See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/369242727

Remote Patient Monitoring Systems: Applications, Architecture, and

Challenges

Article in Scientific African · March 2023

DOI: 10.1016/j.sciaf.2023.e01638

CITATIONS READS

5 251

5 authors, including:

Kegomoditswe Boikanyo Murtala Zungeru Adamu

Botswana International University of Science and Technology Botswana International University of Science and Technology
2 PUBLICATIONS 5 CITATIONS 123 PUBLICATIONS 1,666 CITATIONS

SEE PROFILE SEE PROFILE

Boyce Sigweni Caspar K Lebekwe

Brunel University London Botswana International University of Science and Technology
22 PUBLICATIONS 282 CITATIONS 26 PUBLICATIONS 170 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Murtala Zungeru Adamu on 22 March 2023.

The user has requested enhancement of the downloaded file.

 Scientific African 20 (2023) e01638

Contents lists available at ScienceDirect

Scientific African
journal homepage: www.elsevier.com/locate/sciaf

Information Technology and Engineering

Remote patient monitoring systems: Applications,

architecture, and challenges
Kegomoditswe Boikanyo∗, Adamu Murtala Zungeru∗, Boyce Sigweni, Abid Yahya,
Caspar Lebekwe
Department of Electrical, Computer and Telecommunications Engineering, Botswana International University of Science and Technology,
Private Bag 16, Palapye, Botswana

a r t i c l e i n f o a b s t r a c t

Article history: Research in Remote Patient Monitoring Systems (RPMS) is considered to be one of the
Received 25 October 2022 most crucial fields since it deals with human lives. The rise in usage of RPMS has in-
Revised 1 March 2023
creased since the emergence of the pandemic. Even though there is a rise in these sys-
Accepted 7 March 2023
tems, there are some challenges, such as mobility, heterogeneous networks, standardiza-
tion of RPMSs, automation, and Quality of Service (QoS). Our discussion focuses on RPMS
Editor DR B Gyampoh systems for physiological parameter monitoring in the areas of their applications, archi-
tecture, and challenges. Thus, an in-depth review of RPMS and the analysis of these data
Keywords:
are performed in order to understand where the current RPMS literature stands. The lit-
Remote patient monitoring system (RPMS)
erature shows that research in these RPMS is concentrated on two or more of the follow-
Quality of service (QoS)
Health care ing areas: applications, architecture, methodologies, and their performance. It appears that
Mobile Adhoc networks (MANET) prior to 2020, researchers focused on nearly all aspects of RPMS until the pandemic. Then
Wireless Body Area Network (WBAN) there was a shift in RPMS research to focus more on the applications and architectures
Wearable sensors of these systems. As a result, more companies are developing mobile RPMS. In this paper,
we present a detailed of various existing RPMS with areas of focus on their application,
architecture, technology applied, and challenges faced. We further provided a comparative
and statistical analysis of the existing literature, and, finally, an overview of Quality of Ser-
vice (QoS) as one challenge of RPMS is provided. The surveyed QoS requirements based on
traffic type, data quality, device quality and network metrics are provided, with the aim of
providing the current trend for researchers and industries to adapt to the best approach in
the design of quality-aware RMPS. We then conclude the work by providing future work,
which, when adopted, will brighten the future of RPMS deployment.
© 2023 The Author(s). Published by Elsevier B.V. on behalf of African Institute of
Mathematical Sciences / Next Einstein Initiative.
This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)

Introduction

A growing number of remote patient monitoring systems (RPMS) are becoming popular due to the emergence of highly
infectious diseases such as COVID-19, the demographic shift toward the ageing population, and the rise in health complica-

∗
Corresponding authors.
E-mail addresses: [email protected] (K. Boikanyo), [email protected] (A.M. Zungeru).

https://doi.org/10.1016/j.sciaf.2023.e01638
2468-2276/© 2023 The Author(s). Published by Elsevier B.V. on behalf of African Institute of Mathematical Sciences / Next Einstein Initiative. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 1. General Overview of a Remote Patient Monitoring System Architecture.

tions. Furthermore, technological advances have made RPMS more widely used. In RPMS, specific groups of patients are tar-
geted, such as those with chronic diseases, infectious diseases during isolation, and those with mobility challenges or other
disabilities [1]. The systems are also used for post-surgery patients, newborns, and elderly patients. The goal of healthcare is
to keep patients as comfortable as possible during their everyday lives. Patients are, therefore able to move freely and exer-
cise in a private environment [2]. Patient monitoring systems traditionally employed wired sensors connected to computers
inside hospitals. The disadvantage of these systems was that they restricted patient’s mobility. The devices that were used
were also large and expensive, and could only monitor a few patients. RPMS were introduced when healthcare facilities
expanded to offer home-based care. This system was not user-friendly, just like traditional patient monitoring. Technology
improved over time, allowing researchers to create solutions for remote patient monitoring and wirelessly connected pa-
tient monitoring. The health sector is improving greatly as more researchers and companies use remote patient monitoring
systems to improve care, and the business is growing rapidly [3–5].
In a remote patient monitoring system, physiological data is collected through biomedical sensors in order to evaluate
the well-being of patients outside of the hospital setting [6]. Its purpose is to move traditional clinical settings into people’s
homes, where they live, work, and play. In order for the appropriate decisions to be made, the data collected is wirelessly
transmitted to the healthcare provider [7]. Using this type of tele-health, physiological data such as heart rate/pulse rate,
respiratory rate, body temperature, blood pressure, and blood oxygen saturation can be measured and transmitted [8,9].
RPMS pose advantages of real-time detection of illness and the ability to monitor patients’ health conditions constantly.
When untimely deaths are detected, emergency measures should be taken. Additionally, these systems reduce healthcare
costs by utilizing different communication technologies. A patient can also participate in normal daily activities while on
treatment. In addition, the RPMS improves traffic accident emergency care and mobility [10].
Wireless sensors and wireless communication are used to acquire and transmit data to the hospital during remote mon-
itoring of patients. Sensors such as wearable and implantable (contact-based sensors) and non-contact ones are used during
data acquisition. The acquired data can be processed and transformed to a suitable form using an internal or external con-
troller before being transmitted wirelessly to the hospital. Various forms of short and long-range communications are used
to relay the data to the end-user [11]. Generally, this form of tele-health pursues three phases of communication networks
as shown in Fig. 1.
The growing number of things connected to the internet has formed the concept of present heterogeneous networks.
The Internet of things exploits the incorporation and interoperability of these heterogeneous networks. It also implements
smartness into many physical objects; like smart cars, homes, health, industry, and smart energy according to [12]. There
are also some challenges like mobility, QoS, automated system, and security that have shown some flaws in RPMS. The het-
erogeneous nature of RPMS which reveals the non-uniform architectural design, network communications, locations, events,
and environments makes the Quality of Service (QoS) to be more complex. Therefore meeting QoS requirements regarding
the network parameters will ensure efficiency improvements in RPMS [13]. IoV in one framework that can be used to mit-
igate the challenges of RPMS in the sense of mobility and routing protocols. Firstly IoV will allow the RPMS nodes which
are either inside the car or along the road to share the tasks of data forwarding and computations and the vehicles keep
on entering and leaving the network. Secondly, IoV will greatly reduce transmission distance between the nodes assuming
that the IoV mobility pattern is well designed. Third, the latency of the network will be reduced while the throughput is
increased. Finally, there will always be connectivity even under sparse or disconnected sensor networks. In this paper, a
comprehensive review and up-to-date study of existing RPMS and their characteristics are performed. Further analysis of
the existing remote patient monitoring system based on their application is done. A discussion of other features of RPMS
and their applications was also performed.

2
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 2. Classification of Focus in Existing RPMS literature.

The focal contributions of this paper are as follows:

1 A detailed of various existing RPMS with areas of focuses in to their application, architecture, and technology applied,
and challenges faced.
2 A comparative and statistical analysis of the existing literature in RPMS.
3 An overview of Quality of Service (QoS) as one challenge of RPMS surveying the QoS requirements based on traffic type,
data quality, and device quality and network metrics.
The rest of the paper is organized as follows: Section II highlights the general related works. Section III gives a detailed
literature of the existing RPMS according to their application, architecture, the technology used, and challenges. Section IV
provides an analytical comparison of classified applications. Conclusions are drawn in Section V, while future aspects are in
Section VI.

Related work

The advancement of the use of heterogeneous devices and their communication challenges have influenced most authors
to survey RPMS from areas of their application, architecture, the technology used, and challenges. Remote patient monitoring
systems as one of the smart city applications can utilize the IoV framework to improve the challenges of mobility, energy
efficiency, QoS, and many more. Nonetheless, many researchers already implemented Mobile Adhoc networks (MANET) like
Vehicular Adhoc Networks (VANETs) applications in the context of remote patient monitoring systems in the way of improv-
ing the health sector, and the growth is tremendous [14]. Even though these ideas are effective, there are still several flaws
and challenges of mobility, network lifetime, and energy efficiency that need to be addressed. IoV is one communication
model that can improve and solve the problems faced by RPMS. The literature review in Table 1 summarizes some existing
solutions and experiences in remote patient monitoring systems. The QoS as one challenge faced by RPMS are also reviewed.

Overview of existing remote patient monitoring system

The following section indicates the literature in areas of focus for remote patient monitoring systems. These areas are
depicted in more details in Fig. 2 and Table 2. Fig. 2 also gives a brief summary of the extant RPMS literature. Apparently,
research in these systems focuses on two or more of the following areas: applications, architecture, methodologies, and their
performance. The focus of application in RPMS is mostly on two aspects: usage and targeted clients. Furthermore, the device
is used for vital monitoring or disease diagnosis and the targeted clients are infants or the elderly.
There is also an architectural area that details how the RPMS functions and interacts with other system components in
general. Sensor type, MCU, and wireless connectivity are all mentioned in the architecture. The study of techniques such
as the Markov model, mobility model, MAC protocol, routing protocol, game theory, and cooperative communication is also
part of RPMS research. Another field examines system performance, particularly through the use of QoS measurements.
Table 2 provides an overall summary of these references, with percentages indicating the weights of the area of attention
in RPMS according to the literature collected. The results are further illustrated on the bar chart in Fig. 3. A recent study by
RPMS shows that the global remote patient monitoring market has been growing at an accelerated rate since 2020. This has
led to an increase in companies developing these systems [22,23].

3
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Table 1
Comparison of related works on RPMS.

The uniqueness of this

Title Citations Year Description Pros Cons paper

Remote patient [14] 2019 Reviewed recent Discussed contactless Concentrated only on Contact and
monitoring: a advances in RPMS camera-based method RPMS sensors. contactless sensors are
comprehensive study including contact and and contact-based analysed according to
contactless methods. methods specific diseases
Body Area Sensor [15] 2016 Reviewed the Designed a new WBAN Focuses only on MAC Relate power
Networks Challenges opportunities and time division multiple protocols consumption and data
and Opportunities - challenges of using access protocol based rates from RPMS
University of Virginia WBAN for RPMS on WBAN RPMS sensors to the internet.
Wearable sensors for [16] 2017 Reviewed and Designed several Did not review Represented different
remote health compared low-cost control circuits for technologies and RPMS control systems
monitoring and non-invasive RPMS data collection in algorithms for data based on diseases
for elderly people RPMS aggregation
Internet of Things for [17] 2021 Survey emerging Gives a broad overview Does not address A review of the
Healthcare applications in the of the application of technologies deployed current developments
Technologies Internet of things IoT in RPMS on the in computations of in biomedical
based RPMS. bases of RPMS these IoT based RPMS technology and
solutions, big data and different research
disease predictions issues
A survey on mobility [18] 2017 Survey approaches and Addressed mobility Focus on research Described main
support in wireless important features in model for both challenges for mobility operational procedures
body area networks RPMS mobility models individual and support in WBAN for mobility model and
group-based RPMS compared them.
Health Monitoring [19] 2020 analyse rpms to analyse the existing Does not state the Proposed IoMT-based
Based on Internet of enhance their mobility routing and mac routing and MAC disaster rescuer health
Medical Things: problems protocols research and protocols used by monitoring system
Architecture, Enabling identify the enabling these researchers. with its technologies
Technologies, and technologies and routing protocols.
Applications
Deployment of IoV for [20] 2018 Surveyed and IoV Focuses mainly on Gave a general Proposed a Universal
Smart Cities: paradigms and how IoV can benefit overview of how IoV IoV which allow
Applications, network models in a the smart city benefits smart city vehicles to
Architecture and smart city scenario environment applications. communicate with
Challenges almost everything
connected to the
internet
A new healthcare [21] 2020 Heart-related diseases The monitoring system No mathematical The monitoring system
architecture using IoV associated with IoV allows mobility and model of how the uses the IoV
technology for a are monitored in order improved energy healthcare monitoring technology primarily
continuous health to improve data efficiency and network system is incorporated for heart disease
monitoring system collection and the lifetime to the IoV network. monitoring.
ambulance’s response
time.

Table 2
Overall summary of areas of focus in RPMS.

Focus Of Study Percentage (%) Refs.

Application
Usage 21.9% [1,10,24–35]
Targeted Clients 7.8% [36–40]
Architecture
Sensor type and Mobile Control Unit 12.5% [41–48]
Wireless Communication 12.5% [49–56]
Techniques Applied
Markov Model 6.3% [57–60]
Mobility Model 4.7% [61–63]
MAC Protocol 4.7% [64–66]
Routing Protocol 15.6% [67–76]
Game Theory 3.1% [77,78]
Cooperative Communication 4.7% [79–81]
Challenges
6.3% [73,82,83]

4
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 3. Percentages relatives to the other areas of focus in RPMS literature.

Application of remote patient monitoring systems

This section summarizes the known literature on why RPMS are used (and for whom) (Targeted Clients). The reason for
analysing these two components of RPMS application is since most researchers have not evaluated them as previously said.
They either concentrate on one aspect or ignore them entirely. These two groups are crucial since they complement one
another. Knowing what the RPMS is used for, for example, it will aid in determining which client is of interest.

Application based on usage

RPMS are intended to mirror the traditional method of patient monitoring used by a doctor, nurse, or caregiver. As a
result, they generally execute the aforementioned activities but are not in the hospital setting. Vital Monitoring and Disease
Diagnosis are two examples.

Vital monitoring. Vital Monitoring is a routine check-up that measures fundamental physiological parameters such as body
temperature, heart rate, blood oxygen saturation, blood pressure, and breathing rate. In RPMS, monitoring is performed
away from the hospital. This is significant, as efficiency in hospitals will be enhanced. There will also be fewer hospital
admissions since more patients will be monitored at home. One or more physiological parameters may be considered during
remote patient monitoring. Depending on the sort of patient being observed, monitoring can be done in real-time or on
a regular basis. This is also influenced by the condition being diagnosed. For example, blood pressure may be tested to
diagnose hypertension. Vital monitoring also determines whether the patient is deteriorating or not [24]. As a result, various
researchers devised and applied vital monitoring RPMS to detect deterioration or irregularities in the patient’s body.
A study in [25] described a wearable, low-cost, small reflectance pulse oximeter capable of providing accurate photo-
plethysmography (PPG) data without the need of filters. The PPG signal is transferred at 240 Hz to a host computer through
a Zigbee transceiver module or a mini-USB interface. To remove ambient noise and slow-varying motion distortions, com-
puter processing is required. The RPMS can measure heart rate, breathing rate, blood oxygen saturation, and blood pressure.
These RPMS measure more physiological variables, making them more reliable.
Researchers in [26] reported a body temperature RPMS with 0.1 °C precision throughout a 16–42 °C range. It uses two
digital temperature sensors to monitor the temperature of the two ear canals and communicates the data through Bluetooth
for further analysis. Using ear canal measures RPMS can cause ear injury. Sensor probes can cause tiny fluctuations in
measurements while in motion. As a result, wearable RPMS will be preferable.
For automatic screening of potential COVID-19 patients, Authors in [27] used XGBoost and logistic regression algorithms.
Physiological indicators such as respiration, heart rate, body movements, sleep quality, and sleep quality were monitored
using ultra-wideband radio-impulse radar. Their study compared 140 radar monitoring data from 23 COVID-19 patients
with 144 radar monitoring data from healthy controls. XGBoost + LR showed excellent discrimination surpassing other
single machine learning algorithms. The study, however, had some drawbacks as well. Environmentally factors and other
parameters, may affect controlled trials in RPMS. The accuracy index of this study may be lower in practice given the small
sample size and lack of multiple age groups. In addition, the proposed RPMS are limited to monitoring one patient at time,
reducing its utility in a real-world hospital. The authors verified if there were any variations in nightly respiration, heart
rate, body movement, and sleep quality between COVID-19 patients and healthy volunteers as shown in Fig. 4.
The authors of [29] used a temperature and heart beat RPMS. The microcontroller controls the temperature sensor read-
ing, which is displayed on an LCD panel. They also employed an RF transmitter to send data to an RF receiver, which uses a
microcontroller to display the data on an LCD screen. The systems in the preceding studies were designed to monitor one or
two physiological parameters, whereas the authors of [28] constructed a multi-vital monitoring system that monitors all five
physiological parameters. The enhanced system is built using Internet of Things (IoT) sensors that collect these parameters
from the patient remotely. The WBAN in the RPMS is used to send data to the doctor’s smartphone. The phone includes an
application that assists doctors in making decisions. This kind of RPMS will also administer life-saving medications to the
patient while the ambulance is in motion. It will also provide the doctor with a virtual presence within the ambulance.

5
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 4. Distribution of (a) heart rate and respiratory rate and (b) mean body dynamic density and wake up times in healthy subjects and COVID-19 patients
[27].

Fig. 5. Wristwatch device model: (a) exploded view, (b) on a Speag phantom arm.

Another study [30] describes an RPMS for measuring blood oxygen saturation and heart rate. A unique 868 MHz wrist-
watch subsystem with an emphasis on optical PPG is disclosed. The 868 MHz and 2.45 GHz bands are being compared.
Using the designed wrist-worn sensor platform, the system was successfully applied in a clinical context for the measure-
ment of SpO2 and heart rate. The developed sensor platform can be used in future RPMS applications as a consequence of
the successful clinical testing. In the future, the gateway will be able to communicate directly with the cloud, eliminating
the need for mobile phones. The Fig. 5 illustrate the Wristwatch device model that can be used in monitoring of patients.
Vital monitoring RPMS can help detect disease development. This resulted in the development of RPMS for disease diag-
nostics.

Disease diagnostic monitoring. Disease diagnostic RPMS involve examination of physiological indicators and other health
problems in order to diagnose a specific disease. High blood pressure and dizziness can indicate hypertension, while a
higher temperature with headaches can indicate several cold-related disorders. Researchers have created a disease diagnostic
method that only targets one or more forms of sickness.
Pathinarupothi et al. [24] describe RPMS-based fall detection. For data collection, a fall detection sensor is used in con-
junction with an Amazon Echo speaker, and a webcam. As a controlling device, they employed a Raspberry Pi model B+,
and WI-FI was used to communicate the collected data to the doctor. When the hub detects a fall, it interacts with the user
via speakers. Amazon Echo devices have voice recognition capabilities and can gain user confirmation of falls. Because the
RPMS confirms falls before connecting with associated parties and obtaining assistance, it decreases the false positive rate
as it relies on user confirmation beforehand. The writers of the research report [32] use a devised RPMS to track mentally
bipolar people. This system collects speech while monitoring and also contains a questionnaire system, medicine intake
monitoring, and sleep activity monitoring. It is built on an autonomic nerve system made of textile. This approach can mon-
itor mentally bipolar people, although employing a questionnaire may result in erroneous results because the patients are
mentally handicapped.
Prabhakar and Rajaguru’s latest study [33] includes the development of a remote patient monitoring system for epilepsy.
The primary focus of this work is on the use of spatial reduction on obtained EEG data. FMI was employed to do this. In this
paper, the Singular Value Decomposition Based Partial Transmit Scheme (SVD-PTS) is introduced as a data transfer method,
while the Gaussian Kernel Based Support Vector Machine (G-SVM) is used as a classification approach. They report a 95.38%
accuracy rate with a time delay of only 2.19 s. Their proposed model is described in the block diagram on Fig. 6. The study’s
[34] researchers created an automatic feedback message for a generation in a diabetic management system. The device may
track food intake, blood pressure, insulin levels, calories, medications, and activity. It also has chat functionality between the
doctor and the patient, as well as a message repository from which the doctor can send a message to the patient as needed
These RPMS solutions allow interaction between the patient and the doctor, which helps improve trust in these devices.
Applications are important in disease diagnostic because it allows the patient to understand when they are healthy and
un-healthy. Stavropoulos et al. [35] developed a free Android app for retrieving disease-related data in RPMS. Each user
can diagnose remotely and communicate with the doctor via phone or messaging. A cost-cutting communication approach

6
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 6. Block diagram of the work proposed in [33].

is also proposed. Because it does not describe how physiological parameters would be gathered, the system may not be
feasible. When embedded in a vital monitoring system, the system becomes more robust.

Application based on targeted clients

Knowing the user is essential when creating RPMS because it dictates the size, material, and aesthetic. Users can be
infants, children, the elderly, males or females, disabled or not disabled.
Chen et al. [36] present a non-invasive RPMS using a wearable reflectance sensor for long-term monitoring of new-
born babies’ oxygen saturation. This device transmits data via an RF transceiver, which assess heart rate and blood oxygen
saturation. Design improvements will reduce motion artefacts that can create erroneous readings.
The authors of the research paper [37] also suggest a video-based remote patient monitoring system for new-borns. A
new algorithm is utilized to calculate respiratory rates for neonates aged 33 to 40 weeks. The final method used a patented
technique of micro motion and stationary detection to measure respiratory motions and background noise (MSD). The corre-
lation between MSD and the current hospital standard, the ECG impedance pneumographic, was r = 0.9485. When compared
to typical continuous monitoring, the inaccuracy in this video-based system was 1.33 breaths per minute and the root mean
square of 6.36 breaths per minute. The system should be capable of monitoring a large number of patients for an extended
period of time.
For elderly person, the work in [31] proposes to use the Internet of Things (IoT) to deploy a bed-exit patient monitoring
system for use in clinical settings, which can warn healthcare staff and the elderly to the presence of a patient requiring
medical help. Researchers examined two datasets derived from two separate accelerometer-based measures, namely obser-
vations from cell phones and RFID-based accelerometers. It has been found that using the proposed technique is useful in
tracking bed escapes and identifying other ambulatory activities during monitoring. Furthermore, the proposed work reduces
the average end-to-end system time.
The research paper [39] provides a framework for a cloud-based remote patient monitoring system (CloudDTH). Wear-
able devices were used with this framework to monitor, diagnose, and predict persons’ health, with the goal of improv-
ing their health, particularly as they age. As a result, the actual and virtual worlds will be combined in this system. Fur-
thermore, the system was able to reduce the end-to-end delay by classifying and alerting workers to potentially alarm-
ing condition shifts in one-tenth of a second. A real-world test and validation of the system would be advantageous as
well.

Remote patient monitoring based on architecture

The architecture of RPMS describes the system’s structure, how it is layered, and how the layers interact with one an-
other. When it comes to human life, the architecture for RPMS is troubled with chaos and several obstacles. To begin, what
sensor brands should be used to gather accurate and reliable data from patients? Second, which processing methods and
communication protocols will be more appropriate for this system, and finally, whether the medical community and patients
would approve the idea [4]?
In this section, a description of RPMS will be examined using Fig. 7. The figure shows how the amount of tiers developed
might affect the architecture. This section will go over the following subtopics: sensor type, Mobile Control Unit used, and
wireless connection. Because some researchers focus solely on one of the tiers and ignore the other, the RPMS are classified
as described above in this section.

7
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 7. Remote Patient Monitoring System Achitecture.

Fig. 8. Parts of a Wireless Sensor.

Fig. 9. Exploded view of Sensor Type 1.

Phychological data acquisition and processing in remote patient monitoring system

Tier 1 is typically used for psychological data acquisition and processing in a remote patient monitoring system. Tier 1
entails collecting physical biomedical data with biomedical sensors, signal conditioning, processing, and storing the data for
viewing or distribution to medical service providers [84]. This tier is critical in RPMS because it is where an entire system’s
accuracy, sensitivity, and speed are tested. Data capture and processing utilizing electronic devices has grown easier, more
dependable, and more accurate as a result of technological innovation. These processes can be carried out without the
need for human interaction. The sensor nodes are small, literally the size of a coin, with low-cost computation that collects
physiological data from the patient and relays it to the MCU for further processing. Body temperature, heart rate, respiratory
rate, blood oxygen saturation, and blood pressure are examples of physiological data. As shown in Fig. 8, each sensor node
is equipped with a sensor device, an actuator, an energy supply, a CPU, storage, and radio transceivers, according to the
authors of [15,16,85,86]
A measuring system, also known as a sensor, will often include a transducer that converts the measured quantity into
an electrical quantity, as well as a signal conditioner for amplification, filtering, modulation, demodulation, and analogue to
digital conversion (ADC). The sensor will also comprise storage, an energy source, and a radio transceiver that will send the
measured parameters to a control centre or via a processor that will use a microprocessor (P), a microcontroller (C), or a
personal device.
In this survey, at least three types of sensor nodes in the Remote Patient Monitoring System are studied from the lit-
erature [80], and an exploded view of the sensors in Fig. 7, are shown in Fig. 9 to Fig. 11. The first type of sensor has an
external processor and can connect with it wirelessly or wired, as shown in Fig. 6. The processor does the processing and
transfers the gathered data to the hospital through the internet via a relaying node (WWAN). Fig. 10
The second form of sensor, as shown in Fig. 7, has an external processor but also a gateway between the sensor and the
controller. The gateway serves as a connection point between the sensor and the processor. Because sensors have various
topologies, frequencies, or are heterogeneous, a gateway is appropriate for connecting the two devices. The processors also

8
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 10. Exploded view of Sensor Type 2.

Fig. 11. Exploded view of Sensor Type 3.

Fig. 12. Wearable Sensors used in RPMS.

used in this sort of sensor to transfer the obtained data to the hospital through WWAN. As shown in Fig. 8, the third type
of sensor combines an internal sensor and a processing unit. A variety of factors influence the selection of a sensor in a
remote patient monitoring system. The first is the observed measurand, which specifies the location in or around the body
as well as the type of sensor to be employed. The second factor is the patient’s gender and age.
The environment in which the patient will be monitored also influences the type of sensor utilized. For example, if the
patient will be observed while mobile, a tiny and comfortable sensor will be used, whereas wired sensors can be used when
viewed inside a house. According to the authors of [16], in order for these sensors to be durable and comfortable to use,
they must meet medical and ergonomic requirements. Sensors must be tiny, flexible, non-toxic, and hypoallergenic to the
human body. The sensors encounter issues such as high power consumption owing to sensing, processing, and transmitting,
as well as quality of service, scalability, and energy efficiency.

Type of sensors in RPMS

According to the authors of [85,87,88], the sensors can be either contact-based or contactless to the patient’s body. Body
contact-based sensors are wearable or implantable sensors that are put on the surface of the human body. An example of a
wearable sensor is a flexible sensor that can be attached to clothing, an elastic band, or directly on the body, and integrate
electronics or computing devices to provide smart functions, as shown in the diagram Fig. 12 below. Fig. 13 depicts the
surgical technique required to integrate the implanted sensors into the patient’s organs. These sensors can be situated in
several locations throughout the human body, or they can be a single sensor that obtains many parameters. Body contactless
sensors are sensors that are put around the patient but are not attached to the patient.
They often collect patient data via infrared, RFID, or optical. Because of the ambient parameter, these types of sensors
are typically less precise than contact-based sensors. The examples are shown in Fig. 14. Depending on which physiological
parameter is being evaluated, contact-based or contactless sensors can be employed. The physiological parameters are mea-

9
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 13. Implantable Sensors in RPMS.

Fig. 14. Contactless Sensors.

sured with a different type of device in different postures, and their normal and pathological values differ. Wearable sensors
can now be found in watches, clothing, finger rings, eye glasses, ear buds, and even socks [85]. Implantable sensors, on the
other hand, can be placed inside the heart, brain, stomach, or fingertip to monitor these conditions.

Summary of existing architectures of RPMS

This section provides a summary of the present Remote Patient Monitoring architecture, focusing on the system layers,
sensor type (contact or contactless), mobile control unit employed, and system placement.
In a study publication [41], an electronic sensor system based on rule-based (IF-THEN rules) is presented that can detect
falls without requiring a remote connection to a database. It is a three-layered system, with the first layer being the data
layer, which performs data collecting and pre-processing. The second layer is the decisional layer, which analyses decisions
using a rule-based manner. This action layer is unique in two ways: 1) the rules are extracted automatically, and 2) they are
extracted and processed by the same application layer.
A study paper [42] proposes an oral device that uses a System-on-a-Chip (SoC) to monitor sleep and can autonomously
acquire physiological information about tongue motions throughout the therapy process. A tunnelling sensor array with a
high degree of sensitivity accurately detects tongue pressure. When the gadget detects a wireless network, it will remotely
grab temporary data and wirelessly transmit it to the user’s computer and cloud storage. External RF power will be gathered
via the platform to recharge the battery. To show continuous tongue movement detection, a small prototype module with
dimensions of 4.5 × 2.5 × 0.9 cm3 is implemented and implanted in the oral appliance. The proposed design is feasible
and the design can make obstructive sleep apnoea (OSA) therapy more efficient and thorough. Portable, light, low-cost,
and interactive bio-sensor nodes have become increasingly possible as sensors, wireless communication, and low-power
integrated circuits have advanced in recent years. These nodes can be used to establish a wireless body area network to
monitor, detect, process, and communicate crucial parameters throughout the mobile body.
The researchers of study [43] proposed a RPMS, based on the biomedical and environmental data collected by deployed
sensors. The mobile device is used to transmit the patient’s health parameters (such as temperature, heart rate, and ECG) to
a medical server, caregiver, and medical practitioner. This system is a three tier based architecture where the first tier consist
of the wearable body sensor that acquire the physiological parameters. The prepossessing is performed in a microcontroller
before being forwarded to an android mobile phone in tier two which transmit the data to a medical server using GPRS.
The Medical server performs in the last tier. This is where decisions are made and emergency measures will be taken.
As opposed to legacy systems that monitor only one or two physiological parameters, this system incorporates multiple
physiological parameters. This paper emphasizes authentication, power consumption, and accuracy in transferring health
data to a medical server as well as hardware, software, and system implementation in terms of hardware, software, and
computer implementation. An automated system will improve the usability of the system.

10
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

The study [27] created a COVID-19 non-contact screening method. The researchers used ultra-wideband impulse-radio
radar to identify physiological characteristics such as respiration, heart rate, body movements, and sleep quality in their
study. A total of 140 radar monitoring data from 23 COVID-19 patients were compared to 144 data from 144 healthy controls
at Wuhan Tongji Hospital. The XGBoost algorithm and logistic regression (XGBoost + LR) approach were then used to classify
the data into sick and healthy people. Environmental and other factors could have an impact on the controlled experiments.
Furthermore, because the study only covered one age group and did not include numerous age groups, the accuracy index in
practice was lower. Monitoring more than one patient at a time may be impossible in a real-world hospital setting, limiting
the efficiency of monitoring with the proposed technique and system.
Nemati et al. [44] created a four-tiered architectural remote patient monitoring system. The first tier includes the ECG
electrode sensors, while the second tier handles signal processing. The third tier involves wireless communication, and the
fourth layer is for decision making, in which a caregiver uses a personal computer. A stretchable belt with three capaci-
tive electrodes implanted in a cotton T-shirt is used in this system. The authors integrated their wireless ECG monitoring
technology, which can be used to perform ECG measures over the cloth. Cotton, which served as a dielectric substance, was
placed diametrically opposite the electrode. The signal processing and communication units were housed on a tiny two-layer
PCB board. To reduce power consumption, a low-power electronic system, wireless connection using the ANT protocol, and
signal sampling in the idle mode are used. Users may find it challenging to work with hard electrodes, which may result in
motion artifacts in the signal.
The authors of [45] devised a concept for remote patient monitoring. The system’s components include pulse oximeters
and temperature sensors. The system is controlled using the Arduino UNO microcontroller. The graphical user interface
program connects the computer to the sensor circuit. The end user (caregiver) can retrieve patient data via an easy-to-
use web interface to calculate blood oxygen saturation and temperature values using Arduino programming. The Android
system is simple to use and efficiently monitors the patient. Nonetheless, future studies should include more physiological
characteristics. Patients’ medical histories should be stored on a local server for quick analysis and any medical procedures.
It is possible to develop an implantable wireless and battery-free bladder pressure monitor system that measures bladder
storage in real time by implanting a compact packed sensor that provides its feedback signal to the external receiver via
BLE (Bluetooth Low Energy).
The study’s [47] researchers created a self-powered implant for monitoring blood pressure. Using a thin film piezoelectric
sensor, a model of aorta wall stress and produced electrical potential related to systolic blood pressure was obtained. During
in vitro testing, a solid linear relationship was seen between the device’s peak output voltage and flow pressure. Researchers
achieved a high sensitivity of 173 mV/mmHg, which outperforms prior results. A maximum instantaneous power of 2.3 W
was observed in vitro, showing an efficient self-powered system. Furthermore, after more than 50,0 0 0 cycles of operation,
the gadget displayed exceptional stability. If hypertension is detected without a built-in battery, the device will sound an
alarm. With this novel method, the implantable healthcare monitoring field offers promise to the RPMS research. Testing
and validation of the system will benefit also benefit health sector enormously. Examining the sensor placement is quite
intriguing.
A person may worry whether installing a specific sensor at a specific location will improve the system’s results. Wireless
sensors are critical for biomedical devices, and a recent research explored the ideal placement of these sensors on the
human body. The issue of fall detection is explicitly addressed in study [48]. The sensors were inserted at the following
places during testing: the right ankle and thigh, the chest, and the waist. When body position is taken into consideration,
the chest and hip numerical accuracy increase to 51% and 51%, respectively. According to their findings, an accelerometer
worn at the waist performs better as well as a chest-worn accelerometer.

Wireless communication used in remote patient monitoring

The use of wireless communications to improve the delivery of health services has been reviewed in several studies.
Zigbee, Wi-Fi, Bluetooth, Bluetooth Low Energy (BLE), LORA, ANT, and UWB technologies tend to be used as short-range
communication technologies in remote patient monitoring systems, as reported by the authors of [49,89–92]. These tech-
nologies are normally used in conjunction with long-range communication technologies such as satellites, cellular networks,
ad-hoc networks and WIFI. In this study, we will discuss Wi-Fi, Zigbee, Bluetooth and Bluetooth Low Energy (BLE) under
short-range communication technologies in addition to cellular networks and Ad-hoc Networks under long-range commu-
nication technologies. The summary of the explained wireless communication is shown on Table 3.

Bluetooth based RPMS

A Bluetooth network uses an open protocol that complies with IEEE 802.15.1 for wireless communication as well de-
scribed in [89]. They further explained that in WPAN, it can be used to connect many different devices, both for data and
voice transmission as part of a short-range network topology, known as a piconet. Bluetooth devices or nodes can range
between two and eight. These authors defined piconets as synchronized networks composed of Bluetooth devices acting
as the master, while all the rest are slaves to the master as shown in Fig. 15. Ref. [93] also explained Bluetooth network
configuration as piconet and scatternet. They explained that piconet consists of one master node with seven slaves. If the
Bluetooth device is connected in different piconets it is normally referred to as scatternet as illustrated in Fig. 12. With

11
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Table 3
Summary of the wireless communication in RPMS.

Energy Network
Frequency Consumption Data Rates Supported Notes Topology
Wireless Comm. Tech. Band (GHz) Range (m) (mW) (Mbps) (Max) Supported Refs.

Bluetooth (IEEE 802.15.1) 2.4→ 2.5 1→100 2.5→100 1 →3 1 master with 7 Star, P2P [49,89–92]
slaves
Bluetooth Low Energy 2.4→ 2.5 1→100 10 1 1 master with 7 Star, P2P [49,89–92,94]
slaves
Zigbee (IEEE 802.15.4) 2.4→ 2.5 10→ 100 35 250 65,533 Tree, P2P, [25,49,89–92]
mesh and
star
Wi-Fi (IEEE 802.11) 2.4→ 2.5 150→ 200 1000 54 255 Star, P2P [31,49,89–92]

Fig. 15. Bluetooth Network Topology.

Bluetooth, time slots are distributed alternately between Master and Slave through a Master-driven TDD (Time Division Du-
plex) scheme. Slave packets can only be sent to the Master in odd-numbered slots immediately after being received from
the Master, whereas the Master can only be sent packets in even-numbered slots [1].
Bluetooth is attractive because it is capable of connecting and communicating with an almost limitless range of Bluetooth
enabled devices as explained by [94]. In addition, devices connected by line-of-sight can communicate without requiring the
device to be in line of sight of the other device. In this way, it can be used to connect a variety of personal devices in order
to support both data and voice. For interference-free operation, Bluetooth units operate in the 2.4 GHz ISM band (Industrial,
Scientific, and Medical band). They employ frequency hopping within 79 1 MHz channels at a rate of 1600 hops/sec to
achieve their reduction in power consumption. According to the standard, there are three kinds of devices, each with its
own transmit power and coverage range from 1 to 100 m. It offers up to 3 Mbps data rates.
Bluetooth was examined as one popular technology used in RPMS by Ref. [49]. The experts also argued the importance
of Bluetooth integration into RPMS for analysing vital signs.
According to Ref. [50], wireless transmission of vital health information between a mobile phone and a personal com-
puter is possible with a low-cost, secure device. They measured blood oxygen saturation and heart rate. During emergencies,
the mobile phone transmits real-time, secure data to a server that processes the information. To ensure security, both the
sender and the receiver encrypt and decode. Integrity and confidentiality of messages were ensured using Hash Message Au-
thentication Codes (HMACs) and symmetric encryption authentication methods. In another article, De et al. [51] described a
system for recognizing complex daily activities using multiple sensors. An accelerometer and a gyroscope were employed to
measure activity; humidity, temperature, and barometric pressure sensors to gauge the environment; and a Bluetooth beacon
to determine location. A portable ubiquitous-Vital (u-Vital) system consisting of a Vital Block (VB), a Performance Monitoring
Device (PMD), and a Vital Sign Server (VSS) is developed by researchers, [52], allowing for the monitoring and storage of vi-
tal signs. The VBs collect vital signs such as electrocardiograms (ECGs), blood oxygen saturations (SpO2), non-invasive blood
pressures (NiBPs), body temperature (BT) are collected live and transmitted to a VSS through wireless protocols including
Bluetooth and Wi-Fi3.

Bluetooth low energy based RPMS

According to [89] BLE technology provides ultra-low power consumption and cost, allowing it to be deployed in local
area networks. The device facilitates interoperability between portable consumer devices. The technology also allows for
the transmission of very short data packets, as well as short connections for the most efficient transfer of data. Wireless
connectivity of small devices to mobile terminals is possible with Bluetooth LE. The sensors are thin enough that they can

12
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 16. ZigBee Network Topologies.

not only run on a small amount of power but are also inexpensive and can be used for health monitoring applications.
A data rate of up to 1 Mbps is expected with Bluetooth Low Energy technology. As Bluetooth uses only a few channels,
synchronization occurs in milliseconds rather than seconds. An RPMS application like alarm generation or an emergency
response would benefit from this, as they need to happen quickly [94].
Advances in Bluetooth technology like BLE offer tremendous possibilities in RPMS software platforms that require less
power as described by writers of [49]. The writers also argued that BLE consumes less power compared to Wi-Fi and Zigbee.
Remote Patient monitoring has been able to make use of this technology especially on wearable kinds of RPMS or WBAN
based RPMS. This technology transmits the data from the sensor to a mobile device or microcontroller.
A study by [53] designed an Empatica E3 which is a wearable wireless multisensory device that provides data collection
and biofeedback in real-time. The device is light, compact, comfortable, and can be used for almost any real-life application.
In addition to performing real-time data processing via Bluetooth low energy, the E3 also records data to its internal storage.
E3 allows for ambulatory data acquisition in a comfortable and distraction-free way outside the lab.

Zigbee based RPMS

A similar wireless technology called Zigbee finds application in today’s remote patient monitoring systems to review [49].
Zigbee, or IEEE 802.15.4, is a high-layer communication standard that uses WPANs for short-range wireless communication.
This technology developed by the Zigbee Alliance is effective and eco-friendly compared to other wireless monitoring tech-
nologies. It is fully suited for RPMS purposes, where low-cost, low-power, and long-lasting battery power is critical. It is
more affordable and simpler than other WPANs, like Bluetooth and Wi-Fi. IEEE 802.15.4 operates with data rates ranging
from 20 Kbps to 250 Kbps. It is possible to set up star, mesh and cluster tree network topologies when using Zigbee in
RPMS as it appear in Fig. 16. Zigbee technology (IEEE 802.15.4) is used in Ref. [54] to develop and demonstrate a reliable,
cheap, low powered and accurate RPMS that is wearable. During a medical procedure, the system measures a patient’s heart
rate, temperature, and detects distress by plotting the measured physiological parameters using receiver units. The system
presented in [95] integrates wireless sensor networks into a continuous patient monitoring network that sends data to a
centralized ARM server. Every patient monitoring system is connected to a Zigbee node that is small, consumes low power,
and consumes little space. A common database stores the patient’s information, which physicians can access from their
mobile devices.

WiFi based RPMS

Wi-Fi (wireless fidelity) is a popular wireless technology for transferring data and transferring information. A local
area network (LAN) uses the Ethernet standard IEEE 802.11 to enable wireless communication according to the authors of
[89].These authors further explained the main advantage of Wi-Fi is its high transmission speed. Throughout the network,
devices are interoperable using stars and point-to-points. A Wi-Fi network consists of several electronic devices that can
connect to the local network or the Internet by using an access point with an average distance of 100 metres. According to
the IEEE standard, you can access the Internet at speeds as fast as 54 Mbps. However, power consumption is relatively high
with this technology. Sensors and devices in RPMS that can be continuously monitored by Wi-Fi represent a good candidate.
These include a Wi-Fi-based system implemented by authors used for monitoring blood oxygen saturation and heart rate.

Cellular network based RPMS

If such application needs data to be transmitted over a longer distance with little latency, Wi-Fi and cellular networks
are the best solutions. The gadget has greater transmission, processing, and storage capacity than shorter communication
types. During data delivery, these communication models send multimedia data such as speech and video. [17]. Researches

13
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Table 4
Features of WSN and WBAN in RPMS.

Characteristics WSN- RPMS WBAN-RPMS

Node Type Homogeneous Heterogeneous

Coverage Area From a few m→km From a few cm→m
Number of Nodes 100→1000 10→100
Node Task Dedicated or Single Multiple
End-to-end-delay Higher Low and critical
Node replacement Easy Difficult
Data loss Accepted Not Accepted
Node Lifetime Long Long with small batteries

in [96] have introduced RPMS using 5 G network. The combination of various technologies such as computers, medicine,
and telecommunications will greatly increase patient efficiency.

RPMS based on wireless sensor network (WSN) and wireless body area network (WBAN)
WSNs consist of large-scale, highly distributed and self-organized networks, mainly developed for surveillance and mon-
itoring. Their components consist of a large number of cheap and tiny nodes that are capable of computing, sensing, and
communicating [97]. Originally, WSNs were meant for surveillance purposes on large battlefields. During the development
of these applications, WSNs were supposed to have millimetre-sized computers called smart dust. The concept has not yet
been realized, to the best of our knowledge as defined in [98]. Even so, this technology paves the way for a host of new
opportunities. So WSNs will be used in a broad range of applications, such as monitoring environmental conditions, observ-
ing animal migrations, monitoring security, monitoring industrial processes, monitoring personal healthcare, etc. [99]. The
healthcare industry uses a completely different type of WSN than other industries. The WSN in healthcare must cover five
core criteria: data security, energy efficiency, multicasting, mobility support, and real-time data dissemination. It is expected
that sensors will perform without maintenance and that data will be collected depending on their location. To achieve the
best performance, low costs, and application to the healthcare network system, many researchers have focused on routing
protocols. In the field of medicine, wireless technology is being developed for integrating larger, more expensive medical
equipment with other machines and procedures needed to be used. One such technology is Wireless Body Area Network
[100].
Recent developments have shown that Wireless Body Area Networks (WBANs) offer a viable alternative to remote pa-
tient monitoring systems [101,102]. WBAN based RPMS involves a set of tenable, heterogeneous sensor nodes that can be
implanted or attached to the body to measure physiological data. The coordinators, such as PADs or smartphones, gather and
deliver the information to remote centres, such as hospitals, over wireless connections. The mobile nature of users makes
WBANs based RPMS different from WSN based RPMS. Furthermore, WBAN based RPMS consist of three main communication
channels: intra-BAN communication, inter-BAN communication, and beyond-BAN communication. Intra-BAN communication
is the communication between wireless sensors attached to a patient’s body and the master node of the WBAN. A commu-
nications inter-BAN is a communication between a master node and a network of personal devices such as laptops, mobile
phones and so on.
Both WBAN and WSN RPMS have sensor nodes, but the number of nodes in WSN is larger than that of the WBAN.
Furthermore, the WSN’s coverage area is larger since the nodes in WBAN are implanted close to the body. Both networks
have slightly different data rates and node tasks as shown in the Table 4 [99].
According to authors in [103], WBAN based RPMS, reliability and efficiency are determined by how fast and accurate the
system is in sending and receiving data between nodes, which ultimately depends on the routing protocols or algorithms
used. An in-body sensor node or on-body sensor transceiver emits radiation similar to those emitted by WSNs to transmit
data. Although WSN routing protocols have been the understudy for the past few years, they cannot be used for WBANs
due to their strict requirements. When it comes to WSNs, the main focus is on minimizing routing overhead and maxi-
mizing throughput rather than reducing energy consumption. In addition, while WSNs are mostly homogeneous, WBANs
are heterogeneous as well according to reference [104]. Homogenous networks have the same functional capabilities they
can acquire process and transmit the same data. In heterogeneous networks, the sensors nodes have different functional
capabilities in the sense that data traffic is different, with different data rates and different energy consumption levels.

Remote patient monitoring system as a node

There are numerous requirements to be considered when implementing a remote patient monitoring system as a node.
These may comprise the requirements on the type of sensor device used and its applications or the location of implementa-
tion and its surroundings. These requirements are important in defining the quality of data, routing protocols and the com-
munication networks to be deployed. However, the end-to-end delivery of vital sign data depends on the communication
network. Therefore requirements of the network determine the implementation of a RPMS [100]. However, the increasing
number of things connected to the internet and rapid changes in network communications have increased the number of
RPMS nodes and conveyed new dynamics of the RPMS operating environment. These have also brought unique challenges

14
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

to the architectural design of RPMS. Therefore new solutions are needed to overcome some of these challenges which may
include but are not limited to Quality of service, node mobility, Real-time data delivery, operational cost management, secu-
rity, latency, fault tolerance, scalability, and IEEE standardizations, interoperability, power management, automated systems.
The requirements are outlined and defined based on the network communication as well-defined by the authors in [82].

Characteristics of RPMS
Quality of service (QoS). QoS is generally regarded as the performance of a network communication path’s data delivery
service. This performance is measured in terms of quality of service characteristics. QoS can be measured according to the
throughput, bitrate, packet loss, jitter, and reliability in the network communication. Therefore in RPMS, the overall QoS is
determined by the network communication between the Body Area Network (BAN) subsystems, Personal Area Networks,
Local Area Network and Wide Area Network (WAN) for continuous delivery of data [12].

Node location. The location where the RPMS is deployed and used is important during data acquisition and delivery. The
node in rural areas will face challenges in network connectivity compared to the ones in the urban areas. The mobile node
will also pose challenges of changing results due to variation in speed and network work connectivity when compared to a
static node. The node in crowded traffic areas will face delays in data deliveries compared to the node in fewer traffic areas.

Real time delivery. This is the ability to deliver critical data at the speculated time without any delays. In RPMS timing is
important to reduce any complications in patients during emergencies.

Fault tolerance. Redundant connections allow alternative paths in case one link fails. The communication between patients
and doctors or caregivers should always be there to avoid end-to-end delays.

Power management. Battery life should be considered both in the sensor applied and the mobile devices used. WBAN are
more reliable in energy usage compared to WSN.

Reliable communication. It is the capability of sending and receiving different forms of data at a very low bit error rate. It is
important in the RPMS network, especially during emergencies. This aspect is influenced by the routing protocols therefore
choosing an appropriate routing protocol is significant.

Device mobility management. The communication network should have the ability to adapt to any changes in the link quality
as patients and their caregivers are always on mobile. Therefore the routing protocol should be able to distribute and balance
the data accordingly on the network regardless of the speed the patient or relay node moves at.

Latency. It determines the end-to-end delay in the communication network. It measures the amount of time taken to send
data from a sink to a source and vice versa. In RPMS delays are not tolerated as human lives are considered and the feedback
should be given immediately. It is influenced by load distribution and balancing.

Scalability. This is the ability of the network algorithm to adapt and respond accordingly in case of changes in network size
or events around it. In RPMS patient can be monitored when in transit, in vehicles, at home and in shopping centres in a
town or rural area. Therefore these locations can influence a change in the number of nodes in a network and even the
events around those nodes.

Security. Safety and privacy are important when dealing with patients data. The Health Insurance Portability and Account-
ability Act of 1996 (HIPPA) was created mainly to improve the flow of healthcare data and address policies of how personal
information of the patients in insurances and medical facilities should be handled to protect frauds and theft. There is also
some existing research that addresses security in RPMS.

IEEE standards. Most sensor nodes normally use the (Industry Scientific and Medical) ISM band, (Ultra Wide Band) UWB
band, and Zigbee according to the IEEE 802.15 standard. WSNs tend to use license-free communication frequencies: 173,
433, 868, 915 MHz and 2.4 GHz [9]. Authors in the study [100] noted that RPMS, can use IEEE 802.15.4 standards for the
source node (patient) and IEEE 802.15.6 standards can be applied to the sink node (hospital).

15
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Table 5
Overview of data rate and time of arrival for the physiological parameters.

Physiological Parameters Parameter range Data Rate (kbs) Data Arrival Time (sec)

ECG 0.5 - 4 mv 6.0 0.002

Blood Pressure 10 - 400 mmHg 6.8 - 7.8 1.2 0.01
Body Temperature 32 - 40°C 0.0024 5
Oxygen Saturation (SpO2) 0.01 - 0.85/s 0.16 0.16
Respiratory Rate 2 - 50 breaths/min 0.24 0.24

Table 6
Priority weights.

Physiological Parameters Data rate Latency Priority

ECG High Low Level 0

Blood Pressure Low Low Level 1
Body Temperature Low High Level 2
Oxygen Saturation (SpO2) High High Level 3

QoS in RPMS
QoS is a loaded term with different definitions and opinions as it can be characterized in a vast number of applications.
Even though it may be self-explanatory, there is little consent on the exact and standard definition of Quality of Service
especially in the context of the remote patient monitoring system. Therefore people and communities observe and under-
stand QoS differently and these make it difficult to evaluate. Recommendation E.800 of the International Telecommunication
Union (ITU) defined QoS as an overall feature of a telecommunication service to provide and satisfy the needs of its users of
the service. This definition shows that communication networks used to relay the medical data are the basis for the quality
of patient health during remote monitoring.
QoS can also mean the ability of a communication network to deliver data reliably and efficiently according to authors
in [105]. In generally more reviews evaluate QoS in e-Health services compared to QoS in RPMS. Most authors argue that
QoS should be addressed in Tier 1 as compared to Tier 2 and Tier 3 due to its importance in data acquisition and data
transmission. The highest priority in data transmission of critical data is important in RPMS [82]. The success rate also is
one important aspect of RPMS because RPMS are used in medical applications, their inability to transmit data accurately
in real-time poses a high requirement for throughput. Even so, in general, co-located RPMS share a limited number of
channels. To mitigate inter-and intra-RPMS interference, the data rate for sensors must be adjusted reasonably to improve
system quality of service [73]. For example, a patient monitored for the cardiac disease which involves acquiring heart
activity is more important than a body temperature monitored patient. Also depending on the patient’s clinical history, the
priority may change due to the quick deterioration.

QoS requirements. To provide QoS in RPMS, all three tiers should be considered in terms of sensor data, the computation and
communication involved in relaying the patient’s data. The QoS requirements may be also addressed following the protocol
layers of each tier, therefore, making it more complex to assess. Mostly the patients are not aware of the QoS provided
by the medical services; it is upon the service provider to define different data and their priority on the communication
network. Therefore certain QoS metrics are used to define the requirements of QoS in RPMS. The below QoS requirements
are defined based on the type of traffic, quality of data and the resource;
(a) Type of traffic
Nowadays RPMS involves multimedia which mainly include text, graphics, audio and video [106]. Most of these data is
in the form of analogue signals during acquisitions, therefore an Analogue to Digital Converter (ADC) is needed for convert-
ing this data to digital format. Physiological parameters in RPMS have different priorities or data rates therefore they are
sampled differently and have different acceptances of delays [83]. Table 5 give an overview of data rate and their time of
arrival according to authors in [83].
Classification of priority level may be assigned according to the data rates and latency. According to Abidoye etal. low
latency means a quick response to a critical signal transfer and should ideally be as short as possible while high data rates
indicate critical signals that must be transmitted very fast and with high reliability. The priority weights of the physiological
parameter is classified as High data rate and low latency, low data rate and high latency, low data rate and low latency and
both high in data rate and latency. The Table 6 show the priority weight of the five physiological parameters.
(a) Quality of data

Accuracy: This is the closeness of the measured value compared to the expected value. It is normally related to sensor
measurement. In RPMS, the sensor nodes must provide accurate data to the applications or the caregivers for ap-
propriate decision making. Therefore gathering sensor measurements as close to the point of the subject improves
accuracy. Accuracy cannot be directly calculated but can be calculated based on the errors.

16
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 17. Individual and Group based RPMS.

Reliability: This is a system’s ability to remain connected even after certain network failures. It is dependant on BAN
sensor readings and delivery in RPMS. The communication dependability of the communication routes, protocols, and
technologies used in medical data transmission is equally important. All RPMS tiers are accountable for ensuring
patient monitoring reliability.
Data Timeliness: Out-of-date data is useless in a real-time application and can harm the system. In RPMS, QoS needs
may be assessed based on real-time and non-real-time applications. Timeliness needs in real-time applications can be
further classified as hard or soft real-time. The system should ensure a predictable end-to-end latency in hard real-
time and is delay intolerant. Soft real-time systems ensure probabilistic end-to-end latency and can tolerate minor
delays. RPMS should offer either deterministic or probabilistic end-to-end latency to facilitate data timeliness in real-
time applications. The technologies and equipment used in the tiers of RPMS, as well as their protocol layers, are to
blame for data transmission delays.
Universal Access: The medical providers and caregivers should be able to access the patient’s information anytime and
everywhere. Restricting access to RPMS resources could make the system to be inflexible. All the tiers devices, proto-
cols and technologies are responsible for universal access in RPMS‘s resources.
Interpretability: Refers to how easy the user can understand the data. Any uncertainty in understanding the data may
delay some feedbacks or even cause some wrong treatment. All the devices in the three tiers are responsible for
processing and interpreting of these collected data.

Individual and group based RPMS

The RPMS can be classified as either individual or group-based. Individual RPMSs use short-range communication, such
as Bluetooth or Low-power Bluetooth, to connect the body nodes to a personal device. The technology collects medical
data at the nearest access point, then forwards it to the hospital for decisions and feedback. With a personal device, an
ambulance can track the location of a patient in case of emergency. It is a one-way communication method. A group-
based RPMS, by contrast, focuses primarily on communication between individual RPMS. Health monitoring systems can be
organized centrally, decentralized, or clustered depending on the overall connections of each individual. A coordinator is
responsible for gathering and transferring data for each individual in a centralized structure. Individuals in the distributed
structure are self-organized and communicate through a coordinator. When clustered structures are used, it is caused by
a group of individuals joining together and selecting a leader for the cluster. Each cluster is connected by the coordinator.
The group-based RPMS is shown in Fig. 17. A further overview of how RPMS nodes behave in an individual and a group
based is explained. Firstly a general communication including some technologies and routing and MAC protocols in both the
individual and group-based RPMS will be shown with reference given in the base of using WSN, WBAN and Heterogeneous
networks.

Individual based RPMS. Given two types of nodes, being nD and nSi , where nD represent the root node (destination) located
in the centre, and nSi represent any source node that forwards their packets with a single hop communication to node
nD . The node nSi is deployed at a distance di,0 from the root nodenD . Then we can show a single-hop communication or
individual based RPMS as shown in Fig. 14. Recall that in this kind of communication we can have four scenarios, where;

1. Both nSi and nD are static

2. nD is static and nSi is mobile
3. nD is mobile nSi is static
4. Both nD and nSi are mobile

Then di,0 which can also define the link coverage or the distance between these two nodes will change in case one of
the nodes is mobile. The network lifetime will be reduced if the destination node is far from the source node because it
will take more energy to send a single packet of data and avoid losing it along the long transmission path. Battery drains
at a fast rate as the distance di,0 between receiving and sending nodes increases, which results in a reduction of a lifetime

17
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 18. Markov Model for patient’s posture [57].

Fig. 19. Markov model for walking into three posture state [58].

[75]. The other factors like link quality, delays, energy consumption can also affect the communication between the two
nodes. When single-hop communication is used, information is directly forwarded from the sensor nodes to the destination
node. Different technologies and applications have been implemented in individual-based RPMS nodes which can increase
the lifetime of nodes because the nodes on the human body can form a network and perform energy-saving technologies
[19].
The big data produced by these systems have made researchers bring forth technologies that can improve the communi-
cation between these nodes.

a Markov Model for Individual-based RPMS

A Markov model describes the stationary stochastic process and can easily be used in Individual-based RPMS to describe
the user’s postures such as standing, sitting, lying down, walking, and running. The authors of [57] modelled postures ac-
cording to a Markov chain, as shown in Fig. 18. According to Sharma et al., as the human body is mobile, it makes the
distance between the sink and the nodes in RPMS to change. Therefore the nodes selected are based on the probability of
posture change within the RPMS network. They assumed that if the sink node is located in the chest of the patient, and
then during movement the distance between the sink and the nodes would be changed. Consequently, the changes affect
energy consumption, throughput, and network lifetime. To correct these flaws, the researchers developed a mobility model
for multi-hop and forwarder based routing techniques. This model only focuses on the sensor node network but does not
show the interaction in the whole RPMS framework.
Based on the study performed by Liu et al., they divided walking into three posture states, as shown in Fig. 19, by using
a Markov model to describe a mobility model suitable for RPMS nodes [58]. Researchers in [59] used the Markov model to
determine whether RPMS are available for a virtual grid in this paper. Energy is a constraint on wireless sensor networks,
so their energy use should be limited as a way to ensure their quality of service. This paper investigates how an energy-
optimized algorithm and Markov processes can be used to improve availability and service quality in RPMS. Markov model
is used in a vast number of aspects concerning wireless communication in RPMS. Other researchers have applied this model
in the stochastic channel model in RPMS. The researcher in [60] described six states of the channel’s dynamic behaviour.
They then developed a Semi-Markov model that measured both the LOS (Line of Sight) and non-LOS (Non-Line of Sight)
scenarios. Based upon Ref. [31], the dynamical sensors schedule was constrained by Markov decision processing.

a Mobility Model For Individual-Based RPMS

Individualized RPMS have the advantage of not limiting human movements, e.g. they allow users to walk at their leisure.
Despite this, the motion mode can have a significant impact on network performance. RPMS mobility model is currently
being studied by some researchers as a way to mitigate problems faced by these systems. In most mobility studies, the
movement of individuals has been described in terms of single node mobility. Models such as Random Walks, Random
Waypoints, and Random Directions are well-known and cited models that do not utilize memory [107]. Based on a case
study in Ref. [61], it was examined how average mobility of a patient with RPMS would be in a hospital room. Ref. [62] pro-
poses a mobility model where the user chooses a random room, walks to the room at a random speed, and stays there for a
specified period. Interestingly, patients do not always know where they wish to go, according to [63]. Therefore, the authors
restricted the activities of patients on a single floor, and they then applied a model for nomadic mobility.

18
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

a Mac Protocol For Individual-Based RPMS

MAC protocol with multi-level access was proposed [108] for enhancing the Quality of Service (QoS) of the RPMS on
the user, data, and time levels. The quality of service can be improved by differentiating data and user priorities. A unique
solution can be used to optimize QoS and optimize energy consumption. The new MAC protocol proposed by [64] solves
the energy and emergency data problems by using IEEE 802.15.4. Various methods were used, depending on the type of
data. They reduced the transmission delay by sending an emergency beacon message to enable emergency data transmis-
sion. Then they allocated a fixed amount of bandwidth for periodic data. Standard data was transmitted using a slotted
CSMA/CA mechanism. The similarities of MAC protocols for WSN based and WBAN based RPMS have been studied by many
researchers (e.g., a WBAN is a type of WSN). There are two main types of MAC protocols: contention-based and schedule-
based. Scheduling-based protocols, like Time Division Multiple Access (TDMA), have higher energy efficiency and quicker
response than contention-based MAC protocols (e.g., Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA)). Scala-
bility and synchronization remain problems for schedule-based protocols [65]. Based on IEEE 802.15.6, Yuan et al. proposed
a new MAC protocol that distinguishes between different traffic types and users’ priorities, which significantly reduced the
network’s power consumption and extended its lifetime [66].

a Routing Protocols In Individual-Based RPMS

Addressing, routing, and forwarding data packets is the responsibility of a network layer in a WBAN based RPMS. The
implementation of QoS-aware routing protocols is very important in these systems so that packets are delivered on time
and reliably (e.g., doctors receive critical information from patients) as presented by authors from article [109]. It is not
trivial to develop a routing protocol that is efficient, including forwarding data, and QoS-aware in RPMS. There is a dynamic
environment on and in the human body and outside the human body due to wireless technologies [65]. The standard routing
protocol in ordinary WSNs was studied and improved to correspond to the RPMS [97,99]. To resolve the survival time of the
RPMS nodes, a uniform clustering algorithm-based routing structure has been invented [76]. Compared to existing wireless
body sensor networks, this method is environmentally friendly and has a greater lifetime.
Several studies have suggested adaptive greedy buffer allocation algorithms [110]. Due to its residual energy and long
lifetime, this algorithm can produce more residual energy and lifetime at the node, meaning the sensor can operate under
more challenging conditions, including mines, tunnels, and holes. According to [67], a cost-based energy-efficient routing
protocol is designed that eliminates thermal effects and delivers excellent performance. Using this method, a wearable de-
vice in RPMS can improve its energy efficiency under most interference conditions. In the study by Abbaneh et al., an
energy-balanced rate allocation protocol and routing method were developed to improve energy consumption, network life-
time and utility. This protocol transmits data via a link based on the link utilization to reduce the load on the nodes, and
data rates are determined intelligently according to a designed utility.

Group based RPMS. Healthcare data can be shared between RPMS based on groups and RPMS based on individuals. In a
group health monitoring system, individuals can transmit each other’s data. Thus, multiple individuals form a multi-hop
network. If two individuals are close to each other, their sensors can interfere. Based on time-sharing, channel switching,
and CSMA/CA technologies, Alma et al. proposed three coexistence strategies to reduce channel interference [111]. Managing
network throughput can also be difficult in a large network. The cross-layer carrier sensing scheme was proposed by Shimly
et al. to provide high reliability and efficiency for data transmission with high throughput was designed so that channel
state information could be passed from the physical layer to the network layer of RPMS. In RPMS networks with many users
moving freely, routing can be difficult, especially when it is stable and reliable [112]. Unfortunately, currently, there have
been no routing standards developed for RPMS. Thus, the study examines the current trends for current universal routing
protocols, as well as a hybrid adaptation routing protocol in a similar mobile ad hoc network [68], and then adds human
mobility characteristics to these protocols.
To facilitate this proposed system, it is necessary to use some key technologies outlined below which are obtained from
existing literature.

a Game Theory In Group-Based RPMS

Game theory is an efficient mathematic tool to model the interaction amongst independent decision-makers, which is ap-
propriate to be implemented in RPMSs competing for limited resources selfishly. Several game theory model-based schemes
have been studied to resolve the interference amongst RPMS, especially power control schemes that can mitigate interfer-
ence and decrease energy consumption effectively [73]. Multi-user wireless networks compete for a finite range of spectrum
and channels, resulting in interference between users. All users strive to achieve high utility levels during the competition
process. So, game theory can help allocate resources and share channels effectively in RPMS [113]. To avoid socially aware
interference, Liu et al. proposed a game-theoretic method to allocate Wi-Fi channels to individuals and Zigbee channels to
human nodes for a two-stage channel allocation scheme as shown in Fig. 20.
In a study by Kazemi et al., there is a link between interference and increased power consumption. The authors used a
pricing mechanism to minimize inter-human interference between close individuals [78]. To meet the large data access and
tailored service needs for many Quality of Service (QoS) factors of RPMS, efficient compute offloading techniques are more
necessary. In this research [114], they presented a Two-Stage Potential Game Based Computation Offloading Strategy (TPOS)

19
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 20. WBAN interfering with each other [77].

Fig. 21. The network model of TPOS [114].

to improve resource allocation while taking task and user priorities of RPMS into account. First, they build a system utility
maximization problem for task QoS. To model computation offloading, the reward, cost, and penalty functions are provided.
Their model is illustrated in Fig. 21. Wang et al. proposed a cooperative Nash bargaining game for human interaction and a
Stackelberg negotiation game for individual human interaction to enhance the network and mitigate interference [73].

a Mobility Model In Group-Based RPMS

Users’ mobility models significantly impact the inter-user routes in the group health monitoring system based on IoMT,
where entity mobility models and group mobility models can be distinguished. The most commonly used entity mobility
models include the random waypoint model, the random walk model, the random direction model, and the Gauss-Markov
mobility model [107]. Mobility models for small groups consist of a reference point mobility model, column mobility model,
and pursue mobility mode. According to Liu et al., a Gauss-Markov mobility model explains the most about human mobility
and it has the most similarities to the human body [58]. In contrast, the Gauss-Markov mobility model represents the
mobility of entities, not the characteristics of the RPMS with multiple nodes. As a result, a three-dimensional motion model
of key nodes of the human body was added to their Gauss-Markov mobility model. By using this new model, the walking
process can be better explained in a RPMS network.

a A Cooperative Communication Approach for Grouped-based RPMS

A cooperative approach is beneficial to achieving the desired network performance [115]. To mitigate interference in
wireless networks, cooperative communication is a common technique that is always used. By using spatial diversity, re-
duces the transmission power and maximizes the capacity of specific nodes when two or more nodes come together to
build a virtual antenna array for mutual communication. As a result of channel interference in crowded environments, the

20
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

RPMS will perform less well and be less useful [116]. For interference mitigation, cooperative communication has been stud-
ied by some researchers in the RPMS. The distributed orthogonal code allocation scheme and the time reference correlation
scheme were proposed by Ali et al. in the context of overlapping super frames and channel sharing [79]. As a remedy for
resolving data transmission interruption, Manirabona et al. proposed a routing protocol based on three cooperative commu-
nication scenarios: (1) between two coordinators; (2) between a coordinator and node, and (3) between two nodes [80]. An
inter-hop relay-assisted cooperative communication methodology was presented by Dong et al. to enhance coexistence and
minimize interference in RPMS [81]. Through an experimental study [117], the performance of two-hop on-body relay-aided
cooperative communication has been investigated for RPMS. The three-branch diversity gain was implemented with two
low-complexity relays. A realistic RPM system was used to conduct this study in both scenarios of external radio interfer-
ence and no external interference. Using channels gains measured from BAN links, first- and second-order statistics were
calculated to evaluate the effectiveness of the scheme.

a Routing Protocols In Group Based RPMS

The routing algorithm is the most critical part of multihop networks. The routing algorithm is what determines whether
the packets will arrive at their destinations. Researchers of [69] explained that a protocol opportunistic routing protocol
based on random linear network coding was proposed for data transmission over the RPMS. In contrast to traditional rout-
ing protocols, using this protocol uses a distance-unstable chain that is unavailable in traditional methods, and enables
significant reductions in end-to-end latency and network throughput.
Ref. [70] presented a design of the remote patient monitoring system along with an improved version of the acquisition
tree protocol as well as the fuzzy logic routing algorithm. A sensor node like this can transmit data in real-time, ensuring
high reliability and balancing node energy consumption, extending network lifetime and maximizing the value of network
resources. Authors in Ref [71] proposed a Collaborative RPMS using IEEE 802.15.4 WBAN to reduce transmission delay and
energy consumption in these systems. Local sensors are clustered on multiple parts of the body to create the proposed
architecture. There is a wireless local gateway (WLG), which resides inside the patients’ premises. The WLG communicates
with each cluster head. Through a wireless gateway, data from the WLG is transmitted to a remote hospital gateway (HG),
where it is used to compile diagnoses or create medication profiles. It calculates criticality level based on input variables
such as patient age, body temperature, blood oxygen saturation percentage, and blood pressure using a fuzzy logic decision
model.
Furthermore, the work in Ref. [103] proposed an energy-aware objective function that could improve network lifetime
based on a low-power and low-loss high-speed routing protocol. This objective function takes into account the expected
transmission count of packets and the energy remaining in the nodes to optimize the routing of packets. A routing protocol
based on location-assisted routing was designed by Meharouech and colleagues to address any problems associated with
QoS, mobility of users, and energy-efficiency considerations [72]. RPMS employ many traditional routing protocols. However,
these routing protocols do not work well for this mobile version of RPMS. RPMS include plenty of nodes as well as are
sparse and mobile, so a steady end-to-end communication path between the ordinary sensor nodes and the sink node
cannot be maintained. As a result, data transmission relies on node movement and can only be achieved when nodes are
within one another’s range. Immediately after that, the Delay-Tolerant Mobile Sensor Network (DTMSN) emerged as an
essential research area in wireless networks as explained by Ref. [81]. Authors in [68] described SHARP as a hybrid routing
protocol that uses AODV and OLSR. It describes various routing mechanisms of the protocol that addresses different network
constraints and traffic requirements.

Analytical comparison of areas of focus in RPMS

As mentioned in Section III, the existing literature review was reviewed and weighted based on other areas of focus.
Fig. 22 gives a percentage distribution of areas of focus in RPMS literature from the years of 2011 to 2022. As can be seen,
a majority of studies focused on the application and techniques applied in RPMS. The two areas show growth from 2011 to
2019 and decline from 2020 to 2022. In the years 2011 to 2013, the RPMS literature charted a 4.7% growth in the area of
applications. This compares to 7.8% in Architecture, 10.9% in Techniques and only 4.7% in challenges. Between 2014 and 2019,
the percentages in application and techniques continued to rise, while architecture and challenges declined. From 2020 to
2022 there was a total decline in all areas of focus, showing a decline in the literature of RPMS around this time.
Also, Table 7 through Table 8 analyse the areas of focus in RPMS and give a comparison amongst those studies.
Table 7 gives a summary and compares the literature in RPMS in the area of application. There was an observation that
in RPMS literature, one or more physiological parameters such as Heart Rate, Respiratory Rate, Blood Oxygen Saturation,
Blood Pressure, or Body Temperature were implemented. Most of these systems are targeted at elderly people or general
adults. The RPMS use wearable sensors with different mobile control units as well as using different low power commu-
nication networks. Mobile RPMS have shown to have the advantage of giving the patient increased freedom. Additionally,
mobility affects the throughput and network lifetime of nodes as they join and leave the network. They are more likely to
be network failures because the node may keep moving away from the sink.
Table 8 presents a summary and comparison of literature in areas of architecture of RPMS. In several studies, researchers
used RPMS based on three to four metrics. Several sensors are placed on the body, from the tongue to the chest to the

21
 K. Boikanyo, A.M. Zungeru, B. Sigweni et al.
Table 7
Comparison of application based on vital monitoring RPMS.

Targeted Mobile Control

Refs. Year Vital Monitored Clients Sensor Type/ Placement Unit (MCU) Wireless Comm. Uniqueness Limitation

[25] 2010 Vital Monitoring Adults Reflectance pulse Computer Zigbee transceiver Provide accurate
(HR,RR,SP02, and BP) oximeter module or a mini-USB photoplethysmograph (PPG)
signals without requiring
filters.
[26] Vital Monitoring (BT) Adults Two digital Bluetooth Limited to measuring only one
temperature probes/ parameter. Long term may
two ear canals affect the ear canal.
[28] Vital Monitoring Adults Internet of things (IoT) Smartphone WBAN offer the doctor a virtual
(HR,RR,SP02, and BP, BT) sensors presence within the
ambulance
[29] Vital Monitoring (BT, HR) Adults Microcontroller RF transmitter
/Receiver
[30] Vital Monitoring (HR, Adults Wristwatch/wrist Mobile phone Cellular network Can be applied on clinical The system rely on the mobile
22

SPO2) setting phone, when it’s off the will

be network failure.
[31] fall detection Adults fall detection raspberry pi WI-FI Have voice recognition Limited to diagnosing one
sensor/web cam/ model B+ capabilities and can obtain type of disease
Arms, knees confirmation about falls from
the user which reduces false
positive rate.
[32] Mentally bipolar Adults textile-based Includes a questionnaire Using a questionnaire may
autonomic nervous system, a medication intake give inaccurate data as the
system./ Head monitoring system, and sleep patients are mentally
activities monitoring impaired.
[33] Diabetes Adults Camera/ Close to the RFID Capability to monitor food Prone to error due to its
patient body intake, blood pressure, insulin distance between the camera
levels, calories, medications, and the patient’s body.
and activity.
[35] Cardiovascular Adults Mobile app Mobile phone WI-FI Can detect several diseases Does not give a clear overview

Scientific African 20 (2023) e01638

based on the physiological of how the sensor will be
parameter observed deployed in the body
 K. Boikanyo, A.M. Zungeru, B. Sigweni et al.
Table 8
Comparison of architectures of RPMS.

Sensor Mobile Control Unit Wireless No of

Refs. Vital Monitored Sensor Type Placement (MCU) Comm. Tiers Uniqueness Limitation

[42] Monitor sleep System-on-a-Chip Tongue Microcontroller Bluetooth Through the platform, external May cause discomfort due to
(SoC) RF power will be harvested to its placement on the tongue
recharge the battery.
[43] Vital Monitoring wearable body sensor Upper body WI-FI 3 emphasizes authentication, The system is not automated
Microcontroller/android power consumption, and which results in errors and
mobile phone accuracy in transferring health delays during manual
data to a medical server operation
[27] RR, HR, body ultra-wideband Away from Smartphone Radar using ultra-wideband Limited to only one age group
movements, and impulse-radio radar the body impulse-radio radar to detect which results in a lower
sleep quality various physiological variables accuracy index Can monitor
23

one patient at a time which

limits the efficiency of
monitoring
[44]. Vital Monitoring ECG capacitive Chest microcontroller ANT 4 utilizing ANT protocol, and It may be difficult for users to
(BT, HR) electrodes signal sampling in the idle work with rigid electrodes and
mode are employed to reduce may lead to motion artifacts
power consumption in the signal.
[118] Vital Monitoring Pulse oximeter and Finger, Arduino UNO WI-FI 3 The Android system is user Medical history of patients
(BT, SPO2) Body temperature armpit microcontroller friendly and effectively should also be kept on a local
monitors the patient server for quick analysis and
any medical procedures.
[46]. Bladder pressure Implanted miniature bladder BLE 3 shown that the system could Implemented in rabbit,
monitor packaged sensor meet the bladder pressure therefore introducing it to
monitoring requirements in an human is a need.
in vitro experiment

Scientific African 20 (2023) e01638

K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Fig. 22. Percentage Distribution of area of focus in RPMS from year 2014–2022.

armpits, to monitor vitals. There are also implantable sensors, which have been tested on rabbits. Different wireless tech-
nologies are used by the RPMS, including Bluetooth, WIFI, RADAR, ANT, and BLE. The review has shown that the higher the
number of tiers the better the network lifetime. This is because more tiers decrease the number of hops during a transmis-
sion and the more the hops the higher the network lifetime. Architecture has shown that these systems are much smaller
in size and portable for daily use and during mobility.

Conclusion

RPMS are essential to our daily lives. This has been proven by technology advancements in RPMS. As part of this paper,
RPMS were analysed in four areas of focus being; the application, architectures, the techniques applied and also the chal-
lenges. RPMs can be applied in elderly people, neonates, and disabled. They have shown to have different structures, some
being wearable, while others are non-contact. The challenges faced by RPMS include throughput, network lifetime, QoS, and
energy efficiency. Several solutions to these challenges were presented, including the use of Markov models, game theory,
mobility models, cooperative communications and routing protocols. Architecture, communication models were extensively
examined as potential solutions to routing challenges in RPMS. Furthermore a statistical analysis and comparison in the
area of focus mentioned above was done to show the distribution of the literature from the year 2011 to 2022. The results
showed that there was a growth in RPMS research in all areas in the years 2011 to 2019. However, a number of declines
occurred from 2020 to 2022. The pandemic that hit the world in 2020 and 2021 contributed greatly to the decline.

Future aspects

RPMS are expected to have a bright future after the aforementioned challenges and issues have been resolved. There have
been several efforts in improving the performance of RPMS. Therefore the key design challenge in RPMS lies in extending
the lifespan of sensor devices while ensuring that the messages are delivered reliably and in a timely manner. Even though
routing protocol has been implemented to improve the performance in the sense of QoS, the future to addressing these
challenges of RPMS is the IoV framework. IoV have the advantages of having the ability of a vehicle communicating with
other vehicles, a person, a building, the road and almost everything that is connected to the internet. If RPMS is connected
to everything, imagine how much better the performance will be. The above reviewed and simulated protocols may offer
new opportunities to find optimal routing solutions in the Internet of Vehicle and RPMS. In the future, efficient RPMS will
improve clinical decision-making, self-management, and adherence to care plans. Payers and providers will be able to reduce
the cost of care. Additionally, it will reduce patients’ expenses and boost productivity at work. Patients will become more
engaged with the RPMS since they will be actively involved in their monitoring. In addition to optimizing clinical staff
efficiency, the systems will also prevent clinical staff shortages. Infectious diseases and hospital-acquired infections will be
prevented in some ways. Monitoring is carried out everywhere and at any time.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have
appeared to influence the work reported in this paper.

24
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

Funding

ORDI of the Botswana International University of Science and Technology

Public interest statement

Remote Patient Monitoring Systems (RPMSs) are considered one of the most critical research fields, as it deals with hu-
man lives. These solutions have gained prominence due to the emergence of highly infectious diseases. Due to this, the
world is also adopting IoT-based approaches to remotely monitor patients using real-time detection systems to mitigate fur-
ther spread. However, several challenges and flaws have been encountered with patient monitoring solutions. These include
mobility, the use of heterogeneous networks, and standardization of RPMS, automation, and Quality of Services. Studying the
literature and analysing the data, focusing on the applications, structures, techniques, and challenges faced, led to proper and
detailed analysis and classification of RPMS.

Acknowledgement

We would like to give thanks to Botswana International University of Science and Technology for providing the necessary
infrastructures and support for completing this research. I would like to give many thanks to the aforementioned supervisor
for the guidance and patience during the process of this research.

References

[1] B. Talbot, et al., Patient and clinician perspectives on the use of remote patient monitoring in peritoneal dialysis, Can. J. Kidney Heal. Dis. 9 (2022).
[2] T. Malche, et al., Artificial intelligence of things- (AIoT-) based patient activity tracking system for remote patient monitoring, J. Healthc. Eng. 2022
(2022).
[3] H. Fouad, A.S. Hassanein, A.M. Soliman, H. Al-Feel, Analyzing patient health information based on IoT sensor with AI for improving patient assistance
in the future direction, Meas. J. Int. Meas. Confed. 159 (2020) 107757.
[4] A.C. va. Graan, B. Scrooby, Y. Bruin, Recording and interpretation of vital signs in a selected private hospital in the KwaZulu-Natal province of South
Africa, Int. J. Africa Nurs. Sci. 12 (April 2019) (2020) 100199.
[5] J. Mohammed, C.H. Lung, A. Ocneanu, A. Thakral, C. Jones, A. Adler, Internet of things: remote patient monitoring using web services and cloud
computing, in: Proceedings of the IEEE International Conference on Internet of Things (iThings) 2014, 2014, pp. 256–263. 2014 IEEE Int. Conf. Green
Comput. Commun. GreenCom 2014 2014 IEEE Int. Conf. Cyber-Physical-Social Comput. CPS 20.
[6] M.A. Panhwar, et al., Energy-efficient routing optimization algorithm in WBANs for patient monitoring, J. Ambient Intell. Humaniz. Comput. 12 (7)
(2021) 8069–8081.
[7] F.A.C.D. De Farias, C.M. Dagostini, Y.D.A. Bicca, V.F. Falavigna, A. Falavigna, Remote patient monitoring: a systematic review, Telemed. e-Health 26 (5)
(2020) 576–583.
[8] M.U. Ahmed, M Björkman, A. Čaušević, H. Fotouhi, M. Lindén, An overview on the internet of things for health monitoring systems, ICST Inst. Comput.
Sci. Soc. Informatics Telecommun. Eng. (2016) 429–436.
[9] Design and development of patient monitoring system using android application, Int. J. Mod. Trends Eng. Res. 4 (5) (2017).
[10] M. Dhinakaran, K. Phasinam, J. Alanya-Beltran, K. Srivastava, D.V. Babu, S.K. Singh, A system of remote patients’ monitoring and alerting using the
machine learning technique, J. Food Qual. 2022 (2022).
[11] S. Majumder, E. Aghayi, M. Noferesti, H. Memarzadeh-Tehran, T. Mondal, Z. Pang, M.J. Deen, "Smart homes for elderly healthcare," recent advances
and research challenges, Sensors 17 (2017) 2496, doi:10.3390/s17112496.
[12] H.M. Reeve, A.M. Mescher, A.F. Emery, Experimental and numerical investigation of polymer preform heating, Am. Soc. Mech. Eng. Heat Transf. Div.
HTD 369 (6) (2001) 321–332.
[13] F.H. El-Fouly, R.A. Ramadan, R.A. Ramadan, E3AF: energy efficient environment-aware fusion based reliable routing in wireless sensor networks, IEEE
Access 8 (2020) 112145–112159.
[14] L.P. Malasinghe, N. Ramzan, K. Dahal, Remote patient monitoring: a comprehensive study, J. Ambient Intell. Humaniz. Comput. 10 (1) (2019) 57–76.
[15] P. Gope, et al., Body area sensor networks challenges and opportunies - Univeristy of Virginia, IEEE Sens. J. 13 (5) (2016) 415–418.
[16] S. Majumder, T. Mondal, M.J. Deen, Wearable sensors for remote health monitoring, Sensors (Switzerland) 17 (1) (2017).
[17] I. Krogstad, Internet of Things for Healthcare Technologies, 73, 2021.
[18] B.S. Kim, K.H. Kim, K. Il Kim, A survey on mobility support in wireless body area networks, Sensors (Switzerland) 17 (4) (2017) 1–18.
[19] K. Wei, L. Zhang, Y. Guo, X. Jiang, Health monitoring based on internet of medical things: architecture, enabling technologies, and applications, IEEE
Access 8 (2020) 27468–27478.
[20] L. Minn, A. Senior, K.P. Seng, G. Ijemaru, Deployment of IoV for smart cities : applications, architecture and challenges, IEEE Access PP (December)
(2018) 1.
[21] T. Thirugnanam, M.R. Ghalib, A new healthcare architecture using IoV technology for continuous health monitoring system, Health Technol. (Berl) 10
(3) (2020) 289–302.
[22] Bizvibe, “Top 10 Remote Patient Monitoring Companies in the World 2020, Remote Patient Monitoring Market.” [Online]. Available: https://blog.
bizvibe.com/blog/top-remote-patient-monitoring-companies. [Accessed: 21-Jul-2022].
[23] S. Dolanl, “Top Remote Patient Monitoring Companies & Startups in 2022.” [Online]. Available: https://www.insiderintelligence.com/insights/
remote- patient- monitoring- companies- startups/. [Accessed: 21-Jul-2022].
[24] R.K. Pathinarupothi, M.V. Ramesh, E. Rangan, Multi-layer architectures for remote health monitoring, in: Proceedings of the " IEEE 18th International
Conference on e-Health Networking, Applications and Services (Healthcom), 2016 2016.
[25] K. Li, S. Warren, A wireless reflectance pulse oximeter with digital baseline control for unfiltered photoplethysmograms, IEEE Trans. Biomed. Circuits
Syst. 6 (3) (2012) 269–278.
[26] P. Oğuz, G. Ertaş, Wireless dual channel human body temperature measurement device, IEEE Xplore, no. ICECCO 2013 (2013) 52–55.
[27] C. Dong, et al., Non-contact screening system based for COVID-19 on XGBoost and logistic regression, Comput. Biol. Med. 141 (2022) Feb..
[28] N. Kumar, S.N. Panda, P. Pradhan, R.K. Kaushal, IoT based hybrid system for patient monitoring and medication, EAI Endorsed Trans. Pervasive Healh.
Technol. 5 (19) (2019).
[29] C. K. Das, M. W. Alam, M. I. Hoque, A wireless heartbeat and temperature monitoring system for remote patients, Int. Conf. Mech. Eng. Renew. Energy
(2013) 1–7.

25
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

[30] S. Kumar, et al., A wristwatch-based wireless sensor platform for IoT health monitoring applications, Sensors (Switzerland) 20 (6) (2020).
[31] S. Greene, H. Thapliyal, D. Carpenter, IoT-Based fall detection for smart home environments, in: Proceedings of the 2016 IEEE International Symposium
on Nanoelectronic and Information Systems (iNIS), 2017, pp. 23–28.
[32] A. Lanata, G. Valenza, M. Nardelli, C. Gentili, E.P. Scilingo, Complexity index from a personalized wearable monitoring system for assessing remission
in mental health, IEEE J. Biomed. Health Inf. 19 (1) (2015).
[33] S.K. Prabhakar, H. Rajaguru, Adaboost classifier with dimensionality reduction techniques for epilepsy classification from EEG, IFMBE Proc. 66 (2018)
185–189.
[34] A. Fioravanti, G. Fico, D. Salvi, R.I. García-Betances, M.T. Arredondo, Automatic messaging for improving patients engagement in diabetes management:
an exploratory study, Med. Biol. Eng. Comput. 53 (2015) 1285–1294.
[35] T.G. Stavropoulos, A. Papastergiou, L. Mpaltadoros, S. Nikolopoulos, I. Kompatsiaris, Iot wearable sensors and devices in elderly care: a literature
review, Sensors (Switzerland) 20 (10) (2020).
[36] W. Chen, I. Ayoola, S.B. Oetomo, L. Feijs, Non-invasive blood oxygen saturation monitoring for neonates using reflectance pulse oximeter, in: Proceed-
ings of the Design, Automation & Test in Europe Conference & Exhibition (DATE 2010), 2010, pp. 1530–1535.
[37] S.L. Rossol, et al., Non-contact video-based neonatal respiratory monitoring, Children 7 (171) (2020) 1–10.
[38] M. Awais, et al., An internet of things based bed-egress alerting paradigm using wearable sensors in elderly care environment, Sensors (Switzerland)
19 (11) (2019).
[39] Y. Liu, et al., A novel cloud-based framework for the elderly healthcare services using digital twin, IEEE Access 7 (2019).
[40] J. Mathew, J. Lail, A.C. Chang, J.L. Jefferies, Outpatient Monitoring and Self-Care, Elsevier Inc., 2018.
[41] G. Sannino, I. De Falco, G. De Pietro, A supervised approach to automatically extract a set of rules to support fall detection in an mHealth system,
Appl. Soft Comput. 34 (2015) 205–216.
[42] K. Y. Yeh, et al., A wireless monitoring system using a tunneling sensor array in a smart oral appliance for sleep apnea treatment, Sensors (Switzer-
land) 17 (10) (2017).
[43] V. Wahane, D.P.V. Ingole, Interactive mobile health monitoring system, Int. J. Adv. Comput. Sci. Appl. 8 (4) (2017).
[44] E. Nemati, M.J. Deen, T. Mondal, A wireless wearable ECG sensor for long-term applications, IEEE Commun. Mag. 50 (1) (2012) 36–43.
[45] M.A. Panhwar, K.A. Memon, I.A. Koondhar, M. Channa, S. Abbasi, Design of android-based remote patient monitoring system, Int. J. Adv. Comput. Sci.
Appl. 9 (6) (2018).
[46] Y. Zhong, et al., Development of an implantable wireless and batteryless bladder pressure monitor system for lower urinary tract dysfunction, IEEE J.
Trans. Eng. Heal. Med. 8 (2020) 1–7.
[47] X. Cheng, et al., Implantable and self-powered blood pressure monitoring based on a piezoelectric thinfilm: simulated, in vitro and in vivo studies,
Nano Energy 22 (2016) 453–460.
[48] H. Gjoreski, M. Luštrek, M. Gams, Accelerometer placement for posture recognition and fall detection, in: Proceedings of the 2011 7th International
Conference on Intelligent Environments IE 2011, 2011, pp. 47–54.
[49] K. Sruthi, E. V Kripesh, K. A. U. Menon, A survey of remote patient monitoring systems for the measurement of multiple physiological parameters,
Heal. Technol. Springer 7 (2017) 153–159.
[50] M. Kamel, S. Fawzy, A. El-Bialy, A. Kandil, Secure remote patient monitoring system, IEEE Xplore (2011) 339–342.
[51] D. De, P. Bharti, S. K. Das, S. Chellappan, Multimodal wearable sensing for fine-grained activity recognition in healthcare, IEEE Internet Comput. 19
(2015) 26–35.
[52] T.W. Bae, K.K. Kwon, K.H. Kim, Vital block and vital sign server for ECG and vital sign monitoring in a portable u-vital system, Sensors (Switzerland)
20 (4) (2020).
[53] M. Garbarino, M. Lai, S. Tognetti, R. Picard, and D. Bender, “Empatica E3 - A wearable wireless multi-sensor device for real-time computerized
biofeedback and data acquisition,” Mobihealth, pp. 1–4, 2014.
[54] C.A. Chin, G.V. Crosby, T. Ghosh, R. Murimi, Advances and challenges of wireless body area networks for healthcare applications, in: Proceedings of
the International Conference on Computing, Networking and Communications ICNC’12, 0869, 2012, pp. 99–103.
[55] Y. Mohammed, A.S. Mohammed, H.T. Abdulkarim, C. Danladi, A. Victor, R. Edoka, Development and implementation of an internet of things (IOT)
based remote patient monitoring system, in: Proceedings of the 15th International Conference on Electronics, Computer and Computation, ICECCO
2019, 2019.
[56] S. Tennina, M. Santos, A. Mesodiakaki, P. Mekikis, E. Kartsakli, A. Antonopoulos, in: WSN4QoL : WSNs for remote patient monitoring in e-health
applications, IEEE Xplore, 2016, pp. 0–5.
[57] M.M. Sandhu, M. Akbar, M. Behzad, N. Javaid, Z.A. Khan, U. Qasim, Mobility model for WBANs, in: Proceedings of the International Conference on
Broadband, Wireless Computing, Communication and Applications BWCCA 2014, 1, 2014, pp. 155–160.
[58] Y. Liu, D. Liu, G. Yue, BGMM: a body gauss-markov based mobility model for body area networks, Tsinghua Sci. Technol. 23 (3) (2018) 277–287.
[59] A. Karimi, B. Kiamanesh, F. Zarafshan, S.A.R. Al-Haddad, Markov process modeling for wireless sensor network availability with QOS constraints, Appl.
Mech. Mater. 330 (2013) 1054–1058.
[60] K. Takizawa, et al., Channel models for wireless body area networks, in: Proceedings of the 30th Annual International Conference of the IEEE Engi-
neering in Medicine and Biology Society, EMBS’08 - "Personalized Healthcare through Technology", 2008, pp. 1549–1552.
[61] A. Awang, U.F. Abbasi, Performance evaluation of cross-layer opportunistic MAC/routing with node’s mobility for wireless body area networks, in:
Proceedings of the 2015 IEEE 12th Malaysia International Conference on Communications (MICC 2015), 2016, pp. 30–35.
[62] S. Misra, J. Mahapatro, M. Mahadevappa, N. Islam, Random room mobility model and extra-wireless body area network communication in hospital
buildings, IET Netw. 4 (1) (2015) 54–64.
[63] Y. Liao, Y. Han, Q. Yu, Q. Ai, Q. Liu, M.S. Leeson, Wireless body area network mobility-aware task offloading scheme, IEEE Access 6 (2018) 61366–61376.
[64] Z.A. Khan, M.B. Rasheed, N. Javaid, B. Robertson, Effect of packet inter-arrival time on the energy consumption of beacon enabled MAC protocol for
body area networks, Proced. Comput. Sci. 32 (2014) 579–586.
[65] M.A. Razzaque, M.T. Hira, M. Dira, QoS in body area networks: a survey, ACM Trans. Sens. Netw. 13 (3) (2017).
[66] X. Yuan, C. Li, Y. Song, L. Yang, S. Ullah, On energy-saving in e-healthcare: a directional MAC protocol for WBAN, in: Proceedings of the IEEE Globecom
Workshops (GC Wkshps), 2015.
[67] M. Roy, C. Chowdhury, N. Aslam, Designing 2-hop interference aware energy efficient routing (Hier) protocol for wireless body area networks, in:
Proceedings of the Lecture Notes in Computer Science (Including Subseries) Lecture Notes in Artificial Intelligence Lecture Notes Bioinformatics),
10340, LNCS, 2017, pp. 262–283.
[68] T.A. Ramrekha, C. Politis, A hybrid adaptive routing protocol for extreme emergency ad hoc communication, in: Proceedings of the International
Conference on Computer Communications and Networks (ICCCN), 2010.
[69] X. Lv, X. Li, L. Liu, Y. Li, Research of opportunistic routing protocol in wearable sensor network, in: Proceedings of the APWCS - 2010 Asia-Pacific
Conference on Wearable Computing Systems, 2010, pp. 99–103.
[70] J. Jiang, Y. Liu, F. Song, R. Du, M. Huang, The routing algorithm based on fuzzy logic applied to the individual physiological monitoring wearable
wireless sensor network, J. Electr. Comput. Eng. 2015 (2015).
[71] S. Rajalakshmi, K. V. Madhav, R. Abhishek, Y. V. Raghava, Collaborative remote patient monitoring system using IEEE 802. 15. 4 wireless body area
networks, Int. J. Comput. Commun. Eng. Collab. 8 (2) (2019) 60–72.
[72] A. Meharouech, J. Elias, A. Mehaoua, Joint epidemic control and routing in mass gathering areas using Body-to-Body Networks, in: Proceedings of the
13th International Wireless Communications and Mobile Computing Conference (IWCMC) 2017, 2017, pp. 1759–1764.

26
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

[73] J. Wang, Y. Sun, Y. Ji, Priority-based capacity and power allocation in co-located WBANs using Stackelberg and bargaining games, J. Supercomput. 74
(7) (2018) 3114–3147.
[74] A.M. Zungeru, L.M. Ang, K.P. Seng, Termite-hill: routing towards a mobile sink for improving network lifetime in wireless sensor networks, in: Pro-
ceedings of the International Conference on Intelligent Systems Modelling and Simulation, ISMS 2012, 2012, pp. 622–627.
[75] L.K. Ketshabetswe, A.M. Zungeru, M. Mangwala, J.M. Chuma, B. Sigweni, Communication protocols for wireless sensor networks: a survey and com-
parison, Heliyon 5 (5) (2019) e01591.
[76] N. Sangeetha Priya, R. Sasikala, S. Alavandar, L. Bharathi, Security aware trusted cluster based routing protocol for wireless body sensor networks,
Wirel. Pers. Commun. 102 (4) (2018) 3393–3411.
[77] A. Meharouech, J. Elias, A. Mehaoua, A two-stage game theoretical approach for interference mitigation in Body-to-Body Networks, Comput. Netw.
95 (June) (2016) 15–34.
[78] R. Kazemi, R. Vesilo, E. Dutkiewicz, G. Fang, Inter-network interference mitigation in wireless body area networks using power control games, in:
Proceedings of the ISC 2010 - 2010 10th International Symposium on Communications and Information Technologies„ 2010, pp. 81–86.
[79] A. Manirabona, S. Boudjit, L.C. Fourati, Energy aware routing protocol for inter WBANs cooperative communication, in: Proceedings of the Interna-
tional Symposium on Networks, Computers and Communications ISNCC 2015, 2015.
[80] J. Dong, D. Smith, Coexistence and interference mitigation for wireless body area networks: improvements using on-body opportunistic relaying, IEEE
Trans. Wirel. Commun. (2013) 1–25.
[81] X. Ma, X. Zhang, R. Yang, Reliable energy-aware routing protocol in delay-tolerant mobile sensor networks, Wirel. Commun. Mob. Comput. 2019
(2019).
[82] Ó. Gama, P. Carvalho, J.A. Afonso, P.M. Mendes, Wireless sensor networks with QoS for e-health and e-emergency applications, in: Proceedings of the
International Workshop on e-Health Services and Technologies, EHST 2008 - Conjunction with ICSOFT 2008, 2008, pp. 60–69.
[83] A.P. Abidoye, N.A. Azeez, A.O. Adesina, K.K. Agbele, H.O. Nyongesa, Using wearable sensors for remote healthcare monitoring system, J. Sens. Technol.
01 (02) (2011) 22–28.
[84] N. Ahmad, et al., Improved QoS aware routing protocol (IM-QRP) for WBAN based healthcare monitoring system, IEEE Access 10 (November) (2022)
121864–121885.
[85] A. Darwish, A.E. Hassanien, Wearable and implantable wireless sensor network solutions for healthcare monitoring, Sensors 11 (6) (2011) 5561–5595.
[86] A.M. Zungeru, L.M. Ang, K.P. Seng, Classical and swarm intelligence based routing protocols for wireless sensor networks: a survey and comparison,
J. Netw. Comput. Appl. 35 (5) (2012) 1508–1536.
[87] A. Samanta, S. Misra, Energy-efficient and distributed network management cost minimization in opportunistic wireless body area networks, IEEE
Trans. Mob. Comput. 17 (2) (2018) 376–389.
[88] M.T. Arefin, M.H. Ali, A.K.M.F. Haque, Wireless body area network: an overview and various applications, J. Comput. Commun. 05 (07) (2017) 53–64.
[89] H. Mshali, T. Lemlouma, M. Moloney, D. Magoni, A. Survey, H. Monitoring, A survey on health monitoring systems for health smart homes, Int. J. Ind.
Ergon. (2018).
[90] A. Sagahyroon, “Remote Patients Monitoring : Challenges,” IEEE, 2017.A. Sagahyroon, “Remote patients monitoring: challenges,” 2017.
[91] N. El-Rashidy, S. El-Sappagh, S. M. Riazul Islam, H.M. El-Bakry, S. Abdelrazek, Mobile health in remote patient monitoring for chronic diseases:
Principles, trends, and challenges, Diagnostics 11 (4) (Apr. 2021).
[92] Center for Technology and Aging, Technologies for Remote Patient Monitoring for Older Adults,” An Initiative of the SCAN Foundation and, Public
Health Institute (, 2010).
[93] Y.I. Joo, T.J. Lee, D.S. Eom, Y. Lee, K.H. Tchah, Power-efficient and QoS-aware scheduling in Bluetooth scatternet for wireless PANs, IEEE Trans. Consum.
Electron. 49 (4) (2003) 1067–1072.
[94] G. Keifer, F. Effenberger, No Title No Title, Angew. Chem. Int. Ed. 6 (11) (1967) 951–952.
[95] P. KARTHICK, C. SURESHKUMAR, P. ARUNPRASAD, S. PUSPARAJ, M. PUSPARAJ, and N. S. VANITHA, “Embedded Based Real-time Patient Monitoring
System,” Int. J. VLSI Embed. Syst., vol. 05, no. March, pp. 773–777, 2014.
[96] Y. Zhang, G. Chen, H. Du, X. Yuan, and M. Kadoch, “Real-Time Remote Health Monitoring System Driven by 5G MEC-IoT,” Electronics, vol. 9, no. 1753,
pp. 1–17, 2020.
[97] Y. Wang, Y. Liu, Analysis and comparison of WSN routing protocols oriented to WBAN, Appl. Mech. Mater. 241–244 (2013) 952–955.
[98] M.B. Yassein, M. Hamdan, H.A. Shehadeh, L. Mrayan, A novel approach for health monitoring system using wireless sensor network, Int. J. Commun.
Antenna Propag. 7 (4) (2017).
[99] R. Kumari, P. Nand, Performance comparison of various routing protocols in WSN and WBAN, in: Proceedings of the 2016 International Conference
on Computing, Communication and Automation (ICCCA) 2016, 2017, pp. 427–431.
[100] M. Hamoud, R. Hassoun, A. Mahmoud, M. Alkubeily, A.S. Ahmad, Comparison of MANET routing protocols used in home health-monitoring system
for elderly pati, Int. J. Sci. Res. 6 (2) (2017) 988–993.
[101] J. Guo, Mathematical modeling and simulation of wireless sensor network coverage problem, in: Proceedings of the Wireless Communications and
Mobile Computing, 2021, 2021.
[102] Y.H. Chu, Y.C. Hsieh, C.H. Wang, Y.C. Pan, R.I. Chang, UPHSM: ubiquitous personal health surveillance and management system via WSN agent on
open source smartphone, in: Proceedings of the 13th IEEE International Conference on e-Health Networking Applications and Services (Healthcom
2011), 2011, pp. 60–63.
[103] V. Bhanumathi, C.P. Sangeetha, A guide for the selection of routing protocols in WBAN for healthcare applications, Hum. Centr. Comput. Inf. Sci. 7 (1)
(2017) 1–19.
[104] Y. Qu, G. Zheng, H. Ma, X. Wang, B. Ji, H. Wu, A survey of routing protocols in WBAN for healthcare applications, Sensors (Switzerland) 19 (7) (2019).
[105] A.R. Basha, A Review on Wireless Sensor Networks: Routing, 125, Springer US, 2022.
[106] R. Chiwariro, N. Thangadurai, Deep learning-based prioritized packet classification and optimal route selection in wireless multimedia sensor net-
works, IETE J. Res. (2023).
[107] T. Camp, J. Boleng, V. Davies, A survey of mobility models for ad hoc network research, Wirel. Commun. Mob. Comput. 2 (5) (2002) 483–502.
[108] L. Hu, Y. Zhang, D. Feng, M.M. Hassan, A. Alelaiwi, A. Alamri, Design of QoS-aware multi-level MAC-layer for wireless body area network, J. Med. Syst.
39 (12) (2015).
[109] R. Goyal, N. Mittal, L. Gupta, A. Surana, Routing protocols in wireless body area networks: architecture, challenges, and classification, Wirel. Commun.
Mob. Comput. 2023 (2023) 1–19.
[110] J.Y. Chang, P.H. Ju, An energy-saving routing architecture with a uniform clustering algorithm for wireless body sensor networks, Futur. Gener. Com-
put. Syst. 35 (2014) 128–140.
[111] M.M. Alam, E. Ben Hamida, Interference mitigation and coexistence strategies in IEEE 802.15.6 based wearable body-to-body networks, Lect. Notes
Inst. Comput. Sci. Soc. Telecom. Eng. LNICST 156 (March) (2015) 665–677.
[112] S.M. Shimly, D.B. Smith, S. Movassaghi, Cross-layer designs for body-to-body networks: adaptive CSMA/CA with distributed routing, IEEE Int. Conf.
Commun. 2018-May (March) (2018).
[113] L. Guo, Z. Ning, W. Hou, B. Hu, P. Guo, Quick answer for big data in sharing economy: innovative computer architecture design facilitating optimal
service-demand matching, IEEE Trans. Autom. Sci. Eng. 15 (4) (2018) 1494–1506.
[114] X. Yuan, H. Tian, H. Wang, H. Su, J. Liu, A. Taherkordi, Edge-enabled WBANs for efficient QoS provisioning healthcare monitoring: a two-stage potential
game-based computation offloading strategy, IEEE Access 8 (2020).
[115] M. Ali, H. Moungla, M. Younis, A. Mehaoua, Inter-WBANs interference mitigation using orthogonal walsh hadamard codes, IEEE Int. Symp. Pers. Indoor
Mob. Radio Commun. PIMRC (2016).

27
K. Boikanyo, A.M. Zungeru, B. Sigweni et al. Scientific African 20 (2023) e01638

[116] S. Huang, et al., A novel framework and the application mechanism with cooperation of control and management in multi-domain WSON, J. Netw.
Syst. Manag. 21 (3) (2013) 453–473.
[117] J. Dong, Y. Ge, D.B. Smith, Two-hop relay-assisted cooperative communication in wireless body area networks: an empirical study, ACM Trans. Sens.
Networks 12 (4) (2016) 1–13.
[118] Y. Xu, J. Brownjohn, D. Kong, A non-contact vision-based system for multipoint displacement monitoring in a cable-stayed footbridge, Struct. Control
Heal. Monit. 25 (5) (2018) 1–23.

28

View publication stats