Comparison of Bluetooth Low Energy (BLE), Wi-Fi, Serial and 5G in

IoMT

Nina Pearl Doe

, Stefan Scharoba

, Marc Reichenbach

and Christian Herglotz

Chair of Computer Engineering, Brandenburg University of Technology Cottbus-Senftenberg, Germany

Chair of Integrated System, Applied Microelectronics and Computer Engineering, University of Rostock, Germany

{ninapearl.doe, Stefan.Scharoba, christian.herglotz}@b-tu.de, marc.reichenbach@uni-rostostock.de

Keywords:

Internet of Medical Things (IoMT), Performance Evaluation, Wireless Communication, Electrocardiogram

(ECG) Monitoring, Real-Time Systems.

Abstract:

With the inception of Industry 4.0, incorporating technologies like the Internet of Things (IoT) into healthcare

has become essential. This integration is commonly referred to as the Internet of Medical Things (IoMT). The

IoMT is the connection of medical devices using wired or wireless data transmission technology to allow data

exchange with the goal of improving the overall healthcare delivery. Despite the numerous advantages that

IoMT brings into the healthcare process, there are potential performance challenges that may occur if factors

such as data quality and reliability of the IoT devices in different environmental settings are not properly

considered. The purpose of this paper is to analyse the performance of connected medical IoT devices that

are used for heartrate monitoring based on the aforementioned factors. The setup of the IoMT consists of

sensor nodes, which transmit the Electrocardiogram (ECG) data through a multi-protocol gateway to a central

server for further data processing. This paper presents the performance analysis of the comparison of four

communication technologies: Serial (UART), Bluetooth Low Energy (BLE), Wi-Fi, and 5G NR for real-time

ECG monitoring applications, while taking notice of environmental factors that may affect performance. The

sensor data transmission is evaluated based on round trip time (RTT) latency, ensuring a desirable throughput

and minimal or no data loss. The data readings were taken at varying distances (0.1m to 17m) and sampling

rates (300Hz and 1000Hz). The experimental results show that while Serial communication achieves the

lowest latency (3.96ms - 4.37ms), Wi-Fi demonstrates consistent Gateway-Server performance (40ms - 60ms

RTT), 5G excels in short-range communication (1.8ms - 2.0ms Sensor Node-Gateway RTT), and BLE provides

balanced performance (4.86ms - 7.57ms latency). Wi-Fi performed better in long-range scenarios (43.48ms -

66.23ms RTT) and maintaining stable performance at longer ranges while 5G shows superior performance in

short-range, high-frequency scenarios.

1 INTRODUCTION

In healthcare, the Internet of Medical Things (IoMT)

has emerged as an evolutionary paradigm which

leverages connected network devices to enhance the

patient’s journey through the hospital, which includes

patient care, monitoring and medical diagnosis, and

decision making (Dimitrov, 2016). IoMT systems

rely on various network communication technologies

to allow exchange of data between the medical de-

vices and the server or monitoring system.

This paper focuses particularly on, Bluetooth Low

Energy (BLE), Wi-Fi, 5G, and UART serial commu-

nication technologies. Bluetooth Low Energy (BLE)

is widely used in IoMT because it of its low power

consumption capability and works well with smart-

phones and tablets. It is especially useful for short-

distance communication in wearable medical device

(Girolami et al., 2020). Wi-Fi has a high advantage

of fast data transfer within local networks (Pahla-

van and Krishnamurthy, 2021). It is also good for

sending large amounts of medical data, such as high-

resolution medical images or continuous streams of

vital signs. The ﬁfth-generation cellular network

technology, or 5G, promises extremely low latency

and high bandwidth, which makes it perfect for real-

time monitoring both locally and remotely (Varga

et al., 2020). Because of its dependability and

simplicity in short-range wired connections between

medical devices and local gateways, UART Com-

munication, despite not being a wireless technology

remains essential in the Internet of medical things

Doe, N. P., Scharoba, S., Reichenbach, M. and Herglotz, C.

Comparison of Bluetooth Low Energy (BLE), Wi-Fi, Serial and 5G in IoMT.

DOI: 10.5220/0013321500003911

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2025) - Volume 2: HEALTHINF, pages 859-866

ISBN: 978-989-758-731-3; ISSN: 2184-4305

859

(Huang and Sheng, 2024).

Ensuring optimal performance of networked med-

ical device setups is critical, as it directly affects

the timeliness and quality of healthcare delivery pro-

vided by IoMT systems. A number of factors, in-

cluding data quality, energy efﬁciency, dependability,

and transfer speed, signiﬁcantly affect how effective

IoMT solutions are.

1.1 Problem Statement

Even though there have been many advancements in

IoMT and its related communication technologies, a

signiﬁcant number of challenges still remain, particu-

larly in the area of ECG heart rate monitoring. Data

loss, latency, and throughput issues may jeopardize

the quality and reliability of vital sign monitoring,

resulting in delayed or inaccurate medical interven-

tions. Even minor data loss during ECG monitoring

may result in the oversight of crucial events such as

arrhythmias or other cardiac problems. High latency

in data transmission may result in delays in detecting

rapid changes in heart rate or rhythm, which is critical

in emergency situations. Inadequate throughput may

limit the frequency of ECG signal sampling, lowering

the granularity of heart rate data and perhaps missing

essential short events (Kwon et al., 2018).

While there are many interesting researches which

have sought to address various aspects of IoMT per-

formance, there remains a critical need for solutions

that simultaneously achieve minimal data loss, desir-

able sampling frequency, ultra-low latency, and high

throughput in ECG heartrate monitoring applications.

1.2 Purpose and Objectives

The objectives of this paper are to address the iden-

tiﬁed issues by developing and evaluating an IoMT

solution which is designed for ECG heart rate moni-

toring:

• To compare the performance of the proposed

setup across different communication technolo-

gies (BLE, Wi-Fi, 5G, and UART);

• To evaluate impact of sampling frequency on per-

formance of the system.

The rest of this paper is organized as follows: Section

2 reviews related work. Section 3 describes the pro-

posed system setup. Section 4 presents the evaluation

results and analysis. Finally, conclusion, recommen-

dations and future work in Section 5.

2 RELATED WORKS

The Internet of Medical Things (IoMT) has garnered

signiﬁcant attention in recent years due to its poten-

tial to revolutionize healthcare delivery. This section

provides an overview of existing research on IoMT,

network communication technologies, and the analy-

sis of performance in networked IoT/IoMT systems.

IoMT research has expanded rapidly, covering

various aspects of healthcare technology integration.

(Dimitrov, 2016) provided a comprehensive overview

of IoMT, highlighting its potential to improve patient

outcomes and reduce healthcare costs. The study em-

phasized the importance of interoperability and data

security in IoMT systems identifying the fact that

ensuring seamless communication between diverse

medical devices and systems remains a signiﬁcant

challenge. Building on this foundation, (Alshehri and

Muhammad, 2020) proposed a framework for IoMT

that addresses key challenges in data collection, trans-

mission, and analysis. Their work highlighted the

need for efﬁcient communication protocols and robust

data analytics in IoMT applications.

Most research on network communication tech-

nologies for IoMT has focused on optimizing perfor-

mance for medical applications. Several key tech-

nologies have emerged as prominent in IoMT re-

search. (Al-Shareeda et al., 2023) conducted a com-

prehensive review of BLE applications in health-

care, noting its advantages in power efﬁciency and

widespread adoption in consumer devices. (Vellela,

2024) explored an IoT-based framework for patient

monitoring in intensive care units, addressing chal-

lenges related to quality of life of patients. (Ahad

et al., 2019) discussed the potential of 5G in revolu-

tionizing IoMT, particularly in enabling real-time re-

mote monitoring and telesurgery applications. While

less prominent in recent literature, UART remains rel-

evant in certain IoMT applications. (Deb et al., 2022)

demonstrated its use in a low-cost ECG monitoring

system.

Performance analysis of networked IoT and IoMT

systems has been a critical area of research. (Vis-

maya et al., 2024) proposed a comprehensive 5G-

based framework for IoT network performance anal-

ysis, considering security issues and long-range pa-

tient monitoring and care. Focusing speciﬁcally on

healthcare, (Rahmani et al., 2018) presented a fog

computing-based approach for analysing and optimiz-

ing IoMT network performance.

Several studies have proposed frameworks and ap-

proaches for measuring IoT network performance,

with potential applications in IoMT. (Arafat et al.,

2024) developed a Quality of Service (QoS) aware

HEALTHINF 2025 - 18th International Conference on Health Informatics

860

Figure 1: Setup of transmission of ECG data through a multi-protocol Gateway to Server.

framework for IoMT networks, which is adapted for

smart healthcare. (Razzaque et al., 2016) reviewed

requirements for middleware IoT, many of which are

applicable to IoMT. (Varga et al., 2020) presented

a comprehensive review of latency in IoT networks,

with a focus on industrial IoT 5G-enabled applica-

tions. (Hameed and Koo, 2024)explored the impact of

active reconﬁgurable intelligent surfaces to improve

throughput optimization in IoT networks. These re-

lated works highlight the dynamic nature of IoMT

research and the ongoing efforts to improve network

performance for medical applications. While signif-

icant progress has been made, there remain ample

opportunities for innovation in addressing the unique

challenges posed by IoMT systems.

3 PROPOSED SYSTEM

Figures 1 and 2 depict the architecture and block di-

agram of the setup respectively. The proposed sys-

tem consists of sensor nodes for measuring vital data

of patients (i.e., ECG). In the experimental setup, an

ECG sensor is attached to the vital positions of a per-

son to read the heart rate signals. These signals are

transmitted through microcontroller boards that are

capable of either wired or wireless communication.

This is followed by a wired or wireless transmission

of the sensor data to the multi-protocol gateway for

processing of the data and forwarding from the gate-

way to a central server which houses the central data

storage. Visualization of stored data and live data

from ongoing measurements can be viewed on the

web interface and also used for further analysis.

As shown in Table 1, the proposed model con-

sists of a setup with AD8232 ECG sensor electrodes

placed on a patient.

Wired / Wireless Sensor Node: Selected micro-

controller units (MCU) with different communication

technologies (UART (Arduino Nano Every), Blue-

tooth BLE (Arduino Nano 33 BLE), Wi-Fi (ESP8266)

via a wireless access point), are used to send the ECG

data to the gateway.

Multi-protocol Gateway: Raspberry Pi 5 is used as

the gateway which is connected to RM520N-GL 5G

USB TO M.2 B KEY dongle to allow for 5G commu-

nication between the gateway and server.

Server System: Another Raspberry Pi 5 connected

to an RM520N-GL 5G USB TO M.2 B KEY dongle,

or a computer connected to the 5G network acts as

the server to receive and process the data from the

gateway and display the real time heart rate signals

on a web user interface.

The sensor data transmission is evaluated mainly

based on the latency using sampling frequencies

(300Hz and 1000Hz) as input parameter. Ensuring

minimal data loss, maintaining desirable sampling

frequency (>100Hz), and achieving ultra-low latency

are crucial for accurate heart rate monitoring (Kwon

et al.,2018). For this reason, the 300Hz and 1000Hz

were chosen to test how the setup can handle the ECG

data transmission and guarantee these sampling rates

without data loss or low throughput.

4 EVALUATIONS

As shown in Table 2, the experiment was done in two

variations with ﬁxed sampling frequencies of 300Hz

and 1000Hz and varied distances. This is to sys-

Comparison of Bluetooth Low Energy (BLE), Wi-Fi, Serial and 5G in IoMT

861

Figure 2: Block diagram of transmission of ECG data from sensor node to server.

Table 1: System Components and Speciﬁcations.

Component Hardware Conﬁg Purpose

ECG Sen-

sor

AD8232 12-bit

ADC

Data ac-

quisition

Micro

controller

Arduino

Nano

Every,

ESP8266

NodeMCU,

Arduino

Nano 33

BLE

ATMega

4809

micro-

controller,

Wi-Fi:

IEEE

802.11

b/g/n,

Bluetooth

5.0

Serial,

Wi-Fi,

BLE

transmis-

sion

5G Device RM520N-

GL 5G

USB TO

M.2 B

KEY

dongle

5G Sub-6

GHz mod-

ule

5G trans-

mission

Gateway Raspberry

Pi 5

Multi-

protocol

support

and data

process-

ing

Sensor

data trans-

mission

Server Raspberry

Pi 5

Data stor-

age and

analysis

Analysis

and stor-

age

tematically evaluate the communication protocols at

different stages of the IoT healthcare system. This

approach isolates and analyses performance charac-

teristics of both edge (sensor node to gateway) and

backhaul (gateway to server) communications inde-

pendently. Table 3 shows the speciﬁcations of the

test environment. The experiment was done indoors

in a lab with 5G antenna and a 5GHz Wi-Fi router

mounted at the same place to the wall 3.7m from the

ground. the room is divided into 3 partitions by partial

concrete walls (2m height).

The ﬁrst part of testing involved sensor node to

gateway edge communication where the testing pro-

tocols used were UART (Arduino Nano Every), Wi-

Fi (ESP8266) and BLE (Arduino Nano 33 BLE). The

test conﬁgurations in this part were made up of the

following distance variations as test points: (a) Base-

line measurement: ﬁxed minimal distance, which is

the starting point, (b) Measurement without obstacles:

Table 2: Experimental Parameters and Settings.

Parameter Values Description

Distances 0.1-17m Multiple

points

Sampling

Rates

300Hz,

1000Hz

Fixed inter-

vals

Duration 10 minutes Per conﬁgura-

tion

Environment Indoor con-

trolled

Temperature:

22±2°C

Distances: 1m, 2m, 5m, 10m and (c) Measurement

with obstacles (Wall): Distances: 2.4m, 3.4m, 5.4m,

6.4m

The second part of the experimental tests was

gateway to server backhaul communication, both

Raspberry Pi 5 devices with in-built Wi-Fi and con-

nected to RM520N-GL 5G USB TO M.2 B KEY don-

gle as test protocols. The ECG sensor was connected

to Arduino Nano Every to the gateway via UART for

this test. Distance test points measured were

Test points: (a) Partition A: 6m (facing 5G an-

tenna and Wi-Fi router), 9m (left), 10m (right), (b)

Partition B: 7.5m (facing 5G antenna and Wi-Fi

router) and (c) Partition C: 17m (facing 5G antenna

and Wi-Fi router)

Table 3: Test Environment Speciﬁcations.

Parameter Conﬁguration

Room Dimensions 25m x 25m

Antenna / Wi-Fi Router Height 3.7m

Wall Material Concrete

Internal Partitions Concrete

For each communication protocol, each distance,

and each sampling rate, other data such as, times-

tamps, sample indices for each ECG data, is gen-

erated alongside the ECG sensor data to help track

the performance of the IoMT devices. In order to

determine data loss, throughput and latency perfor-

mance of the system, the communication devices at-

tach sequence numbers to each ECG data point gener-

ated with the timestamp each data point was received.

This is then transmitted in bulks of 50 to the gateway.

The gateway periodically pings the sensor node and

records the time it receives a response to determine

the roundtrip time. Same process happens between

the gateway and the server devices.

HEALTHINF 2025 - 18th International Conference on Health Informatics

862

The data loss rate is calculated by comparing the

total expected samples from the sensor node with

what the gateway received (gateway sequence) vs

what the server received (server sequence) in the

dataset. The difference, divided by the total expected

and multiplied by 100, gives the percentage of data

loss. Latency is calculated for different stages:

• Sensor Node to Gateway (SnG): Time difference

between sensor node timestamp and Gateway re-

ceived time. T.response: Time of response receipt

and T.request: Time of initial request;

RT T.SnG = T.response − T.request (1)

• b. Gateway to Server: Uses the Roundtrip (RTT)

Latency from the gateway data, where T.proc:

Server processing delay, T.Sresp: Server response

time and T.Greq: Gateway request time;

RT T.GS = (T.Sresp − T.Greq) − T. proc (2)

• End-to-End Latency (E2E): Time difference be-

tween Arduino timestamp and Server processed

time;

E2E = RT T.SnG + RT T.GS (3)

5 ANALYSES

Tables 4 and 5 depict the comparisons after analysis is

done on the collected data. First, we observed no data

losses for experiments. This is due to implementation

of techniques such as connection intervals speciﬁcally

for BLE communication and sending the data in bulk

and in batches to ensure all ECG sensor data gener-

ated are transmitted by the communication devices.

5.1 Edge Communication Performance

(Sensor Node to Gateway Analysis)

For the roundtrip time measurement without obsta-

cles, serial communication shows the lowest latency

(3.96ms at 300Hz, 4.37ms at 1000Hz), which is ex-

pected as it’s a direct connection. BLE and Wi-Fi

show higher latencies, with Wi-Fi generally having

slightly higher latencies than BLE. Wi-Fi is about

1.5ms higher than serial and BLE is about 1.3ms

higher than serial.

The 1000Hz conﬁgurations generally show

slightly higher latencies than the 300Hz conﬁgu-

rations especially at longer distances. Generally,

latency increases with distance for both 300Hz and

1000Hz frequencies. The presence of obstacles tends

to increase latency compared to no obstacles. The

Table 4: Communication Latency Comparison at Different

Distances Without Obstacles and Sampling Rates.

Distance Serial (ms) Wi-Fi (ms) BLE (ms)

300Hz 300Hz 300Hz

Starting

point

3.96 5.48 5.24

1m N/A 6.06 5.45

2m N/A 6.47 6.02

5m N/A 8.07 7.48

10m N/A 10.51 10.04

Distance Serial (ms) Wi-Fi (ms) BLE (ms)

1000Hz 1000Hz 1000Hz

Starting

point

4.37 5.74 4.86

1m N/A 6.21 5.22

2m N/A 6.20 6.04

5m N/A 8.44 7.97

10m N/A 10.99 9.78

highest latencies are observed at the greatest dis-

tances (10m), particularly with obstacles present. The

table compares end-to-end latency across different

methods and frequencies:

• Wi-Fi has the highest average latency, followed

closely by BLE, while serial communication has

the lowest latency;

• For Wi-Fi and BLE, the 1000Hz frequency shows

slightly higher latency than 300Hz, but the differ-

ence is small;

• Serial communication shows the least variation in

latency between frequencies;

Table 5: Wi-Fi and BLE Latency Comparison at Different

Distances With Obstacles and Sampling Rates.

Distance

300Hz (ms) 1000Hz (ms)

Wi-Fi BLE Wi-Fi BLE

2.4m 7.97 7.24 7.59 7.73

3.4m 8.44 7.91 8.43 7.59

5.4m 9.61 9.07 9.43 8.83

6.4m 10.22 9.65 10.74 9.64

Like BLE, Wi-Fi latency generally increases with

distance. Obstacles tend to increase latency compared

to no obstacles. There’s high variability in latency,

especially at longer distances. The 1000Hz frequency

often shows higher latency than 300Hz, particularly

with obstacles. The highest latencies are observed at

6.4m and 10m distances.

5.1.1 Baseline Performance Analysis

The initial testing at the starting point of measurement

revealed distinct performance characteristics for each

protocol. Serial communication established the base-

Comparison of Bluetooth Low Energy (BLE), Wi-Fi, Serial and 5G in IoMT

863

line for optimal performance, achieving a mean la-

tency of 3.96ms (±0.06ms at 95% conﬁdence inter-

val) at 300Hz sampling rate. This performance saw

a slight degradation to 4.37ms (±0.07ms) when in-

creasing to 1000Hz sampling rate, representing only

a 10.4% increase despite more than tripling the data

rate. Bluetooth Low Energy demonstrated remark-

able performance for a wireless protocol, with mean

latencies of 5.24ms (±0.09ms) at 300Hz and 4.86ms

(±0.08ms) at 1000Hz. Notably, BLE showed im-

proved performance at the higher sampling rate, sug-

gesting effective packet bundling and transmission

optimization. WiFi performance established the base-

line for IP-based communication, with mean latencies

of 5.48ms (±0.11ms) at 300Hz and 5.74ms (±0.12ms)

at 1000Hz. The relatively small latency increase of

4.7% between sampling rates indicates robust scala-

bility for higher data rates.

5.1.2 Distance Impact Analysis

The impact of distance on the wireless protocols re-

vealed some important patterns for healthcare deploy-

ment planning. BLE demonstrated a linear degra-

dation rate of approximately 0.48ms per meter (R²

= 0.982) at 300Hz sampling rate. This predictable

degradation allows for reliable performance estima-

tion in hospital environments. WiFi showed a similar

linear pattern but with a steeper degradation rate of

0.503ms per meter (R² = 0.975) at 300Hz. The differ-

ence in degradation rates became more pronounced at

1000Hz sampling rate where BLE had 0.492ms/m (R²

= 0.978) and WiFi had 0.525ms/m (R² = 0.971)

5.1.3 Obstacle Effects

The introduction of concrete walls created signif-

icant but predictable impacts on wireless perfor-

mance. At 2-meter distance BLE latency increased by

20.3% (±1.2%) and WiFi latency increased by 23.2%

(±1.4%). This difference in obstacle impact becomes

particularly relevant for hospital deployments where

multiple walls may separate sensors from gateways.

5.2 Backhaul Communication

Performance (Gateway to Server

Analysis)

Table 6 shows the performance comparison between

Wi-Fi and 5G and these are the key observations: As

observed in the ﬁrst set of experiments, both Wi-Fi

and 5G show excellent performance with no observ-

able data loss across all distances and sampling rates.

Both technologies maintain the target sampling rates

(300Hz and 1000Hz) with no data loss. There’s no

signiﬁcant difference between Wi-Fi and 5G in terms

of maintaining the desired sampling rate.

Table 6: Wi-Fi and 5G Latency Comparison at Different

Distances.

Distance Wi-Fi (ms) 5G (ms)

300Hz 1000Hz 300Hz 1000Hz

6m 78.64 44.32 204.53 213.64

7.5m 60.16 47.82 181.21 163.53

9m 65.19 62.02 172.75 169.39

10m 66.23 44.84 187.91 160.68

17m 43.48 54.16 290.63 202.81

End-to-end round trip time latency measurements

gave the following results. At 300Hz sampling fre-

quency, Wi-Fi had an initial high latency (78.64ms)

at 6m which showed at 7.5m to 60.16ms. It then be-

came relatively stable (60ms - 66ms) at 9m and 10m

distances. Its best performance was at maximum dis-

tance (43.48ms at 17m). At 1000Hz sampling fre-

quency, it had a consistent performance range (44-

62ms) with more stable latency across the distances.

The average RTT improvement of 39.8 percent com-

pared to 300Hz.

5G also had moderate latency (204.53ms) at 6.0m

initially at 300Hz sampling frequency. It could be

observed that there was a progressive improvement

until 9m (172.75ms) but performed quite poorly be-

yond 10m with a signiﬁcant decline in performance

(290.63ms) at maximum distance. At 1000Hz sam-

pling frequency, higher initial latency (213.64ms) was

recorded at 6m which improved with distances 7.5m

to 10m. It had 30.2 percent lower RTT at 17m com-

pared to 300Hz.

Wi-Fi consistently outperforms 5G in terms of

Gateway-Server RTT across all distances and sam-

pling rates. Wi-Fi RTT remains relatively stable

(mostly under 70ms) even at longer distances. 5G

shows higher RTT values (160ms-290ms) and more

variation with distance.

Increasing the sampling rate from 300Hz to

1000Hz doesn’t signiﬁcantly impact the performance

of either technology in terms of data loss or through-

put. There’s a slight increase in Arduino-Gateway

RTT for both technologies at 1000Hz, but it’s mini-

mal. Wi-Fi maintains consistent performance across

different distances. 5G shows more variation in

Gateway-Server RTT as distance increases, particu-

larly noticeable at 17m.

5.2.1 Protocol Comparison

The backhaul testing revealed substantially different

performance characteristics compared to edge com-

HEALTHINF 2025 - 18th International Conference on Health Informatics

864

munication. At the 6-meter baseline position the

average performance of 5G was 300Hz: 204.53ms

(±3.21ms) and 1000Hz: 213.64ms (±3.45ms), whiles

average Wi-Fi performance was 300Hz: 78.64ms

(±1.89ms) and 1000Hz: 44.32ms (±1.56ms)

6 DISCUSSION

6.1 Clinical Implications

The performance characteristics revealed in this ex-

periment have some implications for different types

of medical monitoring scenarios. It is, therefore, im-

portant to understand these implication to help with

protocol selection for speciﬁc healthcare applications.

6.1.1 Critical Care Applications

In critical care settings, where immediate detection of

life-threatening arrhythmias is essential, these results

indicate that protocol selection can signiﬁcantly im-

pact the reliability of the system. For instance, in ven-

tricular ﬁbrillation detection, where every millisecond

impacts survival rates, the measured latency differ-

ences become clinically signiﬁcant.

The consistent communication of UART of sub-

4ms latency makes it ideal for bedside monitoring

equipment where physical connections are feasible.

The average latency of 3.96ms ensures that rhythm

analysis algorithms receive data with minimal delay,

which can be crucial for real-time detection of dan-

gerous arrhythmias.

BLE’s performance (5.24ms average) makes it a

viable wireless alternative for critical care applica-

tions, particularly important for maintaining mobility

of the patient while ensuring timely data transmission.

The stability of BLE performance, which is indicated

by its low coefﬁcient of variation (3.44%), provides

the reliability necessary for critical care monitoring.

6.1.2 General Ward Monitoring

For general ward monitoring, where patients require

continuous observation but with quite ﬂexible tim-

ing requirements, these results obtained support more

ﬂexible protocol selection. The measured Wi-Fi la-

tencies (5.48ms to 10.51ms depending on distance)

fall well within acceptable ranges for routine vital

sign monitoring, where updates every 50-100ms are

typically sufﬁcient.

6.1.3 Remote Monitoring Considerations

The backhaul communication results obtained indi-

cates suitability for remote monitoring implemen-

tations. The measured 5G latencies (204.53ms to

290.63ms) prove suitable for non-critical remote

monitoring applications while requiring careful con-

sideration for any critical care implementations re-

quiring real-time response.

6.2 Implementation Recommendations

Based on our ﬁndings, we propose the following im-

plementation strategies for different healthcare sce-

narios:

For critical care environments, we recommend

for primary monitoring, serial connections are used

where possible, because provided the most reliable

and lowest latency data transmission. For mobile

monitoring, BLE implementations with redundant re-

ceivers may be used to maintain coverage. For max-

imum distance, maintaining BLE sensor-to-gateway

distances under 5m can ensure latency remains be-

low 7.5ms. And for sampling rate, 300Hz provides

optimal balance of data resolution and system perfor-

mance.

For general ward monitoring, our results suggest

Wi-Fi is suitable to be used as a primary protocol

for its broader coverage and acceptable latency char-

acteristics. For access point placement, a maximum

10m separation from the gateway to maintain sub-

10ms latency is suitable. And for sampling rate, both

300Hz and 1000Hz are viable depending on the spe-

ciﬁc monitoring needs.

7 CONCLUSION

This paper investigated the performance characteris-

tics of different data transmission methods (Serial,

Wi-Fi, and BLE comparison and Wi-Fi and 5G com-

parison) for an ECG monitoring system. We ﬁnd that

for the Serial, Wi-Fi, and BLE comparison, all the

communication technologies maintain data integrity

with no data loss. The system achieves expected

throughput based on the target sampling frequencies

accurately. Latencies are within expected ranges, with

serial being fastest, followed by BLE, then Wi-Fi. For

the Wi-Fi and 5G comparison, the experiment demon-

strates that while both Wi-Fi and 5G are viable op-

tions for ECG monitoring systems, their optimal use

cases differ. At distance 0.1 to 1m, the best options

are Serial (3.96ms) or BLE (4.86ms) which consis-

tently gave low latency. This is critical for real-time

Comparison of Bluetooth Low Energy (BLE), Wi-Fi, Serial and 5G in IoMT

865

monitoring which can be suitable for Critical Care

Units. Wi-Fi and 5G are best suited for mid-range

to long-range monitoring.

There were some limitations of this experiment

which has help to suggest the directions for future

research. Our testing environment, while controlled,

represents a simpliﬁed version of actual hospital en-

vironments. Speciﬁc limitations include limited inter-

ference sources compared to active hospital environ-

ments, single-story testing versus multi-ﬂoor hospital

scenarios, and absence of dynamic obstacles (moving

equipment and people).

The technical aspect of the experiment also had

some constraints including RTT-based latency mea-

surements versus true end-to-end timing and limited

number of simultaneous devices tested. While our re-

sults align with theoretical requirements for medical

monitoring, additional validation would strengthen

the clinical relevance such as testing with standard-

ized ECG datasets from PhysioNet, validation in ac-

tive hospital environments, and long-term stability as-

sessment in clinical settings.

In future work, we plan to implement a corre-

sponding digital twin of this physical setup and a

probe to properly monitor the performance and par-

ticularly to obtain a more accurate latency calculation

of the system. There will also be power consump-

tion analysis and testing of the characteristics of ECG

signals during rest vs exercise (for wearable devices)

to determine how both may affect performance of the

system. And to investigate performance under higher

network load conditions.

ACKNOWLEDGEMENTS

This research is funded by the German Federal

Ministry for Digital and Transport (reference no.:

45FGU111E).

REFERENCES

Ahad, A., Tahir, M., and Yau, K. L. (2019). 5g-based

smart healthcare network: Architecture, taxonomy,

challenges and future research directions. IEEE Ac-

cess, PP:1–1.

Al-Shareeda, M., Ali, M., Manickam, S., and Karuppayah,

S. (2023). Bluetooth low energy for internet of things:

review, challenges, and open issues. Indonesian Jour-

nal of Electrical Engineering and Computer Science,

31:1182–1189.

Alshehri, F. and Muhammad, G. (2020). A comprehensive

survey of the internet of things (iot) and edge comput-

ing in healthcare. IEEE Access, PP:1–1.

Arafat, M. Y., Pan, S., and Bak, E. (2024). Qqar: A

q-learning-based qos-aware routing for iomt-enabled

wireless body area networks for smart healthcare. In-

ternet of Things, 26:101151.

Deb, S., Islam, M. R., Robaiatmou, J., and Islam, M. T.

(2022). Design and implementation of low-cost ecg

monitoring system for the patient using smart device.

In International Conference on Inventive Computa-

tional Sciences (ICICCS).

Dimitrov, D. V. (2016). Medical internet of things and big

data in healthcare. Healthcare Informatics Research,

22(3):156–163.

Girolami, M., Mavilia, F., and Delmastro, F. (2020). Sens-

ing social interactions through ble beacons and com-

mercial mobile devices. Pervasive and Mobile Com-

puting, 67:101198.

Hameed, I. and Koo, I. (2024). Enhancing throughput in iot

networks: The impact of active ris on wireless pow-

ered communication systems. Electronics, 13:1402.

Huang, W. and Sheng, G. (2024). Analysis and research

on uart communication protocol. In 2024 4th Asia-

Paciﬁc Conference on Communications Technology

and Computer Science (ACCTCS), pages 768–771.

Kwon, O., Jeong, J., Kim, H. B., Kwon, I. H., Park, S. Y.,

Kim, J. E., and Choi, Y. (2018). Electrocardiogram

sampling frequency range acceptable for heart rate

variability analysis. Healthcare Informatics Research,

24(3):198–206.

Pahlavan, K. and Krishnamurthy, P. (2021). Evolution and

impact of wi-ﬁ technology and applications: A his-

torical perspective. International Journal of Wireless

Information Networks, 28:3–19.

Rahmani, A. M., Gia, T. N., Negash, B., Anzanpour, A.,

Azimi, I., Jiang, M., and Liljeberg, P. (2018). Exploit-

ing smart e-health gateways at the edge of healthcare

internet-of-things: A fog computing approach. Future

Generation Computer Systems, 78:641–658.

Razzaque, M. A., Milojevic-Jevric, M., Palade, A., and

Clarke, S. (2016). Middleware for internet of things:

A survey. IEEE Internet of Things Journal, 3(1):70–

95.

Varga, P., Peto, J., Franko, A., Balla, D., Haja, D., Janky,

F., and Kozma, G. (2020). 5g support for industrial

iot applications—challenges, solutions, and research

gaps. Sensors, 20(3):828.

Vellela, S. S. (2024). Iot based icu patient monitoring sys-

tem. International Journal for Modern Trends in Sci-

ence and Technology, 10.

Vismaya, O., Kumar, A., and Syethima, K. (2024). 5g en-

abled iot network and security in healthcare.

HEALTHINF 2025 - 18th International Conference on Health Informatics

866