sensors

Review
A Survey of Autonomous Vehicles: Enabling Communication
Technologies and Challenges
M. Nadeem Ahangar 1 , Qasim Z. Ahmed 1, *, Fahd A. Khan 2 and Maryam Hafeez 1

1 School of Computing and Engineering, University of Huddersfield, Huddersfield HD1 3DH, UK;
m.ahangar@hud.ac.uk (M.N.A.); M.Hafeez@hud.ac.uk (M.H.)
2 School of Electrical Engineering and Computer Science, National University of Sciences and Technology,
Islamabad 44000, Pakistan; fahd.ahmed@seecs.edu.pk
* Correspondence: Q.Ahmed@hud.ac.uk

Abstract: The Department of Transport in the United Kingdom recorded 25,080 motor vehicle fatali-
ties in 2019. This situation stresses the need for an intelligent transport system (ITS) that improves
road safety and security by avoiding human errors with the use of autonomous vehicles (AVs). There-
fore, this survey discusses the current development of two main components of an ITS: (1) gathering
of AVs surrounding data using sensors; and (2) enabling vehicular communication technologies.
First, the paper discusses various sensors and their role in AVs. Then, various communication tech-
nologies for AVs to facilitate vehicle to everything (V2X) communication are discussed. Based on the
transmission range, these technologies are grouped into three main categories: long-range, medium-
range and short-range. The short-range group presents the development of Bluetooth, ZigBee and
ultra-wide band communication for AVs. The medium-range examines the properties of dedicated
short-range communications (DSRC). Finally, the long-range group presents the cellular-vehicle to
everything (C-V2X) and 5G-new radio (5G-NR). An important characteristic which differentiates



each category and its suitable application is latency. This research presents a comprehensive study of


AV technologies and identifies the main advantages, disadvantages, and challenges.
Citation: Ahangar, M.N.; Ahmed,
Q.Z.; Khan, F.A.; Hafeez, M. A Survey Keywords: autonomous vehicles; Bluetooth; dedicated short-range communications; intelligent
of Autonomous Vehicles: Enabling transport system; V2I; V2V; V2X; VANET; vehicular communications; ultra-wide band; ZigBee
Communication Technologies and
Challenges. Sensors 2021, 21, 706.
https://doi.org/10.3390/s21030706

1. Introduction
Academic Editor: Fatih Kurugollu and
Francisco J. Martinez Technology facilitates humans, improves productivity and leads to a better quality
Received: 18 December 2020 of life. Technological developments and automation in vehicular networks will lead to
Accepted: 18 January 2021 better road safety and lower congestion in present urban areas where the traditional
Published: 21 January 2021 transport system is becoming increasingly disorganised and inefficient [1]. Therefore, the
development of the intelligent transport systems (ITS) concept has been proposed, with the
Publisher’s Note: MDPI stays neu- aim and focus on improving traffic safety and providing different services to its users [2,3].
tral with regard to jurisdictional clai- There has been considerable research in ITS resulting in significant contributions (see [4]
ms in published maps and institutio- and references therein). Figure 1 shows the three main applications of the ITS system: (1)
nal affiliations. Applications for transport efficiency; (2) Safety-critical applications; and (3) Infotainment
applications. The main task of the transport efficiency application includes the calculation
of the optimal speed and the route to navigate the vehicle to the destination by taking into
Copyright: © 2021 by the authors. Li-
account the traffic information [5]. Safety-critical applications deal with the response to
censee MDPI, Basel, Switzerland.
emerging vehicles and intersection collision avoidance [5]. Infotainment applications deal
This article is an open access article
with cooperative local services such as the location of petrol stations, hotels, etc. [5].
distributed under the terms and con- One of the most significant contributions towards the ITS has been the development
ditions of the Creative Commons At- of autonomous vehicles (AVs). These vehicles perceive the environment through sensing
tribution (CC BY) license (https:// using various sensors, and then this information is utilised to drive without the need for
creativecommons.org/licenses/by/ any human intervention [6]. In Figure 2, the technical evolution of AVs is illustrated in
4.0/). detail. The modern cruise control system was developed in 1948 and since then self-driving

Sensors 2021, 21, 706. https://doi.org/10.3390/s21030706 https://www.mdpi.com/journal/sensors

Sensors 2021, 21, 706 2 of 33

cars have been gradually developing different features [7,8]. Significant developments
were made in the early 2000s when the lane departure warning system (LDWS), adaptive
cruise control (ACC), self-parking assistance (SPA), auto-pilot and traffic sign recognition
(TSR) were developed for AVs [9–11]. It is planned that by 2022 we will have intelligent
speed adaptation systems and by 2030 fully automated AVs will be available with no
driver backup required. This advancement will further lead to the next emerging and
evolutionary stage of the Internet of vehicles (IoV). In IoV networks, besides human as
users of the Internet, physical sensors, machine-type devices and vehicles will be the users
of the IoV ecosystem.

Figure 1. Different applications for intelligent transport system (ITS).

Figure 2. Technical evolution in autonomous cars.

The Society of Automotive Engineers (SAE) classified the vehicles into six different
levels of automation. The same classification is also adopted by other organisations
such as the International Organisation of Motor Vehicle Manufacturer (OICA) and the
German Federal Highway Research Institute (BASt). The US National Highway Traffic
Safety Administration (NHTSA) initially had a different classification, as compared in
Table 1 [12]. Since then, NHTSA has also adopted SAE’s six levels of automation in the
Federal Automated Vehicle Policy. Levels 0–2 can be broadly categorised as driver assisted,
while Levels 3 and 4 can be termed as semi-automatic and Level 5 is fully autonomous.
The detailed features of each level are described below:
• Level 0 is not automated, and all the tasks are performed by the driver.
• Level 1 provides driver assistance with some functions such as ACC, TSR, etc., but
the driver controls the accelerator and the brakes while monitoring the surroundings.
• Level 2 provides partial driving assistance, and the vehicle can perform the steer-
ing and acceleration functions. However, the driver is responsible for many safety-
critical actions.
• Level 3 provides conditional driving automation where the vehicle performs the
entire monitoring of the surroundings. The driver is no longer responsible for safety-
critical issues.
Sensors 2021, 21, 706 3 of 33

• Level 4 provides high driving automation, and the driver holds control only if the
automated situation turns unsafe. Steering, braking, acceleration and surrounding
check are performed by the vehicle.
• Level 5 represents complete automation where there is no need for human intervention
and the driver is a passenger.
Figure 3 summarises the above discussed features and functions achieved at each
level according to the SAE J3016 standard [12].

Table 1. Comparison between different automation levels.

Organisation Level 0 Level 1 Level 2 Level 3 Level 4 Level 5

BASt, SAE, Driver Driver Partial Conditional High Full
OICA Only Assisted Automation Automation Automation Automation
NHSTA No Function Combined Limited Fully
Specific Function Self-Driving Self-Driving

Figure 3. Different levels of automation.

Autonomous vehicles have several advantages including improved safety and lower
road congestion resulting in lower fuel/energy consumption [4,6]. Besides these advan-
tages, there are some issues which need to be resolved for AVs such as who bears the legal
responsibilities of the AVs, what will the course of action be if the AV controller is hacked,
etc. A summary of the main pros and cons of AVs are listed in Table 2. Overall, it can be
claimed that, if the drawbacks are resolved or minimised, then autonomous cars will be a
considerable technological development in coming years as they will facilitate people’s
lives and increase road safety [10,13].
This paper aims to review the advancement of autonomous cars focusing on the
various sensors deployed in them along with the different communication technologies
employed. Therefore, the objectives of this survey are as follows:
• To identify and describe the different type of sensors used in AVs.
• To classify communication technologies based on their transmission range and identify
their main advantages and disadvantages. This implies:
1. Discussing short-range technologies such as Bluetooth, ZigBee and ultra-wide
band (UWB)
2. Assessing medium-range technologies such as dedicated short-range communi-
cation (DSRC) and Wireless Fidelity (Wi-Fi)
3. Analysing long-range technologies such as Cellular Vehicle To Everything (C-
V2X) and Fifth Generation-New Radio (5G-NR) technology
The survey is organised as follows. Section 2 presents the literature review. Section 3
discusses the system architecture for the AVs with emphasis on the various types of
sensors. Section 4 analyses the Vehicular Ad-hoc Networks (VANETs) and the Vehicle-
to-Everything (V2X) technologies. Section 5 discusses the classification of the different
wireless technologies based on the transmission range and their applications and challenges
Sensors 2021, 21, 706 4 of 33

in AVs and ITS. Finally, in Section 6, the conclusions and future research directions, along
with the decision and planning strategies are presented.

Table 2. Advantages and disadvantages of AVs.

Advantages Disadvantages
Casualty: AVs can significantly reduce the Law: The definition of legal responsibili-
number of accidents. ties can hinder the implementation of AVs.
Fewer Expenses: Precise autonomous driv- Threat: AVs can be more vulnerable to
ing can reduce fuel consumption and in- network hacks because of the present
crease the conservation of other parts. computer-controlled functions.
Productivity: The journey can be produc- Employment: There will be many job
tive by performing other activities than losses due to AVs in the transportation sec-
driving. tor.
Comfort: Interiors of AVs can be comfort- Price: The price of AVs is initially high, but,
able and spacious. after greater adoption, the price is going to
decrease.

2. Literature Review
The focus of the paper is to cover the number of sensors employed for AVs and
the virtual server for data transmission. Most applications for ITS require a multitude
of sensors, which assist automation and communication. These are mounted within a
vehicle or on the network infrastructure. Depending on the transmission range of the
wireless technology, the sensors are divided into three main categories, namely short-range,
medium-range and long-range. Each category has a separate communication latency and,
thus, is deployed in different scenarios. A summary of the wireless technologies, their
categorisation and use cases in existing literature are referenced and discussed in Table 3.
In [14], the architecture design and implementation of an AV is discussed, and, in [15],
various strategies of multi-sensor fusion are discussed. The recent developments and
advancements in the perception and sensor technologies for AVs are presented in [16].
In [17], multiple-target and multiple-source are combined to form a framework for on-
board sensors. In [1], a five-layer architecture for AVs is discussed. However, none of
these survey papers discuss the key wireless technologies for data communication. In [18],
resource allocation schemes for DSRC and C-V2X are discussed. However, both these
technologies are suitable only for medium- and long-range communication and lack low-
latency application.
This paper contributes to the current state-of-the-art by providing a holistic review of
the key enabling wireless technologies for ITS. It not only identifies the main aspects of net-
work architecture behind AVs, but it also analyses the evolution of the technological trends,
the challenges and the constraints that AVs must address before being fully implemented.
The paper links abstract technological concepts with the expected real implications that AVs
can have in human mobility, transportation systems and infrastructure. Most of the recent
papers address these issues separately and they focus on technological innovations, adop-
tion or regulatory issues. Therefore, this paper tries to gather the conceptual frameworks
with the practical aspects by illustrating the main architecture concepts and compiling
the outcomes of some of the latest research about VANET technologies, including their
applications, advantages and limitations.
For short-range communications where transmission range is less than 25 m, four
technologies, namely, Bluetooth [19], Bluetooth low energy (BLE) [20], ZigBee [21] and
UWB [22], have been proposed. Several papers have been published which compare these
wireless technologies, evaluating their main features, such as power consumption, data
rates and transmission time (see, e.g., [23,24] and references therein).
Sensors 2021, 21, 706 5 of 33

For medium-range communication, where transmission range is between 25 and 100 m,

WiFi and DSRC are useful wireless technologies [25,26]. However, IEEE 802.11p/DSRC
is the prevailing technology for vehicular communications as it supports high-mobility.
In addition, another benefit of the IEEE 802.11p/DSRC is the outside-the-context of a
basic service set (OCB) mode [27,28]. The OCB mode allows communication without
authentication/association and as a result offers lower latency which is necessary for
safety-critical applications. However, this may result in severe security issues [27].
Finally, for the long-range communication, from a 100 m to 5 km, Third Generation
Partnership Project’s (3GPP’s) long-term evolution (LTE)-Advanced (LTE-A) pro with its
Cellular-V2X capabilities and 5G-NR have emerged as the key technologies [29–33]. These
technologies provide ubiquitous coverage and high data rates and can support low-to-
medium latency requirement. However, they are not suitable for safety-critical applications
because the latency varies between 50 and 80 ms depending on the cellular traffic load [34].

Table 3. Related work.

Sichitui et al., 2008 [23] The network protocols from the physical to the transport layer are analysed, and the
security aspects for V2V communication are mentioned in detail.

Zeng et al., 2009 [25] DSRC is developed for V2V and V2I communication. Characteristics, mechanism and
the application of DSRC are explained. Security threats and congestion issues are
discussed.

Sevlian et al., 2010 [35] GPS enabled channel sounding system that can predict the multipath and doppler shift
for an adaptive OFDM vehicular communication system is developed. 5.9 GHz results
in serious multipath and doppler shift issues as compared to the results obtained at
700 MHz band.

Dorle et al., 2010 [36] ZigBee technology is combined with infrared sensors to detect and validate the vehicles
on the intersections. They proposed model utilised ZigBee technology to help the users
to recognise and identify the vehicles for security clearance.

Kenney 2011 [26] Development and deployment of DSRC standards (IEEE 1609.x) is carried out in the
United States. Amendments to IEEE 802.11p for wireless access in vehicular
environment (WAVE) is proposed for network services, security and multi-channel
operations.

Eenennaam et al., 2012 [28] Channel switching is proposed by IEEE 1604.4 will allow the DSRC technology to
support multi-channel operations. Analysis on beaconing with the OMNeT++
simulator is presented.

Habib et al., 2013 [29] First time classification based on range such as “short-range”, “medium-range” and
“long-range” is done considering the available communication technologies.

Jo et al., 2014 [37] Distributive system for an AV is proposed which reduces computational complexities
and modularity. AUTOSAR software application system is proposed.

Zheng et al., 2015 [38] DSRC and LTE technologies are combined to produce the heterogenous vehicular
network (HetVNet) model. HetVNet models combines the advantages of both DSRC
and LTE technologies. Characteristics of MAC and network layers are also proposed for
this model.

Cerio et al., 2015 [39] Proposed Bluetooth model is validated by over several road experiments in which the
vehicle is supposed to move at high speeds and establish connection with its
road-side units.
Sensors 2021, 21, 706 6 of 33

Table 3. Cont.

Gao et al., 2016 [40] DSRC experimental demonstration for truck platooning is done. Different dynamic and
static tests revealed that on hilly and curved roads packet delivery ratio is affected.
Lower delivery ratio on straight paths is achieved due to the reflection of the truck with
nearby vehicles.

Abboud et al., 2016 [30] C-V2X and DSRC are combined together to produce a hybrid architecture. It is
concluded that the combination of both technologies can produce useful results in V2X
communications.

Lei et al., 2016 [41] ZigBee technology was tested and implemented for Forward Collision Warning System
(FCWS) in low speed vehicles.

Zanchin et al., 2017 [12] Categorising the levels of automation by explaining each level according to the SAE
standard.

Zong et al., 2018 [14] Perception and planning stages can be improved by implementing two inner loops for
simultaneously localising and mapping. To enlarge the detection and range of the
sensors, they also developed an algorithm to detect and calculate the range of the
system.

Fitah et al., 2018 [42] Demonstrations on various simulators such as SUMO and NS-2.35 to prove the capacity
of DSRC technology. After conducting different experiments they concluded that the
performance of DSRC is much better than the Wi-Fi for medium-range communications.

Valgas et al., 2018 [31] 3GPP calibrated simulator to evaluate the performance of 5G-NR in V2X
communication. An increase in sub-carrier spacing is suggested to boost the system
protection against interference.

Rosique et al., 2019 [43] The physical fundamentals, principal functioning and electromagnetic spectrum of
sensors that are used to gather the external information of the vehicle are presented.
The sensors are analysed on spider charts in order to compare and highlight their
weaknesses.

Sheikh et al., 2019 [44] VANETs security, architecture and challenges are discussed for traffic management
system for vehicular network from malicious nodes and fake messages.

Naik et al., 2019 [32] IEEE 802.11bd and 5G-NR are developed to overcome the disadvantages of DSRC and
C-V2X technology and satisfy the diverse needs for V2X applications.

Chen et al., 2020 [33] Basic road safety, architecture , key technologies and the standards of C-V2X are
introduced. Technical evolution path from LTE-V2X to NR-V2X is covered.

Noor et al., 2020 [18] Fair allocation of resources is of prime importance in wireless networks [45]. Resource
allocation in vehicular networks has been surveyed in [18] and references therein.

3. System Architecture for Autonomous Vehicles

An ordinary vehicle can be converted into an autonomous one by adding some additional
components including sensors that allow the vehicle to make its own decisions by sensing
the environment and controlling the mobility of the vehicle [12,14,37,46]. Figure 4 illustrates
the overall communication process/protocol in AVs and also lists the sensors, actuators,
hardware and the software control required. The protocol architecture, explained below, is
composed of four main stages and enables a Level 5 fully autonomous vehicle where all
the users are passengers.
1. Perception: This stage involves sensing of the AVs surrounding through various
sensors and also detecting its own position with respect to the surroundings. In
this stage, some of the sensors used by the AV are RADAR, LIDAR, camera, real-
time kinetic (RTK), etc. The information from these sensors is then passed to the
Sensors 2021, 21, 706 7 of 33

recognition modules which process this information. Generally, the AV consists of

adaptive detection and recognition framework (ADAF), a control system, LDWS, TSR,
unknown obstacles recognition (UOR), vehicle positioning and localisation (VPL)
module, etc. This processed information is fused and passed to the decision and
planning stage.
2. Decision and Planning: Utilising the data gathered in the perception process, this
stage decides, plans and controls the motion and behaviour of the AV. This stage is
analogous to the brain and makes decision such as path planning, action prediction,
obstacle avoidance, etc. The decision is made based on the current as well as past
information available including real-time map information, traffic details and patterns,
information by the user, etc. There may be a data log module that records errors and
information for future reference.
3. Control: The control module receives information from the decision and planning
module and performs functions/actions related to physical control of the AV such as
steering, braking, accelerating etc.
4. Chassis: The final stage includes the interface with the mechanical components
mounted on the chassis such as the accelerator pedal motor, brake pedal motor,
steering wheel motor and gear motor. All these components are signalled to and
controlled by the control module.
After discussing the overall communication and sensor architecture of an AV, we
discuss the design, functionality and utilisation of some main sensors.

Figure 4. System Architecture for AVs.

3.1. Ultrasonic Sensors

These sensors use ultrasonic waves and operate in the range of 20–40 kHz [14]. These
waves are generated by a magneto-resistive membrane used to measure the distance to the
object. The distance is measured by calculating the time-of-flight (ToF) of the emitted wave
to the echoed signal. Ultrasonic sensors have very limited range which is generally less
than 3 m [47]. The sensor output is updated after every 20 ms [47], making it not compliant
with the strict QoS constraints of an ITS. These sensors are directional and provide a very
narrow beam detection range [14]. Therefore, multiple sensors are needed to to get a
full-field view. However, multiple sensors will influence each other and can cause extreme
ranging errors [48]. The general solution is to provide a unique signature or identification
code which will be required to discard the echoes of other ultrasonic sensors operating
in near-by range [49]. In AVs, these sensors are utilised to measure short distances at low
speeds. For example, they are used for SPA and LDWS [50]. Moreover, these sensors work
satisfactorily with any material (independent of color), in bad weather conditions and even
in dusty environments.
Sensors 2021, 21, 706 8 of 33

3.2. RADAR: Radio Detection and Ranging

RADARs, in AVs, are used to scan the surroundings to detect the presence and loca-
tion of cars and objects. RADARs operate in the millimetre-wave (mm-Wave) spectrum
and are typically used in military and civil applications such as airports or meteorolog-
ical systems [51]. In modern vehicles, different frequency bands such as 24, 60, 77 and
79 GHz are employed and they can measure a range from 5 to 200 m [52]. The distance
between the AV and the object is calculated by measuring the ToF between the emitted
signal and the received echo. In AVs, the RADARs use an array of micro-antennas that
generate a set of lobes to improve the range resolution as well as the detection of multiple
targets [53]. As mm-Wave RADAR has higher penetrability and a wider bandwidth, and it
can accurately measure the short-range targets in any direction utilising the variation in
Doppler shift [51–54]. Due to longer wavelength, mm-Wave radars have an anti-blocking
and anti-pollution capability that allows them to cope in rain, snow, fog and low-light.
Furthermore, mm-Wave radars have the ability to measure the relative velocity using the
Doppler shift [15]. This ability of mm-Wave radars make them suitable for extensive AV
application such as obstacle detection [55], pedestrian recognition [56] and vehicle recog-
nition [57]. Some applications of RADARs in AVs are forward cross traffic alert (FCTA),
lane change assistance (LCA), blind spot detection (BSD), rear cross traffic alert (RCTA),
etc. The mm-Wave also has some disadvantages such as reduced field-of-view (FoV), less
precision and results in getting more false alarm as a result of emitted signals which gets
bounced from the surroundings [54].

3.3. LiDAR: Light Detection and Ranging

LiDAR utilises the 905 and 1550 nm spectra [58]. The 905 nm spectrum may cause
retinal damage to the human eye, and, therefore, the modern LiDAR is operated in the
1550 nm spectrum to minimise the retinal damage [43]. The maximum working distance
of LiDAR is up to 200 m [15]. LiDAR can be categorised into 2D, 3D and solid-state
LiDAR [59]. A 2D LiDAR uses the single laser beam diffused over the mirror that rotates at
high speed. A 3D LiDAR can obtain the 3D image of the surrounding by locating multiple
lasers on the pod [60]. At present, the 3D LiDAR can produce reliable results with an
accuracy of few centimetres by integrating 4–128 lasers with a horizontal movement of
360 degrees and the vertical movement of 20–45 degrees [61]. The solid-state LiDAR uses
the micro-electromechanical system (MEMS) circuit with micro-mirrors to synchronise the
laser beam to scan the horizontal FoV several times. The laser light is diffused with the
help of a micro-mirror to create the vertical projection of the object. The received signal
is captured by a photo-detector and the process repeats until the complete image of the
object is created. LiDAR is used for positioning, obstacle detection and environmental
reconstruction [15]. 3D LiDAR sensors are playing an increasingly significant role in the
AV system [62]. As a result, the LiDARs can be used for ACC, 2D or 3D maps and object
identification and avoidance. A roadside LiDAR system has shown to reduce the vehicle-
to-pedestrian (V2P) crashes both at intersections and non-intersection areas [63]. In [63], a
16-line real-time computationally efficient LiDAR system is employed. Deep auto-encoder
artificial neural network (DA-ANN) is proposed, which achieves an accuracy of 95% within
a range of 30 m. In [64], a 64-line 3D LiDAR utilising a support vector machine (SVM)-based
algorithm is shown to improve the detection of the pedestrian. Although LiDAR is superior
to a mm-Wave radar in measurement accuracy and 3D perception, its performance suffers
under severe weather conditions such as fog, snow and rain [65]. In addition, its operating
range detection capability depends on the reflectiveness of the object [66].

3.4. Cameras
The camera in AVs can be classified as either visible-light based or infrared-based
depending upon the wavelength of the device. The camera uses image sensors built
with two technologies that are charge-coupled device (CCD) and a complementary metal-
oxide-semiconductor (CMOS) [43]. The maximum range of the camera is around 250 m
Sensors 2021, 21, 706 9 of 33

depending on the quality of the lens [15]. The visible cameras use the same wavelength
as the human eye i.e., 400–780 nm, and is divided into three bands: Red, Green and Blue
(RGB). To obtain the stereoscopic vision, two VIS cameras are combined with known focal
length to generate the new channel with the depth (D) information. Such a feature allows
the camera (RGBD) to obtain a 3D image of the scene around the vehicle [67].
The infrared (IR) camera uses passive sensors with a wavelength between 780 nm and
1 mm. The IR sensors in AVs provide vision control in peak illumination. This camera
assists AVs in BSD, side view control, accident recording and object identification [68].
Nevertheless, the performance of the camera changes in bad weather conditions such as
snow, fog and moment-of-light variation [15].
The main advantages of a camera are that it can gather and record the texture, color
distribution and contour of the surroundings accurately [15]. However, the angle of
observation is limited due to narrow view of the camera lens [69]. Therefore, multiple
cameras have been adopted in AVs to monitor the surrounding environment [70,71]. A
three-stage RGBD architecture using deep learning and convolutional neural networks
was proposed by Ferraz et al. for vehicle and pedestrian detection [72]. However, this
requires the AV to process huge amount of data [15]. Currently, AVs do not possess
such computational resources; therefore, computational offloading may be an appropriate
solution [73].
Table 4 summarises the challenges of the discussed sensor technologies. It can be
observed in Table 4 that the detection capability and reliability of the various sensors is
limited in different environments. This limitation can be overcome and the accuracy of
target detection along with the reliability can be improved through multi-sensor fusion.
Radar–camera (RC) [56,74], Camera–LiDAR (CL) [60,75], Radar–LiDAR (RL) [76] and
Radar–Camera–LiDAR (RCL) [77,78] have been proposed where different sensors are
combined together to improve the perception of the environment. Furthermore, in [14],
three different sensor plans are developed based on range, cost and balance function. In
this study, several different sensors are combined. In Plan A, four cameras, a mm-Wave
RADAR, 32- and 4-layer LiDAR and a GPS+IMU are employed. In Plan B, four cameras,
three mm-Wave RADAR, a four-layer LiDAR and a GPS+IMU are utilised. Finally, in Plan
C, two regular cameras, three mm-Wave RADARs, a surrounding camera and a twelve-unit
ultrasonic sensor are utilised.

Table 4. Comparison of sensor and their challenges.

Sensor Challenges
1. Cannot be used at high speed
Ultrasonic Sensors 2. Maximum range is 2 m [15]
3. Very low resolution as compared to RADAR
1. Generate a large number of false alarms due to surroundings metal objects
RADAR 2. The range is between 5 m and 200 m
3. Generated images are of low resolution as compared to Cameras and LiDARs
1. Suffers extremely from the weather
LiDAR 2. Maximum range is 200 m [15]
3. Very expensive due to high price.
1. Computation overheads which increases the time-critical applications [49]
Cameras 2. The maximum range is 250 m depending upon the lens [15]
3. A big range resolution gap exists between the Cameras and RADARs or LiDARs.
Sensors 2021, 21, 706 10 of 33

3.5. GNSS and GPS, IMU: Global Navigation Satellite System and Global Positioning System,
Inertial Measurement Unit
This technology can determine the exact position of the AV and helps it navigate [79].
GNSS utilises a set of satellites orbiting around the earth’s surface to localise [80]. The
system contains the information of AV’s position, speed and the exact time [80]. It operates
by calculating the ToF between the satellite emitted signal and the receiver [81]. The
AV position is usually extracted from the Global Positioning System (GPS) coordinates.
The extracted coordinates by GPS are not always accurate and they usually introduce
an error in the position with a mean value of 3 m and a standard deviation of 1 m [82].
The performance is further degraded in urban environments and an error in position
can increase up to 20 m [83] and in some extreme cases the GPS position error is around
100 m [84]. In addition to this, the RTK system can also be used in AVs to precisely calculate
the position of the vehicle [85]. Furthermore, dead reckoning (DR) and the inertial position
can also be used in AVs to determine the position and the direction of the vehicle [86].
A technique known as odometry can be used to measure the position of the vehicle by
fixing the rotary sensors to the wheels of the vehicle [79]. To make the AV capable of
detecting slippage or lateral movements, the inertial measurement unit (IMU) is used and
it detects this using accelerometers, gyroscopes and the magnetometer sensor’s data. The
IMU combined with all units can rectify the errors and increases the sampling speed of the
measuring system. Although the IMU cannot provide the position error unless it is not
accompanied by the GNSS system, AVs can get information from different sources such as
RADAR, LiDAR, IMU, GNSS, UWB and camera to minimise the possibilities of error and
perform reliable position measurement [43]. GPS can be combined with IMU techniques
such as DR and the inertial position to confirm and improve the position estimate of the
AV [87].

3.6. Sensor Fusion

Real-time and accurate knowledge of vehicle position, state and other vehicle pa-
rameters such as weight, stability, velocity, etc. are important for vehicle handling and
safety and, thus, need to be acquired by the AVs using various sensors [88]. The process
of sensor fusion is used to obtain coherent information by combining the data obtained
from different sensors [15]. The process allows the synthesis action of raw data obtained
from complimentary sources [89,90]. Therefore, sensor fusion allows the AV to precisely
understand its surrounding by combining all the beneficial information obtained from
different sensors [91]. The fusion process in AVs is carried out by using different types of
algorithms such as Kalman filters and Bayesian filters. The Kalman filter is considered very
important for a vehicle to drive independently because it is utilised in different applications
such as RADAR tracking, satellite navigation system and visual odometry [92].

3.6.1. Kalman Filters

The Kalman filter (KF) is used to calculate the probabilistic state of the dynamic
system in the past (smoothing), present (filtering) or future (prediction) [93]. As the
sensors used in self-driving cars produce noisy and incomplete data, the role of the Kalman
filter is to eradicate the unwanted noise and obtain accurate estimates [94]. The state
of the system (xx ) is composed of the position and velocity of the AV [95]. In addition,
as the data are obtained from different sensors such as Radar, Lidar, ultrasonic, etc., the
measured/observed position and velocity of the AV varies depending upon the accuracy
of these sensors. This uncertainty in the data (of position and velocity) can be reduced by
combining the measured data with the prediction of the states utilising the Kalman filter.
The Kalman filter models the uncertainty using a Gaussian model [16]. The variance of the
Gaussian model, σ2 , measures the amount of uncertainty in the state of the system and thus,
a larger variance implies greater uncertainty [96]. To accurately estimate the state of the
system, with low uncertainty, the Kalman filter compares the current observation with the
previous prediction. The process is recursive unless the measurement data are unavailable.
Sensors 2021, 21, 706 11 of 33

Mathematical Model
The Kalman filter is composed of two stages, prediction and update, to finalise the
actual state of the system [93,97]. Considering a dynamic system moving linearly, its
mathematical model can be briefly explained as follows.

Calculation of the Prediction Stage

In the prediction stage, the state, x 1 , of the system with an uncertainty of P 1 at time t,
is estimated from its previous state x and P at time (t − 1). The formula to calculate the
predicted state is as follows [93]

x1 = Fx + n (1)
P1 = FPF H + R (2)

where F is the transition matrix from (t − 1) to t, n is the added noise and R is the noise
co-variance matrix.

Calculation of the Update Stage

In the update stage, the predictions are modified by comparing them with the data
obtained from sensors. If z represents the measured data from sensors, then the estimated
error between the measurement and the prediction is given as [98]

y = z − H x 1, (3)

where H represents the transition matrix between the measured sensor data [93,99].
Similarly, the estimated system error (E E ) can be given as E = H P 1 H T + R . The Kalman
T − 1
K ) is given by K = P 1 H E . The essential purpose of Kalman gain is to correct
filter gain (K
predictions [99,100]. The actual state of the system and position is predicted as [93]

x = x 1 + Ky (4)
P = (I − K H )P 1 (5)

Even though the measurement will provide data that are originated from sensors, the
information in the predicted and measurement states is always uncertain and therefore the
optimal state estimate will be the one in which the position has a lower uncertainty [101].
From the above equations, it can be noted that, to learn the vehicle position and
state, the knowledge of the transition matrix and the noise co-variance matrix is required.
Traditional Kalman filtering yields the optimal solution if the noise statistics and noise
process is completely defined [102]. These matrices are configured with priori fixed values
that do not adapt with the changing surroundings [103]. Ideally, these matrices should
change, adapt and update dynamically according to their surroundings and environment.
Therefore, various approaches have been adopted in the literature to adapt the matrices
and improve the estimate of the state and the position of the vehicle [104,105].
In this regard, there has been significant research interest in Extended Kalman filters
(EKF), Unscented Kalman filters (UKF) and Cubature Kalman filters (CKF) [105–107].
In [108], the EKF was used to estimate the lateral and longitudinal tire forces to improve
the handling and safety of the vehicle. In [109], a dual extended Kalman filter (DEKF) was
proposed which employs two parallel EKFs to estimate the position and state of the AV.
In [110], a DEKF was applied to estimate the road friction parameters. However, EKFs
are strongly influenced by the approximated noise and diverge when non-linearity of the
model dominates [111]. An adaptive EKF (AEKF) was proposed by Akhlaghi et al. [112],
which adopts a traditional co-variance matching technique. Both matrices are updated
continuously by making the filter residual co-variances consistent with the theoretical
co-variances.
A more accurate generalisation is to employ the Unscented Kalman Filter (UKF) [113].
UKF for vehicle dynamics is proposed in [114,115]. A dual UKF (DUKF) has been applied
Sensors 2021, 21, 706 12 of 33

to simultaneously estimate the sideslip angle and vehicle mass [116]. However, in a higher
dimension filtering problem, the cubature Kalman filter (CKF) has shown better stability
and accuracy [107,117]. The improved performance is because of its ability to solve the
multi-dimensional integral. Furthermore, in [118], the dual CKF (DCKF) is proposed,
which performs better than the CKF under various friction coefficients when estimating
the position and states of the AV. The UKF and CKF fall in the sampling based KFs and
the performance of these algorithms depends upon the accuracy of instantiated sigma
points [105]. Therefore, Particle filter (PF) and its variants, which are discussed in next
section, are ideal for non-linear and non-Gaussian problems.

3.6.2. Particle Filters

The basic assumption of the Kalman filters is that the noise experienced by the system
is Gaussian [101]. However, in most cases, the noise experienced by the system will not
have a Gaussian characteristic [119]. Particle filters (PF) have been proposed for processing
non-Gaussian noise (see, e.g., [120] and the references therein). When applying PF there
is always a trade-off between precision and computational resources [111,121]. Therefore,
an extended particle filter (EPF) [122] and an unscented particle filter (UPF) [111] have
been proposed. Both EPF and UPF use the sigma points, making them computationally
inefficient and difficult to implement in real time [122]. Recently, particle-aided unscented
Kalman filter (PAUKF) has been proposed [101] which does not update the sigma points
resulting in fewer computational resources. For vehicle localisation, the PAUKF algorithm
has been shown to perform consistently better compared to existing methods including PF
and UKF and yields an RMSE of approximately 1.5 m [101].

4. Vehicular Ad-Hoc Networks (VANETs)

VANETs are an emerging sub-class of mobile ad-hoc networks capable of spontaneous
creation of a network of mobile devices/vehicles [123]. VANETs can be used for vehicle-
to-vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication [44,124]. The main
purpose of such technology is to generate security on the roads; for example, during
hazardous conditions such as accidents and traffic jam the vehicles can communicate with
each other and the network to share vital information [125,126]. The main components of
VANET technology are:
1. On-board unit (OBU): It is a GPS-based tracking device embedded in every vehicle
to communicate with each other and with roadside unit (RSU) [124,126]. To retrieve
the vital information, the OBU is equipped with many electronic components such as
resource command processor (RCP), sensor devices and user interfaces. Its main goal
is to communicate between different RSUs and OBUs via a wireless link [44].
2. Roadside Unit (RSU): RSU is a computing unit fixed at specific location on roads,
parking areas and intersections [127]. Its main goal is to provide connectivity
between autonomous vehicle and the infrastructure and also assists in vehicle
localisation [44,127]. It can also be used to connect vehicle with other RSUs using
different network topologies [44]. They have also been powered using ambient
energy sources such as solar power [128].
3. Trusted Authority (TA): It is an authority which manages the entire process for
VANETs, so that only valid RSUs and vehicle OBUs can register and communi-
cate [129]. It provides security by verifying the OBU ID and authenticates the vehicle.
It also detects malicious messages or suspicious behaviour [44].
VANETs have some unique properties which are very different from other ad-hoc
technologies.
• VANETs have very low discovery latency and as a result the vehicles, even at high
speeds, connect to the RSU quickly and rarely face network outage [130,131].
Sensors 2021, 21, 706 13 of 33

• The OBUs can move with predictable and regular path. It can help to detect the actual
trajectory of the vehicle at any point of time [131]. The RSUs in VANETs can localise
the vehicle and also log the path of the vehicle and also predict its trajectory to avoid
any hazard.
• The vehicle sensors and other nodes do not face any energy restrictions because they
can extract energy from the vehicle engine.
• The use of multicast broadcasting in VANETs allows the different vehicles to commu-
nicate with each other simultaneously [132].
Vehicular communication, utilising VANETs, includes V2V communication, V2I commu-
nication and V2X communication, as illustrated in Figure 5. The details are given below.

Figure 5. Vehicular communication (VC) system.

4.1. Vehicle-To-Vehicle (V2V) Communication

It is also called inter-vehicle communication (IVC) that allows the vehicles to communi-
cate with each other and share the necessary information about traffic congestion, accidents
and speed limits [4]. V2V communication can generate the network by connecting different
nodes (Vehicles) using a mesh (partial or full) topology [5]. Depending upon the number
of hops used for inter-vehicle communication, they are classified as single-hop (SIVC) or
Multi-hop (MIVC) systems [9]. The SIVC can be used for short-range applications such as
lane merging, ACC, etc., whereas MIVC can be used for long-range communication such
as traffic monitoring. The V2V communication provides several advantages such as BSD,
FCWS, automatic emergency braking (AEB) and LDWS [4].

4.2. Vehicle-To-Infrastructure (V2I) Communication

It is also known as roadside-to-vehicle communication (RVC) and allows the vehicles
to interact with the RSUs. It helps to detect traffic lights, cameras, lane markers and
parking meters [1]. The communication of vehicles with the infrastructure is ad-hoc,
wireless and bidirectional [10]. The data collected from the infrastructure are used for
traffic supervision and management. They are used to set different speed variables allowing
the vehicles to maximise fuel efficiency as well as control the traffic flow [4]. Depending
on the infrastructure, the RVC system can be divided into the Sparse RVC (SRVC) and
the Ubiquitous RVC (URVC) [23]. The SRVC system provides communication services at
hotspots only, for example to detect available parking spaces or gas stations, whereas the
URVC system provides coverage throughout the road, even at high speeds. Therefore, the
URVC system requires a large investment to ensure network coverage [23].

4.3. Vehicle-To-Everything (V2X) Communication

The V2X communication allows the vehicle to communicate with other entities such
as pedestrians (V2P), roadside (V2R), devices (V2D) and the Grid (V2G) [133]. This com-
munication is used to prevent road accidents with vulnerable pedestrians, cyclists and
motorcyclists [20]. The V2X communication allows the Pedestrian Collision Warning (PCW)
mechanism to alert the roadside passenger before any serious accident takes place. The
Sensors 2021, 21, 706 14 of 33

PCW can access the Bluetooth or Near Field Communication (NFC) of the smartphone and
may use beacon stuffing to deliver critical messages to the pedestrian [4].

5. Different VANET Technologies Based on Transmission Range

An AV can communicate with other vehicles and the network infrastructure by using
different wireless technologies. For example, the AV may transmit and receive real-time
data (Audio/Video) during V2X communication or in case of emergency situations such
as fog or accidents, warning information can be exchanged with neighbouring vehicles
in a short time [29]. As discussed above and shown in Figure 4, the wireless technology
employed depends on the system architecture and particularly on the kind of sensors
employed along with the use case. The wireless technologies used in autonomous vehicles
can be differentiated in terms of their transmission range, as shown in Figure 6.

Figure 6. Different VANET technologies considered for AVs.

The long-range technologies allow the vehicles to transmit data while being miles
apart; examples are C-V2X and 5G-NR communication. The medium-range communica-
tion, using Wireless Local Area Network (WLAN) or DSRC, can provide the coverage for
V2V and V2I communication within a range of tens or hundreds of feet. The short-range
communication technologies include Bluetooth, UWB and ZigBee. Below, these wireless
technologies are discussed in detail.

5.1. Short-Range Communication Technologies

5.1.1. Bluetooth Technology
The Bluetooth technology based on IEEE 802.15.1 protocol can be used as the short-
range communication technology in the vehicular network [20]. It operates in the 2.4 GHz
band and can produce the data rate up to 1–4 Mbps with variable coverage ranges [134].
The range of Bluetooth varies due to the propagation conditions, the sensitivity and gain of
the transmitting and receiving antennas [39]. Wi-Fi and Bluetooth use the same bandwidth;
therefore, to avoid interference, Frequency Hopping Spread Spectrum (FHSS) is used in
Bluetooth [135]. Different versions of Bluetooth use different standards. For example,
Bluetooth 4.0 (BLE) uses low energy for data transmission, but it is not compatible with
older versions of Bluetooth known as “Bluetooth classic” (BC) [136].
Due to the low transmission range of BLE, it is unsuitable for inter-vehicular commu-
nication [137]. However, the range and throughput supported by BLE make it a suitable
candidate to eliminate the complex wiring within an AV and reduce its weight. BLE is
suitable for wireless audio stream transmission and can be used to connect the AV’s info-
tainment system wirelessly to the speakers and microphone. In addition, it may be used
to wirelessly connect sensors to the ECU and OBU and reduce the cost and complexity of
wiring. In [138], BLE was compared to the CAN bus for intra-vehicular communication
Sensors 2021, 21, 706 15 of 33

and it was shown that the BLE is a suitable low cost energy efficient solution for com-
munication with less critical sensors having relaxed latency and throughput requirement.
However, as BLE takes time to discover devices and establish connection, it is not suitable
for critical delay sensitive vehicular applications [139]. In [140], an algorithm utilising the
BLE technology is proposed for indoor positioning and exterior localisation for vehicular
applications. Furthermore, Bluetooth vulnerabilities are highlighted in [141]. Blueborne,
which is identified in [142], can have a serious effect on automotive security. However,
detailed Bluetooth and BLE vulnerabilities for AVs are still missing in the literature [143].

5.1.2. ZigBee Technology

ZigBee technology was developed to address the need for low cost and low power
wireless Internet-of-things (IoT) network [144]. The supported connectivity range is up to
100 m with a transfer data rate of about 250 kbps [145]. It is also recognised as IEEE 802.15.4
for Low Rate Wireless Personal Area Network (LRWPAN) and uses different frequency
bands (868 MHz, 902–968 MHz and 2.4 GHz) for operation [146,147]. ZigBee employs
binary phase shift keying (BPSK) or orthogonal quadrature phase shift keying (OQPSK)
for modulation.
Similar to BLE, ZigBee also is suitable for intra-vehicular communication and to
reduce the complex wiring within an AV. Recently, Volvo deployed a ZigBee based wireless
sensor network in trucks to monitor the tire pressure [148]. It was concluded that ZigBee is
sufficiently robust for non-safety and non-security critical applications requiring relaxed
latency with update rates less than 10 Hz. Pawade et al. proposed the use of ZigBee for
Advanced Driver Assistance Systems (ADAS) [149]. In [150], the performance of ZigBee
was simulated in the QualNet simulator. It was observed that as the number of cars
increased to 50, the packet delivery ratio dropped significantly and the average end-to-end
delay was more than 200 s. As a consequence, it was concluded that ZigBee cannot be
employed for urban or crowded places. Furthermore, ZigBee technology has successfully
been implemented and tested for FCWS in vehicles [41]. ZigBee technology has been
successfully implemented to locate monitored vehicles on the real time in [151]. However,
the distance promised is based on received signal strength indicator (RSSI) which is not
very accurate.

5.1.3. Ultra-Wide Band (UWB) Technology

UWB is considered as the short-range wireless technology using short pulses of
bandwidth 3.1–10.6 GHz [22]. Using UWB, the devices can operate at very low power and
also support multiple users with data rates greater than 480 Mbps [152–154]. With UWB, it
is possible to communicate with a unique randomising code by transmitting Gigabits of
information per second. These signals have low amplitude that makes them more covert
and harder to degrade in the noisy environment [152–154]. This allows the UWB to be more
secure in transmission with low probability of detection and interception. Devices using
the UWB technology are generally designed to have a large processing gain [152–154].
Several UWB nodes were deployed in the PRoPART project for obtaining an accurate
and robust positioning technique. The proposed solution utilised the inertial and odometry
data of the vehicle along with the UWB ranging estimates to estimate the vehicle position.
The estimates obtained were accurate and 95% of the computed positions were shown to
have a positioning distance error of less than 27.1 cm [155]. In [156], machine learning was
employed to improve the RMSE of UWB based ranging estimates, and the positioning
distance error was reduced to less than 10 cm. The UWB transmitter employs low power
signals which are below −41.3 dBm/MHz emission limit. This low emission with time
hopping (TH) code, direct sequence (DS) code or a combination of DS/TH codes protect
the data from eavesdropping attack.
A comparison of these four technologies is given in Table 5.
Sensors 2021, 21, 706 16 of 33

Table 5. Comparison of short-range technologies.

Standards Bluetooth BLE ZigBee UWB

Specifications IEEE 802.15.1 IEEE 802.15.1 IEEE 802.15.4 IEEE 802.15.3
Frequency 2.402–2.481 2.402–2.481 868/902–968 MHz, 2.4 GHz 3.1–10.6 GHz
Bandwidth 1 2 0.3/0.6 MHz, 2 MHz 500 MHz–7.5 GHz
Rate 1–3 Mbps 1 Mbps 20–250 kbps 480 Mbps
Range 10 m 50 m 75–100 m 75 m
Latency (msec) 100 6 30 0.1
Modulation GFSK GFSK BPSK,O-QPSK BPSK-QPSK
Data Protection 16-bit CRC 24-bit CRC 16-bit CRC 32-bit CRC

5.1.4. Applications
The short-range communicating technologies can assist various vehicular applications.
Some of the application are listed below.
• Vehicle Localisation: The GPS used in the vehicles can easily determine the position
of the vehicle but in the dense urban environment its localisation becomes more
challenging. However, these short-range technologies, especially UWB, can assist
localisation in dense environments as they do not require the Line of Sight (LoS) and
can easily penetrate in the obstacles [155]. UWB can be regarded as the technology
of choice for range-based localisation especially for the vehicles moving in dense
clustered environments [157].
• Real Time Driving Assistance: These short-range communication technologies can be
utilised for communication between RSUs and the vehicle’s OBU for gathering of real-
time driving data which may help in lane assist and avoidance of road accidents [155].
• Vehicle Identification: As already discussed, these technologies consumes less energy;
therefore, they can be useful to implement at toll plazas where the vehicle identification
and classification plays a crucial role for tax collection [36].
• Forward Collision Warning Under Low Speed Circumstances: These short-range
technologies can be used for short-range V2V communication in low speeds, which
can be used to prevent collisions between the self-driving vehicles or for lane and
parking assist. This technology offers the best solution to avoid forward collisions
between self-driving cars at low speeds [41].
• Others: The above-mentioned technology can provide additional advantages such
as identifying the car owner, detecting the position of passengers in the vehicle and
deactivating the air bags on emergency cases [140,147,155].

5.2. Medium-Range Communication Technologies

5.2.1. Dedicated Short-Range Communication (DSRC)
DSRC or wireless access in vehicular environment (WAVE) is a standard specifically
proposed for reliable communication between vehicles and the network infrastructure.
The technology can be deployed in OBUs and RSUs to help the vehicles to communicate
effectively and create a wide area network [42]. The WAVE/DSRC is a modified version of
the Wi-Fi technology (IEEE 802.11), and its set of standards such as IEEE 802.11p and IEEE
1609.x provides some useful results while applying for vehicular access. The interference
between DSRC and Wi-Fi can be avoided by setting the frequency spectrum in DSRC
much higher than the Wi-Fi spectrum. The difference between Wi-Fi and DSRC is given in
Table 6.
DSRC is utilized in V2V communication where OBUs of vehicles communicate with
each other. It can also be deployed for V2I communication, where vehicles get assistance
from the network infrastructure regarding traffic signals, accident alerts, etc. [25,26]. In
Sensors 2021, 21, 706 17 of 33

addition, the Quality of Service (QoS) of DSRC communication can be improved by using
the enhanced distribution channel access (EDCA) protocol [38]. DSRC security threats
were classified into three levels: critical, major or minor [158]. Denial of service attack on
DSRC was studied by Biswas et al. [159] and its potential countermeasures were analysed
by Whyte et al. [160]. Finally, it was argued that DSRC cannot be the standalone system for
ITS and a Het-Net approach would be required by Dey et al. [161].

Table 6. Comparison of medium-range technologies.

Parameters Wi-Fi DSRC

Specifications IEEE 802.11a IEEE 802.11p
Width in channel 20 10
Signalling OFDM OFDM
Data Rate (Mbps) upto 54 upto 27
Modulation type upto 64QAM upto 64QAM
Symbol duration (micro sec.) 4 8
Guard Time (micro sec.) 0.8 1.6
FFT Size 64 64
FFT period (micro sec.) 3.2 6.4
Preamble duration (micro sec.) 16 32
Sub carrier spacing (MHz) 0.3125 0.15625
Frequency spectrum (GHz) 5 5.9
Latency (msec) 50 100
Range (m) 100 300

5.2.2. Applications
DSRC technology enables various applications in V2V communication and V2I com-
munication. Some applications are discussed below.
• Traffic Safety: DSRC technology can be used to improve traffic safety [162]. It allows
vehicles to detect the barriers and to avoid collision accidents on the roads [35]. With
the help of DSRC, the vehicle can detect a sharp curve and warn other cars [27]. If
an accident happens on the road, the vehicles can transmit an alert message to other
vehicles to avoid collisions [28].
• Traffic Management System: DSRC can help establish a traffic management system
which includes highway fleet management, safe overtaking, etc. [40]. The vehicles can
safely overtake and maintain a desired gap during lane changes [30]. If a particular
vehicle wants to change its lane, it will send the request to other vehicles for lane
change [14]. Therefore, the vehicle can perfectly manage the traffic using DSRC
communication.
• Vehicle Management: DSRC can be used to identify registered vehicles and provide
an automatic pass to them. Hence, trespassing with a non-registered vehicle can be
blocked [31].

5.2.3. Limitations
The DSRC technology has several benefits such as low latency, secure transmission
and fast network acquisition [163]. The QoS of DSRC is much better than Wi-Fi but the
technology itself faces some technical difficulties [42]. Some technical challenges in DSRC
technology are mentioned below.
Sensors 2021, 21, 706 18 of 33

• Hidden Terminal Problem: As the MAC layer consists of two parts, the IEEE 802.11p
forms the lower layer while as IEEE 1609.4 contains the upper MAC layer. In DSRC
technology, when the packets are transmitted, the ACK mechanism does not exist. In
addition, The clear-to-send mechanism is not set by default in carrier-sense multiple
access with collision avoidance (CSMA/CA) medium access, which results in the
hidden terminal problem [164].
• Congestion Issue: As the number of AVs increase, the communication load also
increases, which causes congestion [164]. This congestion leads to delay and packet
loss that affects the performance of the network. The increased number of AVs also
increases the intensity of the channel will result is severe degradation of performance.
• Handover issue: A vehicle that travels with high speed has to change several net-
work topologies over time. Two types of handover occur: (1) horizontal handover;
and (2) vertical handover. Horizontal handover is the handover of the moving AV
from one RSU to another, whereas vertical handover is handover of an AV between
different cellular LTE base stations. A vertical handover takes longer time compared
to the horizontal handover [30]. This will result into the above-mentioned hidden
terminal problem.
• Doppler Spread: The Doppler spread occurs due to the relative motion of transceivers.
The relative velocities shift the signal frequency and the signal at the receiver is at a
different frequency compared to the broadcast signal. The transmitted signal in DSRC
technology faces this Doppler shift phenomenon due to vehicle motion on road which
will affect predicting the exact location and position of the AV [35].

5.3. Long-Range Communications Technology

5.3.1. C-V2X Technology
The vehicles communicate with their surroundings through the cellular network [165].
The C-V2X technology was introduced by 3GPP release 14, and it was further developed by
3GPP release 15 in order to fulfil the criteria of 5G communications [30]. C-V2X technology
offers multiple benefits listed below [24].
• C-V2X is highly reliable and can perform communication between the vehicles at high
speed.
• The C-V2X technology can perfectly operate in dense traffic conditions, therefore it
overcomes the congestion issues faced by DSRC.
• The C-V2X technology supports both short-range and long-range communication
between the vehicles and with the RSU.
• The C-V2X technology is able to provide a 360-degree view non-line of sight (NLoS)
sensing.
• C-V2X can support much higher data rates compared to the short- and medium-range
technologies discussed and, thus, it is able to support applications which require
higher data rates.
The C-V2X technology has two modes for vehicular communication. Mode-3 allows
the vehicle to communicate via the base-station while Mode-4 allows the vehicles to
communicate with each other directly [166]. Mode-4 uses the PC5 interface at 5.9 GHz
while Mode-3 uses the traditional LTE interface [13,167]. In addition, Mode-3 depends
on centralised scheduling while Mode-4 is completely distributed. The purpose of direct
communication between the vehicles (through Mode-4) reduces the latency requirements,
necessary for most of time-critical applications in ITS. The vehicles may switch between
the desired modes dynamically [167]. LTE and LTE-A have severe security vulnerabilities;
the details are provided in [168] (and references therein). Security threats faced by C-V2X
is provided in [169]. Furthermore, countermeasures for C-V2X are discussed in detail
in [169,170] (and references therein).
Sensors 2021, 21, 706 19 of 33

5.3.2. 5G-NR Technology: Evolution of C-V2X Technology

The aim of the 5G-NR standard, developed by 3GPP, is to provide higher data rates,
lower latency and enable communication between multitude of devices [171,172]. These re-
quirements correspond to three different use cases. First, there is the massive machine type
communication (mMTC), which requires low bandwidth, low energy consumption and
high connection density to connect a large number of sensors [173]. Second, there is the en-
hanced mobile broadband (eMBB) for applications requiring extremely high data rates [174].
Third, there is the Ultra Reliable Low Latency Communication (uRLLC) required for ap-
plications that require extremely low delays along with low error rate [31,165,175]. The
5G-NR offers interoperability with the C-V2X technology and, as a result, vehicles in a
given geographical region can access both technologies. The gains provided by densifica-
tion of remote radio heads in cloud radio networks and radio head configuration schemes
can further contribute towards ubiquitous coverage [128,176]. The comparison between
the two technologies is given in Table 7.

Table 7. Comparison of long-range technologies.

Parameters C-V2X 5G-NR V2X

Subcarrier spacing (kHz) 15 up to 240
Carrier aggregation Up to 32 up to 16
Channel Bandwidth (MHz) 20 400
Latency (msec) <10 <1
Reliability 95–99% 99.9–99.999%
Channel Coding Turbo LDPC, Polar
Network Slicing No YES
Waveform SC-FDMA OFDM
Control and data multiplexing FDM TDM
Modulation 16 or 64QAM 256 QAM
Communication Type Broadcast Broadcast, multicast and unicast
Re-transmission Blind PSFCH
Security and privacy Basic Advanced
Positioning accuracy (m) >1 0.1
Frequency Spectrum 800/1800 MHz 700 MHz/3.6 and 26 GHz
Range 100 m to >5 Km 50 m to >5 Km

5.3.3. Applications
From practical implementation and performance perspective, 5G-NR V2X technologies
have potential to contribute in the following areas of enhancements:
• Security and Privacy are critical for successful AV operation in practical settings and
largely depend upon the security and privacy provisions of the enabling communi-
cation technology. To this end, it is important to ensure the security of one-to-one
interactions, e.g., V2N, V2I, etc., as well as one-to-many interactions such as I2V, V2V,
etc. In 5G, different approaches such as session based authentication and authorisation
can be utilised in the former case; however, for the latter, signing each packet uses a
trusted hardware security module can be used for enhanced security [177].
• Management and Orchestration (MANO) have a direct influence on the network
performance and hence require optimisation. 5G-NR V2X network infrastructure
comprises of various components distributed across the network at various hierar-
Sensors 2021, 21, 706 20 of 33

chical levels. The orchestration of such infrastructure allows coordination across

all the different network components that are involved in the life-cycle of each net-
work resource block. It involves managing services enabling deployment automation,
maintenance, management, configuration, upgrade and repair of each component
of the system [177]. ETSI open source MANO (OSM), Network Function Virtualisa-
tion (NFV) and Software Defined Networking (SDN) are studied in [178] from V2X
perspective for MANO.
• Quality of Service (QoS) has stringent constraints in V2X networks including latency,
capacity, etc., as explained above. Several procedures, such as network-application
negotiation, connectivity establishment, mobility and user plane management, etc.
directly impact the service delivery and its quality. URLLC and eMBB are enabling
features offered by 5G for V2X aiming at providing desired QoS to the users.
• Edge services: 5G offers a unique provision of providing computational resources and
storage at the network edge. This enables the network to host applications closer to
the AVs resulting in reduced latency. Multi-Access Edge Computing (MEC) utilises
the network core and backhaul efficiently, therefore reducing the latency requirement
of the AVs [178]. Since vehicular users are highly mobile, frequent handovers can be
supported via resource management at the edge of the access network [177].

5.3.4. Field Testing

Numerous pilots and feasibility studies have been carried out during the previous
decade or so to evaluate the practical performance and suitability of AVs. A comprehensive
yet non-exhaustive list of different trials and testing facilities of AVs can be found in [179]
for the UK and [180] in Europe and elsewhere. The V2X technologies and communication
interfaces are currently in active testing. The US Department of Transport (DoT) is running
US’s Safety Pilot Model Deployment [181] as well as NYCDOT [182], THEA DOT [183]
and WY DOT [184] pilots that explore DSRC based communication. HarborNet [185]
is a test-bed that allows 4G backhauling and Wi-Fi connectivity along with DSRC. The
testbed allows cloud-based code deployment, remote network control and distributed data
collection from moving container trucks, cranes, tow boats, patrol vessels and roadside
units. Google’s automotive driving and testing facility [186] has conducted extensive road
operation testing over the last few years. Tesla, Tencent and Baidu have used simulators and
road-testing methods together to conduct extensive testing in demonstration zones [187].
5GIC test bed in Surrey, UK and Espoo and Ispra trials sites in Italy are being utilised for
5G-NR V2X testing [188].
Finally, the wireless technologies which have already been implemented in AVs are
discussed in Table 8.

Table 8. Wireless technologies in AVs.

Technology Task References

Wirelessly connect sensors having relaxed latency to the ECU/OBU [138,139]
BLE
Indoor Positioning and Localisation (On-demand content for the passenger) [140,189,190]
Forward Collision Warning System [41]
Prevent Collision Warning System [191]
ZigBee
Wirelessly connect non-safety and non-security critical sensors to the ECU/OBU [148]
Advanced Driver Assistance Systems [149]
RSU are connected with UWB fixed nodes to improve the position accuracy [155]
UWB
where the coverage of GNSS is poor. Performance is further improved to 10 cm. [156]
Sensors 2021, 21, 706 21 of 33

Table 8. Cont.

Technology Task References

Operational concept for vehicle safety is presented. [192]
DSRC Field test shows that lane and position of surrounding vehicles can be identified
[193]
with 100% accuracy when the distance between vehicles is less than 50 m.
V2X framework is discussed in detail. [165]
C-V2X C-V2X is not able to satisfy all the latency requirements, therefore, shortened
[133]
transmission time interval is proposed.
Framework of 3GPP-Version 16 for AV is discussed. [194]
Field trial of mm-Wave with 5G-NR, a stable data rate of 2.8 Gbps within 500 ms
5G-NR V2X [195]
latency is achieved.
Field Trial of 5G-NR V2X satisfying the latency requirements for truck platooning. [196,197]

6. Conclusions and Future Research Directions

Due to the increase in road traffic, the possibility of accidents and collisions is also
increasing. The main goal of the development of self-driving/autonomous vehicles is to
make driving safer and more reliable, as well as reduce congestion. For this purpose, these
vehicles need to precisely communicate with other vehicles and their surroundings. This
process starts at the perception stage when the self-driving vehicle gather information from
the environment using various sensors. Even though the data obtained from the sensors
might be inaccurate, the fusion process allows the AVs to synthesise this raw data and
obtain accurate results. After analysing the surrounding and making a decision, the ad-hoc
network topology enables the AVs to communicate with each other and with the network
infrastructure to ensure road safety and traffic flow.
Different types of vehicular communication (V2V, V2I and V2X) performed by AVs
use different technologies. This survey differentiates these technologies on the basis of
transmission range. As a result, Bluetooth, BLE, ZigBee and UWB are grouped as short-
range technologies, while DSRC and Wi-Fi are considered medium-range technologies.
Similarly, C-V2X and 5G-NR are identified as long-range technologies.
This survey has compared all these technologies with the aim of identifying the
applications of V2X communication where each technology is best utilised. For short-range
communication technology, this survey discusses its key features and proposes its use for
some V2X applications which do not have strict latency requirements such as forward
collision warning, toll check and vehicle identification. However, they cannot support
applications such as remote driving, remote maintenance, etc. due to their lack of data
rates and latency. Out of the four short-range technologies, UWB stands out because it
can produce the highest data rate and low latency, and, hence, it can be used for above
mentioned vehicular applications.
Medium-range technologies such as DSRC and Wi-Fi can provide higher mobility
to the AVs. This survey discussed why DSRC technology provides higher flexibility
and is better suited for V2X communication compared to Wi-Fi. Furthermore, the DSRC
technology facilitates AVs with different applications such as collision avoidance, lane
changing assistance, etc. However, DSRC technology has certain limitations such as
congestion issue, Doppler spread and hand-over issue. Based on these limitations and
the ultra-low latency requirements, the paper concludes that DSRC technology does not
completely fulfil the diverse needs for V2X mobility.
Finally, the survey presents long-range technologies and identifies that C-V2X technol-
ogy and the 5G-NR are enabling technologies for future intelligent transport systems. This
survey discusses the two modes for C-V2X operations: Mode 3 and Mode 4. Mode 3 works
within cellular coverage and Mode 4 is suitable for direct V2V communication using the
PC5 interface. Moreover, different vehicular applications require high dates, low latency
and reliable communication and the 5G-NR technology is able to provide massive machine
type communication (mMTC), enhanced mobile broadband (eMBB) and ultra-reliable low
Sensors 2021, 21, 706 22 of 33

latency communication (uRLLC) and, hence, it is suitable for V2X communication. In

conclusion, this survey identifies 5G technology as one of the important advancements
in the development of autonomous cars and can serve as a backbone of future intelligent
transport systems.

6.1. Future Research Directions

The deployment of the 5G-NR technology in vehicular communication faces certain
technical challenges related to the requirements on the backhaul network, millimetre wave
communication and security issues [198]. Recently, the massive number of devices access-
ing the C-V2X network will cause congestion over the network; therefore, non-orthogonal
multiple access (NOMA) is also considered for V2X communication [199]. However,
NOMA techniques cannot be straightforwardly applied to C-V2X due to reliability and
security constraints. Recently, Riaz et al. [200] (see also the references therein) looked to
improve the reliability of NOMA which can be applied to C-V2X or V2V communication.
Furthermore, these AVs will help form an Internet of Vehicles (IoV) which will generate
big amount of data. Recently, artificial intelligence (AI), machine learning (ML) and deep
learning (DL) techniques have been used to improve the state and position estimation and
reference therein [15,201–203]. Different AI, ML and DL frameworks have been proposed
to learn and adapt these matrices online. Techniques such as artificial neural networks
(ANNs), adaptive neuro-fuzzy inference system (ANFIS), convolutional neural networks
(CNN), deep belief networks (DBN), deep inference for covariance estimation (DICE),
long short-term memory (LSTM), recurrent neural networks (RNN) and reinforcement
learning have been adopted to estimate the noise co-variance and the transition matrix
to improve the position accuracy of the AV [102,203–210]. Further detail and discussion
regarding these AI, ML and DL techniques for sensor fusion for AV can be found in [203]
(and references therein). Hopefully, all these challenges will be rectified in the future paving
the way for AVs and then finally IoVs.

6.2. Decision and Planning Strategies

As previously mentioned, self-driving vehicles and AVs must achieve Level 5 au-
tomation in order to be able to drive without any human intervention [7]. As a result,
vehicles must be intelligent enough to communicate within the traditional transport system
and the surrounding [8]. To achieve this, the AVs need to undergo a transition period in
which they will have to co-exist with the existing non-autonomous vehicles. Therefore,
implementation requires careful and gradual planning both at vehicular and infrastructure
level [38].
For the vehicular communication network development, a key factor is generating a
heterogeneous model in which different AV communication technologies can be stacked
together to perform versatile tasks [40]. For this purpose, DSRC technology seems to be
the most appropriate technology for V2V communication, while LTE technology can be
used for the implementation of V2I communication. Nevertheless, there are many open
issues such as handover, big data management, cross layer design and vehicular cloud
computing systems, which still require improvement [38].
Various driving strategies that the AVs must adapt in order to connect and coordinate
the information, received from the environment and from other vehicles, are presented
in [211]. These driving strategies not only facilitate the implementation of AVs, but also
define the basic vehicular competencies needed for traffic flow and safety. They are
explained in Table 9.
In addition, there are three main challenges for V2X communication in AVs, which
need to be taken into account in any future implementation strategy [212]. Firstly, there
is the high cost required for developing and deploying the proper road infrastructure
for an ITS. Secondly, VANETs expose critical vehicular information and, thus, require
better privacy protection mechanisms. Thirdly, it can be very challenging to build robust
Sensors 2021, 21, 706 23 of 33

VANETs which ensure reliable data transmission between AVs with high mobility and
varying location.

Table 9. Types of driving strategies.

Driving Strategy Explanation

Defensive Holds negative assumptions about its neighbouring AVs.
Competitive Adopts positive assumptions about its neighbouring AVs.
Negotiated Negotiates with other AVs based on its implicit decisions.
Cooperative Communicates with other AVs and accepts unified driving dispatch commands.

Finally, with regards to the infrastructure and environmental aspects, the investment
in development of smart cities should incorporate the development of AVs and ITS [213].
In addition, to achieve an ITS, the existing infrastructure is lacking features and requires
significant up-gradation, further research and development along with financial invest-
ment [214].
Author Contributions: Conceptualisation, M.N.A. and Q.Z.A.; Methodology, M.N.A. and Q.Z.A.;
Writing—original draft preparation, M.N.A.; Writing—review and editing, Q.Z.A., F.A.K. and M.H.;
and Supervision, Q.Z.A. and M.H. All authors have read and agreed to the published version of
the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not Applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

2D 2 Dimensional
3D 3 Dimensional
3GPP 3rd Generation Partnership Project
5G-NR Fifth Generation-New Radio
ACC Adaptive Cruise Control
ADAF Adaptive Detection And recognition Framework
AEB Automatic Emergency Braking
AI Artificial Intelligence
ANFIS Adaptive Neuro-Fuzzy Inference System
ANN Artificial Neural Network
AUTOSAR AUTomotive Open System ARchitecture
AU Application Unit
AV Autonomous Vehicle
BASt German Federal Highway Research Institute
BLE Bluetooth Low Energy
BPSK Binary Phase Shift Keying
BSD Blind Spot Detection
CCD Charge Coupled Device
CCH Control CHannel
CNN Convolutional Neural Networks
CMOS Complementary Metal Oxide Semiconductor
CRC Cyclic Redundancy Check
Sensors 2021, 21, 706 24 of 33

CSMA/CA Carrier-Sense Multiple Access with Collision Avoidance

C-V2X Cellular Vehicle To Everything
DA-ANN Deep Auto-encoder Artificial Neural Network
DBN Deep Belief Networks
DICE Deep Inference for Covariance Estimation
DR Dead Reckoning
DSRC Dedicated Short-Range Communication
EDCA Enhanced Distribution Channel Access
eMBB enhanced Mobile BroadBand
FCTA Forward Cross Traffic Alert
FCWS Forward Collision Warning System
FDMA Frequency Division Multiple Access
FHSS Frequency Hopping Spread Spectrum
FoV Field of View
GFSK Gaussian Frequency Shift Keying
GNSS Global Navigation Satellite System
GPS Global Positioning System
HetVNet Heterogenous Vehicular Network
ID IDentification
IMU Inertial Measurement Unit
IoT Internet of Things
IoV Internet of Vehicles
IR InfraRed
ITS Intelligent Transport Systems
IVC Inter-Vehicle Communication
LBAC Laser Based Adaptive Cruise
LCA Lane Change Assistance
LDPC Low Density Parity Check
LDWS Lane Departure Warning System
LiDAR Light Detection and Ranging
LoS Line of Sight
LRWPAN Low Rate Wireless Personal Area Network
LSTM Long Short-Term Memory
LTE Long-Term Evolution
LTE-A Long-Term Evolution-Advanced
MAC Media Access Control
MEC Multi-access Edge Computing
MEMS Micro-ElectroMechanical System
MIVC Multi-hop Inter-Vehicle Communication
ML Machine Learning
mMTC massive Machine Type Communication
mmwave Millimetre Wave
NFC Near Field Communication
NHSTA National Highway Traffic Safety Administration
NLoS Non-Line of Sight
NOMA Non-Orthogonal Multiple Access
NS2 Network Simulator 2
OBU On Board Unit
OICA International Organisation of Motor Vehicle Manufacturer
OCB Outside the Context of a Basic service set
OFDM Orthogonal Frequency Division Multiplexing
O-QPSK Orthogonal Quadrature Phase Shift Keying
OMNeT Objective Modular Network Testbed in C++
PCW Pedestrian Collision Warning
PDA Personal Digital Assistance
PSFCH Physical Sidelink Feedback CHannel
QAM Quadrature Amplitude Modulation
Sensors 2021, 21, 706 25 of 33

QoS Quality of Service

RADAR Radio-Detection and Ranging
RCTA Rear Cross Traffic Alert
RGB Red, Green and Blue
RGBD Red, Green, Blue, and Depth
RMSE Root Mean Square Error
RNN Recurrent Neural Networks
RSSI Received Signal Strength Indicator
RSU Road Side Unit
RTK Real Time Kinetics
RVC Roadside to Vehicle Communication
SAE Society of Automotive Engineers
SC-FDMA Single Carrier-Frequency Division Multiple Access
SIVC Single-hop Inter-Vehicle Communication
SPA Self-Parking Assistance
SRVC Sparse Roadside to Vehicle Communication
SUMO Simulation of Urban MObility
SVM Support Vector Machine
TA Trusted Authority
TDMA Time Division Multiple Access
ToF Time of Flight
TSR Traffic Sign Recognition
UOR Unknown Obstacles Recognition
uRLLC ultra-Reliable Low Latency Communication
URVC Ubiquitous Roadside to Vehicle Communication
UWB Ultra-Wide Band
V2D Vehicle to Device
V2G Vehicle to Grid
V2I Vehicle to Infrastructure
V2P Vehicle to Pedestrian
V2R Vehicle to Roadside
V2V Vehicle to Vehicle
V2X Vehicle to Everything
VANET Vehicular Ad-hoc network
VPL Vehicle Positioning and Localisation
WAVE Wireless Access in Vehicular Environment
Wi-Fi Wireless Fidelity
WLAN Wireless Local Area Network

References
1. Kaiwartya, O.; Abdullah, A.H.; Cao, Y.; Altameem, A.; Prasad, M.; Lin, C.-T.; Liu, X. Internet of Vehicles: Motivation, Layered
Architecture, Network Model, Challenges, and Future Aspects. IEEE Access 2016, 4, 5356–5373. [CrossRef]
2. Petit, J.; Schaub, F.; Feiri, M.; Kargl, F. Pseudonym Schemes in Vehicular Networks: A Survey. IEEE Commun. Surv. Tutor. 2015, 17,
228–255. [CrossRef]
3. Qazi, S.; Sabir, F.; Khawaja, B.A.; Atif, S.M.; Mustaqim, M. Why is Internet of Autonomous Vehicles not as Plug and Play as We
Think? Lessons to Be Learnt From Present Internet and Future Directions. IEEE Access 2020, 8, 133015–133033. [CrossRef]
4. Arena, F.; Pau, G. An Overview of Vehicular Communications. Future Internet 2019, 11, 27. [CrossRef]
5. D’Orey, P.M.; Ferreira, M. ITS for Sustainable Mobility: A Survey on Applications and Impact Assessment Tools. IEEE Trans.
Intell. Transp. Syst. 2014, 15, 477–493. [CrossRef]
6. Ondruša, J.; Kollab, E.; Verta, P.; Šarić, Ž. How Do Autonomous Cars Work? Transp. Res. Procedia 2020, 44, 226–233. [CrossRef]
7. Ross, P.E. Robot, you can drive my car; Autonomous driving will push humans into the passenger seat. IEEE Spectr. 2014, 51,
60–90. [CrossRef]
8. Rosenzweig, J.; Bartl, M. A Review and Analysis of Literature on Autonomous Driving. Available online: https://michaelbartl.
com/wp-content/uploads/Lit-Review-AD_MoI.pdf (accessed on 20 January 2021).
9. Zheng, Y.; Zhang, Y.; Ran, B.; Xu, Y.; Qu, X. Cooperative control strategies to stabilise the freeway mixed traffic stability and
improve traffic throughput in an intelligent roadside system environment. IET Intell. Transp. Syst. 2020, 14, 1108–1115. [CrossRef]
10. Xie, G.; Li, Y.; Han, Y.; Xie, Y.; Zeng, G.; Li, R. Recent Advances and Future Trends for Automotive Functional Safety Design
Methodologies. IEEE Trans. Ind. Inform. 2020, 16, 5629–5642. [CrossRef]
Sensors 2021, 21, 706 26 of 33

11. Karagiannis, G.; Altintas, O.; Ekici, E.; Heijenk, G.; Jarupan, B.; Lin, K.; Weil, T. Vehicular Networking: A Survey and Tutorial on
Requirements, Architectures, Challenges, Standards and Solutions. IEEE Commun. Surv. Tutor. 2011, 13, 584–616. [CrossRef]
12. Zanchin, B.C.; Adamshuk, R.; Santos, M.M.; Collazos, K.S. On the instrumentation and classification of autonomous cars. In
Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics (SMC), Banff, AB, Canada, 5–8 October 2017;
pp. 2631–2636.
13. Zhao, L.; Li, X.; Gu, B.; Zhou, Z.; Mumtaz, S.; Frascolla, V.; Gacanin, H.; Ashraf, M.I.; Rodriguez, J.; Yang, M.; et al. Vehicular
Communications: Standardization and Open Issues. IEEE Commun. Stand. Mag. 2018, 2, 74–80. [CrossRef]
14. Zong, W.; Zhang, C.; Wang, Z.; Zhu, J.; Chen, Q. Architecture Design and Implementation of an Autonomous Vehicle. IEEE Access
2018, 6, 21956–21970. [CrossRef]
15. Wang, Z.; Wu Y.; Niu, Q. Multi-Sensor Fusion in Automated Driving: A Survey. IEEE Access 2020, 8, 2847–2868. [CrossRef]
16. Llorca, F.D.; Daza, G.I.; Parra, N.H.; Alonso, I.P. Sensors and Sensing for Intelligent Vehicles. Sensors 2020, 20, 5115. [CrossRef]
[PubMed]
17. Xiao, Z.; Yang, D.; Wen, F.; Jiang, K. A Unified Multiple-Target Positioning Framework for Intelligent Connected Vehicles. Sensors
2019, 19, 1967. [CrossRef]
18. Noor-A-Rahim, M.; Liu, Z.; Lee, H.; Ali, G.G.M.N.; Pesch, D.; Xiao, P. A Survey on Resource Allocation in Vehicular Networks.
IEEE Trans. Intell. Transp. Syst. 2020, 1–21. [CrossRef]
19. Sawant, H.; Tan, J.; Yang, Q.; Wang, Q. Using Bluetooth and Sensor Networks for Intelligent Transportation Systems. In
Proceedings of the 2004 IEEE Intelligent Transportation Systems Conference (ITSC), Washington, DC, USA, 3–6 October 2004.
20. Lin, J.; Talty, T.; Tonguz, O.K. On the potential of bluetooth low energy technology for vehicular applications. IEEE Commun. Mag.
2015, 53, 267–275. [CrossRef]
21. Tsai, H.; Tonguz, O.K.; Saraydar, C.; Talty, T.; Ames M.; Macdonald, A. ZigBee-based intra-car wireless sensor networks: A case
study. IEEE Wirel. Commun. 2007, 14, 67–77. [CrossRef]
22. Ahmed, Q.Z.; Park, K.; Alouini, M.S. Ultrawide Bandwidth Receiver Based on a Multivariate Generalized Gaussian Distribution.
IEEE Trans. Wirel. Commun. 2015, 14, 1800–1810. [CrossRef]
23. Sichitiu, M.L.; Kihl, M. Inter-vehicle communication systems: A survey. IEEE Commun. Surv. Tutor. 2008, 10, 88–105. [CrossRef]
24. MacHardy, Z.; Khan, A.; Obana, K.; Iwashina, S. V2X Access Technologies: Regulation Research and Remaining Challenges. IEEE
Commun. Surv. Tutor. 2018, 20, 1858–1877. [CrossRef]
25. Zeng, X.; Tao, C.; Chen, Z. The application of DSRC technology in Intelligent Transportation System. In Proceedings of the IET
International Communication Conference on Wireless Mobile and Computing (CCWMC 2009), Shanghai, China, 7–9 December
2009; pp. 265–268.
26. Kenney, J.B. Dedicated Short-Range Communications (DSRC) Standards in the United States. Proc. IEEE 2011, 99, 1162–1182.
[CrossRef]
27. Strom, E.G. On Medium Access and Physical Layer Standards for Cooperative Intelligent Transport Systems in Europe. Proc.
IEEE 2011, 99, 1183–1188. [CrossRef]
28. Eenennaam, M.V.; Venis A.; Karagiannis, G. Impact of IEEE 1609.4 channel switching on the IEEE 802.11p beaconing performance.
In Proceedings of the IFIP Wireless Days, Dublin, Ireland, 21–23 November 2012; pp. 1–8.
29. Habib, S.; Hannan, M.A.; Javadi, M.S.; Samad, S.A.; Muad, A.M.; Hussain, A. Inter-Vehicle Wireless Communications Technologies,
Issues and Challenges. Inf. Technol. J. 2013, 12, 558–568.
30. Abboud, K.; Omar, H.; Zhuang, W. Interworking of DSRC and Cellular Network Technologies for V2X Communications: A
Survey. IEEE Trans. Veh. Technol. 2016, 65, 9457–9470. [CrossRef]
31. Valgas, J.; Martín-Sacristán, D.; Monserrat, J. 5G New Radio Numerologies and their Impact on V2X Communications. Available
online: http://www.iteam.upv.es/waves/2018/02-waves2018.pdf (accessed on 18 November 2020).
32. Naik, G.; Choudhury, B.; Park, J.-M. IEEE 802.11bd & 5G NR V2X: Evolution of Radio Access Technologies for V2X Communica-
tions. IEEE Access 2019, 7, 70169–70184.
33. Chen, S.; Hu, J.; Shi, Y.; Zhao, L.; Li, W. A Vision of C-V2X: Technologies, Field Testing, and Challenges With Chinese Development.
IEEE Internet Things J. 2020, 7, 3872–3881. [CrossRef]
34. Araniti, G.; Campolo, C.; Condoluci, M.; Iera, A.; Molinaro, A. LTE for Vehicular Networking: A Survey. IEEE Commun. Mag.
2013, 51, 148–157. [CrossRef]
35. Sevlian, R.; Chun, C.; Tan, I.; Bahai, A.; Laberteaux, K. Channel Characterization for 700 MHz DSRC Vehicular Communication. J.
Electr. Comput. Eng. 2010, 2010, 840895. [CrossRef]
36. Dorle, S.S.; Deshpande, D.M.; Keskar, A.G.; Chakole, M. Vehicle Classification and Communication Using ZigBee Protocol. In
Proceedings of the 3rd International Conference on Emerging Trends in Engineering and Technology, Goa, India, 19–21 November
2010; pp. 106–109.
37. Jo, K.; Kim, J.; Kim, D.; Jang, C.; Sunwoo, M. Development of Autonomous Car—Part I: Distributed System Architecture and
Development Process. IEEE Trans. Ind. Electron. 2014, 61, 7131–7140. [CrossRef]
38. Zheng, K.; Zheng, Q.; Chatzimisios, P.; Xiang, W.; Zhou, Y. Heterogeneous Vehicular Networking: A Survey on Architecture,
Challenges, and Solutions. IEEE Commun. Surv. Tutor. 2015, 17, 2377–2396. [CrossRef]
39. De Cerio, D.P.-D.; Valenzuela, J.L. Provisioning Vehicular Services and Communications Based on a Bluetooth Sensor Network
Deployment. Sensors 2015, 15, 12765–12781. [CrossRef] [PubMed]
Sensors 2021, 21, 706 27 of 33

40. Gao, S.; Lim, A.; Bevly, D. An empirical study of DSRC V2V performance in truck platooning scenarios. Digit. Commun. Netw.
2016, 2, 233–244. [CrossRef]
41. Lei, Y.; Wu, J. Study Of applying ZigBee technology into forward collision System (FCWS) under low speed circumstances. In
Proceedings of the 25th IEEE Wireless and Optical Communication Conference (WOCC), Chengdu, China, 21–23 May 2016;
pp. 1–4.
42. Fitah, A.; Badri, A.; Moughit, M.; Sahel, A. Performance of DSRC and WIFI for Intelligent Transport Systems in VANET. Procedia
Comput. Sci. 2018, 127, 360–368. [CrossRef]
43. Rosique, F.; Navarro, P.; Fernández, C.; Padilla, A. A Systematic Review of Perception System and Simulators for Autonomous
Vehicles Research. Sensors 2019, 29, 648. [CrossRef] [PubMed]
44. Sheikh, M.S.; Liang, J. A Comprehensive Survey on VANET Security Services in Traffic Management System. Wirel. Commun.
Mob. Comput. 2019, 2019, 2423915. [CrossRef]
45. Hafeez, M.; Elmirghani, J.M.H. Dynamic Spectrum Leasing for Bi-Directional Communication: Impact of Selfishness. IEEE Trans.
Commun. 2016, 64, 2427–2437. [CrossRef]
46. Jo, K.; Kim, J.; Kim, D.; Jang, C.; Sunwoo, M. Development of Autonomous Car—Part II: A Case Study on the Implementation of
an Autonomous Driving System Based on Distributed Architecture. IEEE Trans. Ind. Electron. 2015, 62, 5119–5132. [CrossRef]
47. Carullo, A.; Parvis, M. An ultrasonic sensor for distance measurement in automotive applications. IEEE Sens. J. 2001, 1, 143–147.
[CrossRef]
48. Kim, H.; Lee, J.-H.; Kim, S.-W.; Ko, J.-I.; Cho, D. Ultrasonic vehicle detector for side-fire implementation and extensive results
including harsh conditions. IEEE Trans. Intell. Transp. Syst. 2001, 2, 127–134.
49. Agarwal, V.; Murali, N.V.; Chandramouli, C. A Cost-Effective Ultrasonic Sensor-Based Driver-Assistance System for Congested
Traffic Conditions. IEEE Trans. Intell. Transp. Syst. 2009, 10, 486–498. [CrossRef]
50. Paidi, V.; Fleyeh, H.; Håkansson, J.; Nyberg, R.G. Smart parking sensors, technologies and applications for open parking lots: A
review. IET Intell. Transp. Syst. 2018, 12, 735–741. [CrossRef]
51. Alluhaibi, O.; Ahmed, Q.Z.; Kampert, E.; Higgins, M.D.; Wang, J. Revisiting the Energy-Efficient Hybrid D-A Precoding and
Combining Design for mm-Wave Systems. IEEE Trans. Green Commun. Netw. 2020, 4, 340–354. [CrossRef]
52. Farooq, A.; Ahmed, Q.Z.; Alade, T. Indoor Two Way Ranging using mm-Wave for Future Wireless Networks. In Proceedings
of the Emerging Tech (EMiT) Conference 2019, Huddersfield, UK, 9–11 April 2019; University of Huddersfield: Huddersfield,
UK, 2019.
53. Li, Z.; Xiang, L.; Ge, X.; Mao, G.; Chao, H.-C. Latency and Reliability of mmWave Multi-Hop V2V Communications Under Relay
Selections. IEEE Trans. Veh. Technol. 2020, 69, 9807–9821. [CrossRef]
54. Alluhaibi, O.; Ahmed, Q.Z.; Pan, C.; Zhu, H. Hybrid Digital-to-Analog Beamforming Approaches to Maximise the Capacity of
mm-Wave Systems. In Proceedings of the IEEE 85th Vehicular Technology Conference (VTC Spring), Sydney, NSW, Australia, 4–7
June 2017; pp. 1–5.
55. Alencar, F.A.; Rosero, L.A.; Filho, C.M.; Osorio, F.S.; Wolf, D.F. Fast metric tracking by detection system: Radar blob and camera
fusion. In Proceedings of the 2015 12th Latin American Robotics Symposium and 2015 3rd Brazilian Symposium on Robotics
(LARS-SBR), Uberlandia, Brazil, 29–31 October 2015; Volume 174, pp. 120–125.
56. Etinger, A.; Balal, N.; Litvak, B.; Einat, M.; Kapilevich, B.; Pinhasi, Y. Non-imaging MM-wave FMCW sensor for pedestrian
detection. IEEE Sens. J. 2014, 14, 1232–1237. [CrossRef]
57. Lee, S.; Yoon, Y.-J.; Lee, J.-E.; Kim, S.-C. Human-vehicle classification using feature-based SVM in 77-GHz automotive FMCW
radar. IET Radar Sonar Navig. 2017, 11, 1589–1596. [CrossRef]
58. Wojtanowski, J.; Zygmunt, M.; Kaszczuk, M.; Mierczyk, Z.; Muzal, M. Comparison of 905 nm and 1550 nm semiconductor
laser rangefinders’ performance deterioration due to adverse environmental conditions. Opto Electron. Rev. 2014, 22, 183–190.
[CrossRef]
59. Li, Y.; Ibanez-Guzman, J. Lidar for Autonomous Driving: The Principles, Challenges, and Trends for Automotive Lidar and
Perception Systems. IEEE Signal Process. Mag. 2020, 37, 50–61. [CrossRef]
60. Asvadi, A.; Garrote, L.; Premebida, C.; Peixoto, P.; Nunes, U.J. Multimodal vehicle detection: Fusing 3D-LIDAR and color camera
data. Pattern Recognit. Lett. 2018, 115, 20–29. [CrossRef]
61. De Silva, V.; Roche, J.; Kondoz, A. Robust fusion of LiDAR and wide-angle camera data for autonomous mobile robots. Sensors
2018, 18, 2730. [CrossRef]
62. Zhao, S.; Yao, H.; Zhang, Y.; Wang, Y.; Liu, S. View-based 3D object retrieval via multi-modal graph learning. Signal Process. 2015,
112, 110–118. [CrossRef]
63. Zhao, J.; Xu, H.; Wu, J.; Zheng, Y.; Liu, H. Trajectory tracking and prediction of pedestrian’s crossing intention using roadside
LiDAR. IET Intell. Transp. Syst. 2019, 13, 789–795. [CrossRef]
64. Wang, H.; Wang, B.; Liu, B.; Meng, X.; Yang, G. Pedestrian recognition and tracking using 3D LiDAR for autonomous vehicle.
Robot. Auton. Syst. 2017, 88, 71–78. [CrossRef]
65. Rasshofer, R.; Spies, H.; Spies, M. Influences of weather phenomena on automotive laser radar systems. Adv. Radio Sci. 2011, 9, 49.
[CrossRef]
66. Wallace, A.M.; Halimi, A.; Buller, G.S. Full Waveform LiDAR for Adverse Weather Conditions. IEEE Trans. Veh. Technol. 2020, 69,
7064–7077. [CrossRef]
Sensors 2021, 21, 706 28 of 33

67. Zhang, Z. A flexible new technique for camera calibration. IEEE Trans. Pattern Anal. Mach. Intell. 2000, 22, 1330–1334. [CrossRef]
68. Kannala, J.; Brandt, S.S. A generic camera model and calibration method for conventional, wide-angle, and fish-eye lenses. IEEE
Trans. Pattern Anal. Mach. Intell. 2006, 28, 1335–1340. [CrossRef] [PubMed]
69. Swief, A.; El-Habrouk, M. A survey of automotive driving assistance systems technologies. In Proceedings of the 2018
International Conference on Artificial Intelligence and Data Processing (IDAP), Malatya, Turkey, 28–30 September 2018; pp. 1–12.
70. Baftiu, I.; Pajaziti, A.; Cheok, K.C. Multi-mode surround view for ADAS vehicles. In Proceedings of the 2016 IEEE International
Symposium on Robotics and Intelligent Sensors (IRIS), Tokyo, Japan, 17–20 December 2016; pp. 190–193.
71. Delavarian, M.; Reza Marouzi, O.; Hassanpour, H.; Parizi, R.M.; Khan, M.S. Multi-camera multiple vehicle tracking in urban
intersections based on multilayer graphs. IET Intell. Transp. Syst. 2020, 14, 1673–1690. [CrossRef]
72. Ferraz, P.A.P.; de Oliveira, B.A.G.; Ferreira, F.M.F.; Martins, C.A.P.d.S. Three-stage RGBD architecture for vehicle and pedestrian
detection using convolutional neural networks and stereo vision. IET Intell. Transp. Syst. 2020, 14, 1319–1327. [CrossRef]
73. De Souza, A.B.; Rego, P.A.L.; Carneiro, T.; Rodrigues, J.D.C.; Filho, P.P.R.; Souza, J.N.D.; Chamola, V.; Albuquerque, V.H.C.D.;
Sikdar, B. Computation Offloading for Vehicular Environments: A Survey. IEEE Access 2020, 8, 198214–198243. [CrossRef]
74. Reina, G.; Milella, A.; Rouveure, R. Traversability analysis for offroad vehicles using stereo and radar data. In Proceedings of the
2015 IEEE International Conference on Industrial Technology (ICIT), Seville, Spain, 17–19 March 2015; pp. 540–546.
75. Shimizu, T.; Kobayashi, K. Development of a person-searching algorithm using an omnidirectional camera and LiDAR for the
Tsukuba challenge. In Proceedings of the 2018 57th Annual Conference of the Society of Instrument and Control Engineers of
Japan (SICE), Nara, Japan, 11–14 September 2018; Volume 2, pp. 810–815.
76. Kwon, S.K.; Hyun, E.; Lee, J.; Lee, J.; Son, S.H. A Low-Complexity Scheme for Partially Occluded Pedestrian Detection Using
LIDAR-RADAR Sensor Fusion. In Proceedings of the IEEE 22nd International Conference on Embedded and Real-Time
Computing Systems and Applications (RTCSA), Daegu, Korea, 17–19 August 2016; p. 104.
77. Steinbaeck, J.; Steger, C.; Holweg, G.; Druml, N. Design of a low level radar and time-of-flight sensor fusion framework. In
Proceedings of the 2018 21st Euromicro Conference on Digital System Design (DSD), Prague, Czech Republic, 29–31 August 2018;
pp. 268–275.
78. Steinbaeck, J.; Strasser, A.; Steger, C.; Brenner, E.; Holweg, G.; Druml, N. Context-Aware Sensor Adaption of a Radar and Time-of-
Flight Based Perception Platform. In Proceedings of the Sensors Applications Symposium (SAS), Kuala Lumpur, Malaysia, 9–11
March 2020; pp. 1–6.
79. Psiaki, M.L.; Humphreys, T.E. GNSS Spoofing and Detection. Proc. IEEE 2016, 104, 1258–1270. [CrossRef]
80. Toledo-Moreo, R.; Zamora-Izquierdo, M.A.; Ubeda-Minarro, B.; Gomez-Skarmeta, A.F. High-Integrity IMM-EKF-Based Road
Vehicle Navigation With Low-Cost GPS/SBAS/INS. IEEE Trans. Intell. Transp. Syst. 2007, 8, 491–511. [CrossRef]
81. Kaplan, E.; Hegarty, C. Understanding GPS Principles and Applications, 2nd ed.; British Library Cataloguing in Publication Data;
Artech House: Boston, MA, USA, 2006; pp. 1–723.
82. Hughes, W.J. Global Positioning System (GPS) Standard Positioning Service (SPS) Performance Analysis Report; FAA GPS Performance
Analysis Report, Tech. Rep. 94; The University of Texas at Austin: Austin, TX, USA, July 2016.
83. Ahmad, K.A.; Sahmoudi, M.; Macabiau, C. Characterization of GNSS Receiver Position Errors for User Integrity Monitoring in Ur-
ban Environments. In Proceedings of the 2014 European Navigation Conference (ENC)-GNSS 2014, Rotterdam, The Netherlands,
14–17 April 2014.
84. MacGougan, G.; Lachapelle, G.; Klukas, R.; Siu, K.; Garin, L.; Shewfelt, J.; Cox, G. Performance analysis of a stand-alone
high-sensitivity receiver. GPS Solut. 2002, 6, 179–195.
85. Toledo-Moreo, R.; Betaille, D.; Peyret, F. Lane-Level Integrity Provision for Navigation and Map Matching With GNSS, Dead
Reckoning, and Enhanced Maps. IEEE Trans. Intell. Transp. Syst. 2010, 11, 100–112. [CrossRef]
86. Alam, N.; Tabatabaei Balaei, A.; Dempster, A.G. A DSRC Doppler-Based Cooperative Positioning Enhancement for Vehicular
Networks With GPS Availability. IEEE Trans. Veh. Technol. 2011, 60, 4462–4470. [CrossRef]
87. Zhao, S.; Chen, Y.; Farrell, J.A. High-Precision Vehicle Navigation in Urban Environments Using an MEM’s IMU and Single-
Frequency GPS Receiver. IEEE Trans. Intell. Transp. Syst. 2016, 17, 2854–2867. [CrossRef]
88. Jin, X.; Yu, Z.; Yin, G.; Wang, J. Improving Vehicle Handling Stability Based on Combined AFS and DYC System via Robust
TakagiSugeno Fuzzy Control. IEEE Trans. Intell. Transp. Syst. 2018, 19, 2696–2707. [CrossRef]
89. Ahmed, Q.Z.; Park, K.; Alouini, M.S.; Aïssa, S. Compression and Combining Based on Channel Shortening and Reduced-Rank
Techniques for Cooperative Wireless Sensor Networks. IEEE Trans. Veh. Technol. 2014, 63, 72–81. [CrossRef]
90. Bello, L.L.; Mariani, R.; Mubeen, S.; Saponara, S. Recent Advances and Trends in On-Board Embedded and Networked Automotive
Systems. IEEE Trans. Ind. Inform. 2019, 15, 1038–1051. [CrossRef]
91. Katsaggelos, A.K.; Bahaadini, S.; Molina, R. Audiovisual Fusion: Challenges and New Approaches. Proc. IEEE 2015, 103, 1635–1653.
[CrossRef]
92. Lovegrove, S.; Davison, A.J.; Ibañez-Guzmán, J. Accurate visual odometry from a rear parking camera. In Proceedings of the
IEEE Intelligent Vehicles Symposium (IV), Baden-Baden, Germany, 5–9 June 2011; pp. 788–793.
93. Faragher, R. Understanding the Basis of the Kalman Filter Via a Simple and Intuitive Derivation [Lecture Notes]. IEEE Signal
Process. Mag. 2012, 29, 128–132. [CrossRef]
94. Nair, N.R.; Sudheesh, P.; Jayakumar, M. 2-D airborne vehicle tracking using Kalman filter. In Proceedings of the International
Conference on Circuit, Power and Computing Technologies (ICCPCT), Nagercoil, India, 18–19 March 2016; pp. 1–9.
Sensors 2021, 21, 706 29 of 33

95. Chai, C.; Wong, Y.D. Automatic vehicle classification and tracking method for vehicle movements at signalized intersections. In
Proceedings of the IEEE Intelligent Vehicles Symposium (IV), Gold Coast, Australia, 23–26 June 2013; pp. 624–629.
96. Proakis, J.G.; Salehi, M. Digital Communications; McGraw-Hill: Boston, MA, USA, 2008.
97. Lefkopoulos, V.; Menner, M.; Domahidi, A.; Zeilinger, M.N. Interaction-Aware Motion Prediction for Autonomous Driving: A
Multiple Model Kalman Filtering Scheme. IEEE Robot. Autom. Lett. 2021, 6, 80–87. [CrossRef]
98. Simon, D.; Chia, T.L. Kalman filtering with state equality constraints. IEEE Trans. Aerosp. Electron. Syst. 2002, 38, 128–136.
[CrossRef]
99. Aidala, V.J. Kalman Filter Behavior in Bearings-Only Tracking Applications. IEEE Trans. Aerosp. Electron. Syst. 1979, AES-15, 29–39.
[CrossRef]
100. Jetto, L.; Longhi, S.; Venturini, G. Development and experimental validation of an adaptive extended Kalman filter for the
localization of mobile robots. IEEE Trans. Robot. Autom. 1999, 15, 219–229. [CrossRef]
101. Lin, M.; Yoon, J.; Kim, B. Self-Driving Car Location Estimation Based on a Particle-Aided Unscented Kalman Filter. Sensors 2020,
20, 2544. [CrossRef] [PubMed]
102. Gao, X.; Luo, H.; Ning, B.; Zhao, F.; Bao, L.; Gong, Y.; Xiao, Y.; Jiang, J. RL-AKF: An Adaptive Kalman Filter Navigation Algorithm
Based on Reinforcement Learning for Ground Vehicles. Remote Sens. 2020, 12, 1704. [CrossRef]
103. Wu, F.; Luo, H.; Jia, H.; Zhao, F.; Xiao, Y.; Gao, X. Predicting the Noise Covariance With a Multitask Learning Model for Kalman
Filter-Based GNSS/INS Integrated Navigation. IEEE Trans. Instrum. Meas. 2021, 70, 1–13.
104. Ngoc, T.T.; Khenchaf, A.; Comblet, F. Evaluating process and measurement noise in extended Kalman filter for GNSS position
accuracy. In Proceedings of the 13th European Conference Antennas Propagation, Krakow, Poland, 31 March–5 April 2019;
pp. 1–5.
105. Cui, B.; Wei, X.; Chen, X.; Li, J.; Li, L. On Sigma-Point Update of Cubature Kalman Filter for GNSS/INS Under GNSS-Challenged
Environment. IEEE Trans. Veh. Technol. 2019, 68, 8671–8682. [CrossRef]
106. Stengel, R.F. Stochastic Optimal Control: Theory and Application; John Wiley and Sons: New York, NY, USA, 1986.
107. Xu, B.; Zhang, P.; Wen, H.Z.; Wu, X. Stochastic stability and performance analysis of cubature Kalman filter. Neurocomputing 2016,
186, 218–227. [CrossRef]
108. Ray, L.R. Nonlinear state and tire force estimation for advanced vehicle control. IEEE Trans. Control. Syst. Technol. 1995, 3, 117–124.
[CrossRef]
109. Wenzel, T.A.; Burnham, K.J.; Blundell, M.V.; Williams, R.A. Dual extended Kalman filter for vehicle state and parameter estimation.
Veh. Syst. Dyn. 2006, 44, 153–171. [CrossRef]
110. Zong, C.; Dan, H.U.; Zheng, H. Dual Extended Kalman Filter for Combined Estimation of Vehicle State and Road Friction. Chin. J.
Mech. Eng. 2013, 26, 313–324. [CrossRef]
111. Julier, S.J.; Uhlmann, J.K. Unscented filtering and nonlinear estimation. Proc. IEEE 2004, 92, 401–422. [CrossRef]
112. Akhlaghi, S.; Zhou, N.; Huang, Z. Adaptive adjustment of noise covariance in Kalman filter for dynamic state estimation. In
Proceedings of the IEEE Power & Energy Society General Meeting, Chicago, IL, USA, 16–20 July 2017; pp. 1–5.
113. Wan, E.A.; Van der Merwe, R. The unscented kalman filter for nonlinear estimation. In Proceedings of the IEEE Adaptive Systems
for Signal Processing, Communications, and Control Symposium 2000, AS-SPCC, Lake Louise, AB, Canada, 4 October 2000;
pp. 153–158.
114. Wielitzka, M.; Dagen, M.; Ortmaier, T. State estimation of vehicle’s lateral dynamics using unscented Kalman filter. In Proceedings
of the IEEE 53rd Annual Conference on Decision and Control (CDC), Los Angeles, CA, USA, 15–17 December 2014; pp. 5015–5020.
115. Wielitzka, M.; Dagen, M.; Ortmaier, T. Joint unscented Kalman filter for state and parameter estimation in vehicle dynamics. In
Proceedings of the IEEE Conference on Control Applications (CCA), Sydney, Australia, 21–23 September 2015; pp. 1945–1950.
116. Cheng, Q.; Correa-Victorino, A.; Charara, A. A new nonlinear observer of sideslip angle with 5unknown vehicle parameter using
the dual unscented Kalman filter. In Proceedings of the 15th International IEEE Conference on Intelligent Transportation Systems,
Anchorage, AK, USA, 16–19 September 2012; pp. 1716–1724.
117. Arasaratnam, I.; Haykin, S. Cubature Kalman filters. IEEE Trans. Autom. Control 2009, 54, 1254–1269. [CrossRef]
118. Sun, Y.; Chen, Q. Joint estimation of states and parameters of vehicle model using cubature Kalman filter. In Proceedings of the
IEEE International Conference on Systems, Man, and Cybernetics (SMC), Budapest, Hungary, 9–12 October 2016; pp. 977–982.
119. Ahmed, Q.Z.; Alouini, M.; Aissa, S. Bit Error-Rate Minimizing Detector for Amplify-and-Forward Relaying Systems Using
Generalized Gaussian Kernel. IEEE Signal Process. Letters 2013, 20, 55–58. [CrossRef]
120. Arulampalam, M.S.; Maskell, S.; Gordon, N.; Clapp, T. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian
tracking. IEEE Trans. Signal Process. 2002, 50, 174–188. [CrossRef]
121. Caron, F.; Davy, M.; Duflos, E.; Vanheeghe, P. Particle Filtering for Multisensor Data Fusion With Switching Observation Models:
Application to Land Vehicle Positioning. IEEE Trans. Signal Process. 2007, 55, 2703–2719. [CrossRef]
122. Aggarwal, P.; Gu, D.; Nassar, S.; Syed, Z.; El-Sheimy, N. Extended particle filter (EPF) for INS/GPS land vehicle navigation
applications. In Proceedings of the Institute of Navigation Satellite Division Technical Meeting (ION GNSS 2007), Fort Worth, TX,
USA, 25–28 September 2007; pp. 2619–2626.
123. Hartenstein, H.; Laberteaux, L.P. A tutorial survey on vehicular ad hoc networks. IEEE Commun. Mag. 2008, 46, 164–171.
[CrossRef]
Sensors 2021, 21, 706 30 of 33

124. Toor, Y.; Muhlethaler, P.; Laouiti, A.; La Fortelle, A.D. Vehicle Ad Hoc networks: Applications and related technical issues. IEEE
Commun. Surv. Tutor. 2008, 10, 74–88. [CrossRef]
125. Li, F.; Wang, Y. Routing in vehicular ad hoc networks: A survey. IEEE Veh. Technol. Mag. 2007, 2, 12–22. [CrossRef]
126. Harri, J.; Filali, F.; Bonnet, C. Mobility models for vehicular ad hoc networks: A survey and taxonomy. IEEE Commun. Surv. Tutor.
2009, 11, 19–41. [CrossRef]
127. Zhang, R.; Yan, F.; Xia, W.; Xing, S.; Wu, Y.; Shen, L. An Optimal Roadside Unit Placement Method for VANET Localization. In
Proceedings of the Global Communications Conference (GLOBECOM), Singapore, 4–8 December 2017; pp. 1–6.
128. Zaidi, S.A.R.; Afzal, A.; Hafeez, M.; Ghogho, M.; Mclernon, D.C.; Swami, A. Solar energy empowered 5G cognitive metro-cellular
networks. IEEE Commun. Mag. 2015, 53, 70–77. [CrossRef]
129. Ferrag, M.A.; Maglaras, L.; Ahmim, A. Privacy-Preserving Schemes for Ad Hoc Social Networks: A Survey. IEEE Commun. Surv.
Tutor. 2017, 19, 3015–3045. [CrossRef]
130. Chour, H.; Nasser, Y.; Artail, H.; Kachouh, A.; Al-Dubai, A. VANET Aided D2D Discovery: Delay Analysis and Performance.
IEEE Trans. Veh. Technol. 2017, 66, 8059–8071. [CrossRef]
131. Abbas, F.; Fan, P.; Khan, Z. A Novel Low-Latency V2V Resource Allocation Scheme Based on Cellular V2X Communications.
IEEE Trans. Intell. Transp. Syst. 2019, 20, 2185–2197. [CrossRef]
132. Huang, Y.; Guan, X.; Cai, Z.; Ohtsuki, T. Multicast capacity analysis for social-proximity urban bus-assisted VANETs. In
Proceedings of the IEEE International Conference on Communications (ICC), Budapest, Hungary, 9–13 June 2013; pp. 6138–6142.
133. Lee, K.; Kim, J.; Park, Y.; Wang, H.; Hong, D. Latency of Cellular-Based V2X: Perspectives on TTI-Proportional Latency and
TTI-Independent Latency. IEEE Access 2017, 5, 15800–15809. [CrossRef]
134. Ahmed, Q.; Che, F.; Shakir, M.Z.; Ahmed, A. Artificial intelligence for localisation of ultra-wide bandwidth (UWB) sensor nodes.
In AI for Emerging Verticals: Human-Robot Computing, Sensing and Networking (Computing and Networks); Shakir, M.Z., Ramzan, N.,
Eds.; IET: London, UK, 2020.
135. Mamdouhi, H.; Khatun, S.; Zarrin, J. Bluetooth Wireless Monitoring, Managing and Control for Inter Vehicle in Vehicular Ad-Hoc
Networks. J. Comput. Sci. 2009, 5, 922–929. [CrossRef]
136. Jeon, K.E.; She, J.; Soonsawad, P.; Ng, P.C. BLE Beacons for Internet of Things Applications: Survey, Challenges, and Opportunities.
IEEE Internet Things J. 2018, 5, 811–828. [CrossRef]
137. Wang, J.; Liu, J.; Kato, N. Networking and Communications in Autonomous Driving: A Survey. IEEE Commun. Surv. Tutor. 2019,
21, 1243–1274. [CrossRef]
138. Mirza, N.; Khan, A.N. Bluetooth Low Energy based Communication Framework for Intra Vehicle Wireless Sensor Networks. In
Proceedings of the International Conference on Frontiers of Information Technology (FIT), Islamabad, Pakistan, 18–20 December
2017; pp. 29–34.
139. Bronzi, W.; Frank, R.; Castignani, G.; Engel, T. Bluetooth low energy for inter-vehicular communications. In Proceedings of the
IEEE Vehicular Networking Conference (VNC), Paderborn, Germany, 3–5 December 2014; pp. 215–221.
140. Cao, Y.; Lu, X.; Zhao, Z.; Ji, X.; Yang, J.; Pang, X. A Comparative Study of BLE-based Fingerprint Localization for Vehicular
Application. In Proceedings of the Ubiquitous Positioning, Indoor Navigation and Location-Based Services (UPINLBS), Wuhan,
China, 22–23 March 2018; pp. 1–7.
141. Onishi, H.; Wu, K.; Yoshida, K.; Kato, T. Approaches for vehicle cyber-security in the US: Vulnerabilities of carry-in devices,
GNSS, & vehicle-to-vehicle communication. Int. J. Automot. Eng. 2017, 8, 1–6.
142. Seri, B.; Vishnepolsky, G. BlueBorne; Technical Report; Armis: Palo Alto, CA, USA, 2017
143. El-Rewini, Z.; Sadatsharan, K.; Selvaraj, D.F.; Plathottam, S.J.; Ranganathan, P. Cybersecurity challenges in vehicular communica-
tions. Veh. Commun. 2020, 23, 100214. [CrossRef]
144. Si, W.; Starobinski, D.; Laifenfeld, M. A Robust Load Balancing and Routing Protocol for Intra-Car Hybrid Wired/Wireless
Networks. IEEE Trans. Mob. Comput. 2019, 18, 250–263. [CrossRef]
145. Wheeler, A. Commercial Applications of Wireless Sensor Networks Using ZigBee. IEEE Commun. Mag. 2007, 45, 70–77. [CrossRef]
146. Lee, J.-S.; Su, Y.-W.; Shen, C.-C. A Comparative Study of Wireless Protocols:Bluetooth, UWB, ZigBee, and Wi-Fi. In Proceedings of
the 33rd Annual Conference of the IEEE Industrial Electronics Society (IECON), Taipei, Taiwan, 5–8 November 2007; pp. 46–51.
147. Park, C.; Rappaport, T.S. Short-Range Wireless Communications for Next-Generation Networks: UWB, 60 GHz Millimeter-Wave
WPAN, And ZigBee. IEEE Wirel. Commun. 2007, 14, 70–78. [CrossRef]
148. Parthasarathy, D.; Whiton, R.; Hagerskans, J.; Gustafsson, T. An in-vehicle wireless sensor network for heavy vehicles. In
Proceedings of the IEEE 21st International Conference on Emerging Technologies and Factory Automation (ETFA), Berlin,
Germany, 6–9 September 2016; pp. 1–8.
149. Pawade, S.; Shah, S.; Tijare, D. ZigBee based intelligent driver assistance system. Int. J. Eng. Res. Appl. 2013, 3, 1463–1468.
150. Yassein, M.B.; Al-Maolegi, M.; Khamayseh, Y.; Krstic, D.; Aljawarneh, S. Smart System for Busy Roads Using Short Range Wireless
Technologies. Int. J. Math. Comput. Simul. 2015, 9, 146–152.
151. Zhang, K.; Zhang, L.; Lu, F.; Zhao, Y. Distance measurement algorithm for freeway vehicles based on ZigBee technology. In
Proceedings of the IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC),
Chongqing, China, 25–26 March 2017; pp. 2007–2010.
Sensors 2021, 21, 706 31 of 33

152. Ahmed, Q.Z.; Yang, L. Performance of Hybrid Direct-Sequence Time-Hopping Ultrawide Bandwidth Systems in Nakagami-m Fad-
ing Channels. In Proceedings of the IEEE 18th International Symposium on Personal, Indoor and Mobile Radio Communications,
Athens, Greece, 3–7 September 2007; pp. 1–5.
153. Ahmed, Q.Z.; Yang, L. Reduced-rank adaptive multiuser detection in hybrid direct-sequence time-hopping ultrawide bandwidth
systems. IEEE Trans. Wirel. Commun. 2010, 9, 156–167. [CrossRef]
154. Ahmed, Q.Z.; Yang, L.; Chen, S. Reduced-Rank Adaptive Least Bit-Error-Rate Detection in Hybrid Direct-Sequence Time-Hopping
Ultrawide Bandwidth Systems. IEEE Trans. Veh. Technol. 2011, 60, 849–857. [CrossRef]
155. San Martín, J.; Cortés, A.; Zamora-Cadenas, L.; aSvensson, B.J. Precise Positioning of Autonomous Vehicles Combining UWB
Ranging Estimations with On-Board Sensors. Electronics 2020, 9, 1238. [CrossRef]
156. Che, F.; Ahmed, A.; Ahmed, Q.Z.; Zaidi, S.A.R.; Shakir, M.Z. Machine Learning Based Approach for Indoor Localization Using
Ultra-Wide Bandwidth (UWB) System for Industrial Internet of Things (IIoT). In Proceedings of the International Conference on
UK-China Emerging Technologies (UCET), Glasgow, UK, 20–21 August 2020; pp. 1–4.
157. Wymeersch, H.; Marano, S.; Gifford, W.M.; Win, M.Z. A Machine Learning Approach to Ranging Error Mitigation for UWB
Localization. IEEE Trans. Commun. 2012, 60, 1719–1728. [CrossRef]
158. Laurendeau, C.; Barbeau, M. Threats to security in DSRC/WAVE. In Lecture Notes in Computer Science, Proceedings of the Ad-Hoc,
Mobile, and Wireless Networks, Ottawa, QC, Canada, 17–19 August 2006; Springer: Berlin/Heidelberg, Germany, 2006; pp. 266–279.
159. Biswas, S.; Mišic, J.; Mišic, V. DDoS attack on WAVE-enabled VANET through synchronization. In Proceedings of the Globecom
2012: Communication and Information System Security Symposium, Anaheim, CA, USA, 3–7 December 2012; pp. 1079–1084.
160. Whyte, W.; Petit, J.; Kumar, V.; Moring, J.; Roy, R. Threat and Countermeasures Analysis for WAVE Service Advertisement. In
Proceedings of the IEEE 18th International Conference on Intelligent Transportation Systems, Las Palmas, Spain, 15–18 September
2015; pp. 1061–1068.
161. Dey, K.C.; Rayamajhi, A.; Chowdhury, M.; Bhavsar, P.; Martin, J. Vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I)
communication in a heterogeneous wireless network - performance evaluation. Transp. Res. Part C Emerg. Technol. 2016, 68,
168–184. [CrossRef]
162. Biswas, S.; Tatchikou, R.; Dion, F. Vehicle-to-vehicle wireless communication protocols for enhancing highway traffic safety. IEEE
Commun. Mag. 2006, 44, 74–82. [CrossRef]
163. Ohyama, T.; Nakabayashi, S.; Shiraki, Y.; Tokuda, K. A Study of Real-Time and Autonomous Decentralized DSRC System for
Inter-Vehicle Communications. In Proceedings of the IEEE Intelligent Transportation Systems (ITSC), Dearborn, MI, USA, 1–3
October 2000; pp. 190–195.
164. Reddy, G.R.; Ramanathan, R. An Empirical study on MAC layer in IEEE 802.11p/WAVE based Vehicular Ad hoc Networks.
Procedia Comput. Sci. 2018, 143, 720–727. [CrossRef]
165. Molina-Masegosa, R.; Gozalvez, J. LTE-V for Sidelink 5G V2X Vehicular Communications: A New 5G Technology for Short-Range
Vehicle-to-Everything Communications. IEEE Veh. Technol. Mag. 2017, 12, 30–39. [CrossRef]
166. Gonzalez-Martín, M.; Sepulcre, M.; Molina-Masegosa, R.; Gozalvez, J. Analytical Models of the Performance of C-V2X Mode 4
Vehicular Communications. IEEE Trans. Veh. Technol. 2019, 68, 1155–1166. [CrossRef]
167. Hegde, A.; Festag, A. Mode Switching Strategies in Cellular-V2X. IFAC-PapersOnLine 2019, 52, 81–86. [CrossRef]
168. Cao, J.; Ma, M.; Li, H.; Zhang, Y.; Luo, Z. A Survey on Security Aspects for LTE and LTE-A Networks. IEEE Commun. Surv. Tutor.
2014, 16, 283–302. [CrossRef]
169. Muhammad, M.; Safdar, G.A. Survey on existing authentication issues for cellular-assisted V2X communication. Veh. Commun.
2018, 12, 50–65. [CrossRef]
170. Kiela, K.; Barzdenas, V.; Jurgo, M.; Macaitis, V.; Rafanavicius, J.; Vasjanov, A.; Kladovscikov, L.; Navickas, R. Review of V2X–IoT
Standards and Frameworks for ITS Applications. Appl. Sci. 2020, 10, 4314. [CrossRef]
171. Andrews, J.G.; Buzzi, S.; Choi, W.; Hanly, S.V.; Lozano, A.; Soong, A.C.K.; Zhang, J.C. What Will 5G Be? IEEE J. Sel. Areas Commun.
2014, 32, 1065–1082. [CrossRef]
172. Parkvall, S.; Dahlman, E.; Furuskar, A.; Frenne, M. NR: The New 5G Radio Access Technology. IEEE Commun. Stand. Mag. 2017,
1, 24–30. [CrossRef]
173. Bockelmann, C.; Pratas, N.; Nikopour, H.; Au, K.; Svensson, T.; Stefanovic, C.; Popovski, P.; Dekorsy, A. Massive machine-type
communications in 5 g: Physical and MAC-layer solutions. IEEE Commun. Mag. 2016, 54, 59–65. [CrossRef]
174. Park, J.; Bennis, M. URLLC-eMBB Slicing to Support VR Multimodal Perceptions over Wireless Cellular Systems. In Proceedings
of the IEEE Global Communications Conference (GLOBECOM), Abu Dhabi, UAE, 9–13 December 2018; pp. 1–7.
175. Boccardi, F.; Heath, R.W.; Lozano, A.; Marzetta, T.L.; Popovski, P. Five disruptive technology directions for 5G. IEEE Commun.
Mag. 2014, 52, 74–80. [CrossRef]
176. Hernandez-Aquino, R.; Zaidi, S.A.R.; McLernon, D.; Ghogho, M.; Imran, A. Tilt Angle Optimization in Two-Tier Cellular
Networks—A Stochastic Geometry Approach. IEEE Trans. Commun. 2015, 63, 5162–5177. [CrossRef]
177. Condoluci, M.; Gallo, L.; Mussot, L.; Kousaridas, A.; Spapis, P.; Mahlouji, M.; Mahmoodi, T. 5G V2X System-Level Architecture of
5GCAR Project. Future Internet 2019, 11, 217. [CrossRef]
178. Slamnik-Kriještorac, N.; de Britto e Silva, E.; Municio, E.; Carvalho de Resende, H.C.; Hadiwardoyo, S.A.; Marquez-Barja, J.M.
Network Service and Resource Orchestration: A Feature and Performance Analysis within the MEC-Enhanced Vehicular Network
Context. Sensors 2020, 20, 3852. [CrossRef] [PubMed]
Sensors 2021, 21, 706 32 of 33

179. Government Report: UK Connected and Autonomous Vehicle Research and Development Projects. 2018. Available on-
line: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/737778/ccav-
research-and-development-projects.pdf (accessed on 20 January 2021).
180. Erdelean, I.; Hedhli, A.; Lamb, M.; Strand, N.; Zofka, E. Catalogue of Connected and automated Driving Test Sites. 2019. Available
online: https://www.cedr.eu/download/D2.1-Catalogue-of-connected-and-automated-driving-test-sites.pdf (accessed on 20
January 2021).
181. Bezzina, D.; Sayer, J. Safety Pilot Model Deployment: Test Conductor Team Report; Report No. DOT HS 812. Available
online: https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/812171-safetypilotmodeldeploydeltestcondrtmrep.pdf (accessed on
20 January 2021).
182. NYC DoT. NYC Connected Vehicle Project for Safer Transportation. 2020. Available online: https://www.cvp.nyc/ (accessed on
20 January 2021).
183. U.S. DoT. Connected Vehicle Pilot Deployment Program—Tampa (THEA) Pilot. 2020. Available online: https://www.its.dot.
gov/pilots/pilots_thea.htm (accessed on 20 January 2021).
184. U.S. DoT. Connected Vehicle Pilot Deployment Program—Wyoming (WY) DOT Pilot. 2020. Available online: https://www.its.
dot.gov/pilots/pilots_wydot.htm (accessed on 20 January 2021).
185. Ameixieira, C.; Cardote, A.; Neves, F.; Meireles, R.; Sargento, S.; Coelho, L.; Afonso, J.; Areias, B.; Mota, E.; Costa, R.; et al.
Harbornet: A real-world testbed for vehicular networks. IEEE Commun. Mag. 2014, 52, 108–114. [CrossRef]
186. Gomez, L.R.P.; Fairfield, N.; Szybalski, A.; Nemec, P.; Urmson, C. Transitioning a Mixed-Mode Vehicle to Autonomous Mode.
U.S. Patent No. 8,078,349, 13 December 2011.
187. Wang, J.; Shao, Y.; Ge, Y.; Yu, R. A Survey of Vehicle to Everything (V2X) Testing. Sensors 2019, 19, 334. [CrossRef] [PubMed]
188. Kostopoulos, A.; Chochliouros, I.; Dardamanis, A.; Segou, O.; Kafetzakis, E.; Soua, R.; Zhang, K.; Kuklinski, S.; Tomaszewski, L.; Yi,
N.; et al. 5G Trial Cooperation Between EU and China. In Proceedings of the IEEE International Conference on Communications
Workshops (ICC Workshops), Shanghai, China, 20–24 May 2019; pp. 1–6.
189. Huang, P.; Zheng, P. BlueID: Enabling robust in-car localization and on-demand personalization using Bluetooth. In Proceedings
of the IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Montreal,
QC, Canada, 8–13 October 2017; pp. 1–7.
190. Emmanuel, R.; Das, A.P.; Sandhya, B.; Thomas, M.; Tripathi, A. Passenger Localization for In-Vehicle Personalization Using BLE
Beacons. In Proceedings of the IEEE 87th Vehicular Technology Conference (VTC Spring), Porto, Portugal, 3–6 June 2018; pp. 1–7.
191. Yang, X.; Lu, C.; Pan, W. A ZigBee-Based Highway Vehicles Prevent Collision Warning System Research. In Proceedings of the
Fourth International Symposium on Computational Intelligence and Design, Hangzhou, China, 28–30 October 2011; pp. 270–273.
192. Jiang, D.; Taliwal, V.; Meier, A.; Holfelder, W.; Herrtwich, R. Design of 5.9 ghz dsrc-based vehicular safety communication. IEEE
Wirel. Commun. 2006, 13, 36–43. [CrossRef]
193. Hussain, S.; Faizan, M.; Hayee, M.I. Real-Time Relative Lane and Position Identification of Surrounding Vehicles Using GPS and
DSRC Based Vehicle-to-Vehicle Communication. In Proceedings of the IEEE International Conference on Communications (ICC),
Kansas City, MO, USA, 20–24 May 2018; pp. 1–7.
194. Ashraf, S.A.; Blasco, R.; Do, H.; Fodor, G.; Zhang, C.; Sun, W. Supporting Vehicle-to-Everything Services by 5G New Radio
Release-16 Systems. IEEE Commun. Stand. Mag. 2020, 4, 26–32. [CrossRef]
195. Zhang, Q.; Sun, H.; Wei, Z.; Feng, Z. Sensing and Communication Integrated System for Autonomous Driving Vehicles. In
Proceedings of the IEEE INFOCOM—IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS),
Toronto, ON, Canada, 6–9 July 2020; pp. 1278–1279.
196. Serizawa, K.; Mikami, M.; Moto, K.; Yoshino, H. Field Trial Activities on 5G NR V2V Direct Communication Towards Application
to Truck Platooning. In Proceedings of the IEEE 90th Vehicular Technology Conference (VTC2019-Fall), Honolulu, HI, USA, 22–25
September 2019; pp. 1–5.
197. Mikami, M.; Ishida, Y.; Serizawa, K.; Nishiyori, H.; Moto, K.; Yoshino, H. Field Experimental Trial of Dynamic Mode Switching
for 5G NR-V2X Sidelink Communications towards Application to Truck Platooning. In Proceedings of the IEEE 91st Vehicular
Technology Conference (VTC2020-Spring), Antwerp, Belgium, 25–28 May 2020; pp. 1–5.
198. Patole, S.M.; Torlak, M.; Wang, D.; Ali, M. Automotive radars: A review of signal processing techniques. IEEE Signal Process. Mag.
2017, 34, 22–35. [CrossRef]
199. Di, B.; Song, L.; Li, Y.; Han, Z. V2X Meets NOMA: Non-Orthogonal Multiple Access for 5G-Enabled Vehicular Networks. IEEE
Wirel. Commun. 2017, 24, 14–21. [CrossRef]
200. Riaz, S.; Khan, F.; Saleem, S.; Ahmed, Q.Z. Reducing the Mutual Outage Probability of Cooperative Non-Orthogonal Multiple
Access. IEEE Trans. Veh. Technol. 2020. [CrossRef]
201. Ma, Y.; Wang, Z.; Yang, H.; Yang, L. Artificial intelligence applications in the development of autonomous vehicles: A survey.
IEEE/CAA J. Autom. Sin. 2020, 7, 315–329. [CrossRef]
202. Burleigh, N.; King, J.; Bräunl, T. Deep Learning for Autonomous Driving. In Proceedings of the Digital Image Computing:
Techniques and Applications (DICTA), Perth, Australia, 2–4 December 2019; pp. 1–8.
203. Fayyad, J.; Jaradat, M.A.; Gruyer, D.; Najjaran, H. Deep Learning Sensor Fusion for Autonomous Vehicle Perception and
Localization: A Review. Sensors 2020, 20, 4220. [CrossRef]
Sensors 2021, 21, 706 33 of 33

204. Yao, Y.; Xu, X.; Zhu, C.; Chan, C.-Y. A hybrid fusion algorithm for GPS/INS integration during GPS outages. Measurement 2017,
103, 42–51. [CrossRef]
205. Nourmohammadi, H.; Keighobadi, J. Fuzzy adaptive integration scheme for low-cost SINS/GPS navigation system. Mech. Syst.
Signal Process. 2018, 99, 434–449. [CrossRef]
206. Jaradat, M.A.K.; Abdel-Hafez, M.F. Enhanced, delay dependent, intelligent fusion for INS/GPS navigation system. IEEE Sens. J.
2014, 14, 1545–1554. [CrossRef]
207. Noureldin, A.; Karamat, T.B.; Eberts, M.D.; El-Shafie, A. Performance enhancement of MEMS-based INS/GPS integration for
lowcost navigation applications. IEEE Trans. Veh. Technol. 2009, 58, 1077–1096. [CrossRef]
208. Haarnoja, T.; Ajay, A.; Levine, S.; Abbeel, P. Backprop KF: Learning discriminative deterministic state estimators. In Proceedings
of the Advanced Neural Information Processing Systems, Barcelona, Spain, 5–10 December 2016.
209. Brossard, M.; Barrau, A.; Bonnabel, S. AI-IMU Dead-Reckoning. IEEE Trans. Intell. Veh. 2020, 5, 585–595. [CrossRef]
210. Liu, K.; Ok, K.; Vega-Brown, W.; Roy, N. Deep inference for covariance estimation: Learning Gaussian noise models for state
estimation. In Proceedings of the IEEE International Conference of Robotic Automation (ICRA), Brisbane, Australia, 21–25 May
2018; pp. 1436–1443.
211. Zhao, C.; Pei, X.; Li, Z.; Wang, F.-Y. A comparative study of state-of-the-art driving strategies for autonomous vehicles. Accid.
Anal. Prev. 2020, 150, 105937. [CrossRef] [PubMed]
212. Bezai, N.; Medjdoub, B.; Al-Habaibeh, A.; Chalal, M.; Fadli, F. Future cities and autonomous vehicles: Analysis of the barriers to
full adoption. Energy Built Environ. 2021, 2, 65–81. [CrossRef]
213. Otayek, E.; Wehbe, R.; Massaad, Z. Autonomous vehicles implementation in Lebanon: Opportunities, challenges, and ethical
decisions. In Proceedings of the MATEC Web of Conferences, International Conference of Engineering Risk (INCER), Beirut,
Lebanon, 3–5 April 2019.
214. Zambom Santana, E.; Covas, G.; Duarte, F.; Santi, P.; Ratti, C.; Kon, F. Transitioning to a driverless city: Evaluating a hybrid
system for autonomous and non-autonomous vehicles. Simul. Model. Pract. Theory 2021, 107, 102210. [CrossRef]