki-a lightweight and flexible operating system for tiny
networked sensors. In 29th annual IEEE international conference on local computer networks (pp. 455-462). IEEE.
[55] Sundmaeker, H., Guillemin, P., Friess, P. and Woelffl´e, S., 2010. Vision and challenges for realising the Internet of Things.
Cluster of European research projects on the internet of things, European Commision, 3(3), pp.34-36.
[56] Aljawarneh, S.A., Elkobaisi, M.R. and Maatuk, A.M., 2017. A new agent approach for recognizing research trends in
wearable systems. Computers & Electrical Engineering, 61, pp.275-286.
[57] Floris, A. and Atzori, L., 2015, June. Quality of Experience in the Multimedia Internet of Things: Definition and practical
use-cases. In 2015 IEEE International Conference on Communication Workshop (ICCW) (pp. 1747-1752). IEEE.
[58] Huang, X., Xie, K., Leng, S., Yuan, T. and Ma, M., 2018. Improving Quality of Experience in multimedia Internet of
Things leveraging machine learning on big data. Future Generation Computer Systems, 86, pp.1413-1423.
[59] Chaudhari, S.S. and Biradar, R.C., 2015. Survey of bandwidth estimation techniques in communication networks. wireless
personal communications, 83(2), pp.1425-1476.
[60] Charles, A.J. and Kalavathi, P., 2018. QoS measurement of RPL using Cooja simulator and Wireshark network analyser.
International Journal of Computer Sciences and Engineering, 6(4), pp.283-291.

494