Introduction

RADAR, short for Radio Detection and Ranging System, is an electromagnetic technology
used to detect the position and distance of objects in relation to the RADAR device. It
functions by emitting electromagnetic energy and analyzing the echoes or reflections from
objects. RADAR operates in the UHF and microwave frequency ranges and serves as a
sensor system for detecting, tracking, and identifying various objects at a distance. It can
detect a wide range of objects, including ships, aircraft, vehicles, and even precipitation or
wildlife. Besides position and velocity, RADAR can also provide information about the shape
and size of targets. It operates in adverse weather conditions and uses microwave frequencies
between 400 MHz and 40 GHz. RADAR was developed during the 1930s and 1940s for
military purposes and continues to be used in various sectors such as air traffic control,
weather monitoring, navigation, environmental monitoring, and more.

 Radar System Classifications:

Radar

Primary Secondary

Continuous Wave Pulse

Modulated Unmodulated MTI Doppler

Figure 1.1: Radar system classifications.

 Radar Types:
 Bistatic Radar
 Monopulse Radar
 Doppler Radar
  Instrumentation Radar
 Passive Radar
 Weather Radar
 Pulse Radar
 Navigation Radar
 Pulse Doppler Radar etc.

Table 1.1: RCS values of various objects

Target Cross Section Area, σ (m^2)
Insect 10^-4
Bird 0.01
Missile 0.5
Person 1
Fighter Jet 5-100
Air Liner 100-1000
Ship 300-10,00,000
B-2 Stealth Bomber 10^-6-10^-4

In Radar system, the target cross sectional area varies from different object to object. The
objects are tracked by radar system. The object may be insects, bird, missiles, person, fighter
plane, ship, rain drop etc. Some objects are obscure and some are not obscure. The radar
tracks the objects and provides information about the targets like size, shape, motion, speed,
color etc. For tracking purpose, many types of trackers are used.

 Radar Basic Block Diagram:

 Figure 1.1: Radar system basic block diagram.

The essential parts of this system include the following.

 Transmitter: It can be a power amplifier. It transmits power. Waveform generator
generates signal at first and amplified by amplifier.
 Pulse Modulator: The main function of modulator is to reserve and collect energy
for a defined period forming high power pulse.
 Duplexer: A duplexer can be used as both a transmitter and a receiver. It is a
bidirectional device that allows bi-directional communication over a single path.
 Receiver: It may be a superheterodyne receiver or any other receiver. It is used to
receive signal that transmitted from transmitter.
 Mixer: The combination of Low noise RF amplifier and local oscillator. It is fed to
IF amplifier.
 Local oscillator: The main function is mixing incoming signal to achieve
intermediate frequency through excitation of frequency.
 Display: The display is used for display purpose. It shows the output and
sometimes records the information for the future purposes
1.1 Thesis Background

Off late, the sound highway and the sound transportation are now a common issue all over
the world. Highway faces horrible accidents due to adverse weather like storm, rain, fog and
strong sunlight etc. Massive loss of time, life and property of common people. To disentangle
those vulnerable situations, this work may help a bit even more. A fast and highly accurate
tracking system can be utilized to guess the vehicle movement statistics from video data,
including the current speed of cars on the road. This information empowers officers to
oversee traffic conditions effectively. In recent years, researchers continue to enhance the
precision of object detectors with the assistance of tracking data [1].

Traffic surveillance systems that employ video capture and processing for speed assessment
have become crucial tools for accident prevention. Road accidents rank among the leading
causes of fatalities and injuries, and their frequency is on the rise [2].

Traffic accidents lead a significant danger, resulting in harm and fatalities to passengers,
pedestrians, and damage to vehicles and road infrastructure. Bangladesh, in particular, has a
high rate of road accidents, despite numerous efforts by authorities to reduce them [3].

The World Health Organization reports approximately 1 million annual traffic accident-
related fatalities worldwide, highlighting the pressing need for future planning to address this
issue. This includes areas like video surveillance and traffic monitoring [4].

A highly effective and innovative approach is introduced to track multiple objects. To

achieve this, two distinct methods are adopted for extracting object characteristics. In recent
years, owing to significant advancements in machine vision, computers become capable of
performing a variety of everyday tasks. One such task is the tracking of multiple vehicles,
which finds extensive application in various sectors [5].

Advanced technology provides various means of collecting traffic data, but different devices
often yield varying levels of accuracy in determining vehicle speed. The lack of uniform
accuracy standards across devices presents a common challenge for both transportation
researchers and practitioners [6].

Car accidents are a serious societal concern, with speeding being a leading cause of fatal
accidents. Studies indicate that a significant portion of serious accidents results from
exceeding speed limits or driving at inappropriate speeds. Vehicle speed plays a pivotal role
in accident severity, emphasizing the importance of setting appropriate speed limits for road
safety. This paper aims to propose a novel approach to controlling car speed limits to create
safer roads [7].
Modern road transport reduces the travel times for motorized vehicles, but the higher speeds
required for these improvements have adverse consequences, including increased energy
consumption, environmental impact, and a higher incidence of road accidents and related
injuries. Managing vehicle speed has become a global challenge for policymakers and road
safety professionals despite advancements in car capabilities and demands for faster roads
[8].

This research project intends to develop a vehicle speed measuring technique using a video-
based approach. Video-based vehicle speed detection systems offer cost-effective and
powerful solutions for intelligent traffic monitoring. These systems utilize camera outputs for
video processing to extract necessary information for speed detection. The measurement
provides better speed measurement. This study presents a new method for vehicle speed
detection through video analysis instead of traditional radar methods [9].

The ESP 32 Camera module monitors and records video footage in areas prone to speeding. It
follows the video processing methods. MATLAB software handles the speed calculation
process. This project aims to increase driver awareness, speed measurement, speed
measurement accuracy enhancement. These common purposes lead to ultimately reducing the
percentage of road accidents that may produce a sound highway [10].

1.2 Literature Review

A highly effective and innovative approach is introduced to track multiple objects. To

achieve this, two distinct methods are adopted for extracting object characteristics. In recent
years, owing to significant advancements in machine vision, computers become capable of
performing a variety of everyday tasks. One such task is the tracking of multiple vehicles,
which finds extensive application in various sectors [5].

v [13].

Accurately detecting and predicting the trajectory of moving objects using a mobile sensor is
essential. However, reliably tracking moving objects from a moving vehicle is challenging
due to apparent shape changes from shifting perspectives, which can be misinterpreted as
motion. To address this issue, stable linear features are extracted, and the uncertainty of
feature positions is characterized. The tracker is validated by assessing the consistency of
recent observations with the estimated motion, and it is designed to operate in real-time on
embedded processors with acceptable performance for collision warning purposes [14].
Traffic monitoring videos often contain noise from weather conditions or video compression.
To address these issues, the authors develop an algorithm capable of tracking and separating
highway vehicles in low-quality monitoring videos where conventional methods based on
shape or edge models are not easily applicable. The segmentation method initially identifies
motion vectors associated with occluded vehicles and then assigns over-segmented image
patches to the group of motion vectors that best represent the movement of the patch. The
proposed algorithm produces excellent results for high-traffic situations and works well with
highly compressed and low-resolution videos [15].

The approach considers both dynamic and geometric properties of tracked vehicles and
introduces the concept of motion evidence to overcome the low signal-to-noise ratio in noisy
urban environments. It is experimentally validated under challenging conditions, including
those encountered during the Urban grand challenges and other urban settings. However, the
approach currently does not model pedestrians, bicyclists, or motorcyclists, which is
necessary for driving in populated areas. Future research in perception for autonomous urban
driving can focus on modeling these additional types of moving objects and exploring sensor
fusion with laser, radar, and vision sensors [16].

The implementing a Probability Hypothesis Density (PHD) filter is the primary focus for
tracking multiple extended targets. The paper introduces a modified version of the PHD filter
to handle extended targets and proposes a method to partition measurements into subsets
containing measurements from the same source. The paper also discusses the Simultaneous
Localization and Mapping (SLAM) problem and the Multiple Target Tracking (MTT)
problem in relation to extended target tracking [19].

An efficient and innovative approach is introduced for detecting moving vehicles and
estimating their speeds using a single camera in well-lit environments. The approach detects
and tracks vehicles passing through the surveillance area and records their positions.
Tracking is based on the relative positions of vehicles in consecutive frames, and the
information can be used in the Automatic Number Plate Recognition (ANPR) system to
detect speed limit violations. The proposed approach achieves an average detection accuracy
of approximately 87.7% and a maximum tracking accuracy of up to 98.3% in the afternoon
session. The approach utilizes parameters such as position, height, and width of vehicles for
detection and tracking, requiring less computation and memory. The system can be easily
integrated into existing traffic management systems [20].
The article provides a comprehensive overview of current research in extended object
tracking, defining it, distinguishing it from other object tracking types, and addressing
modeling aspects. The authors offer a tutorial introduction to two commonly used approaches
for extended object tracking: the random matrix approach and the Kalman filter-based
approach for star-convex shapes. They also discuss techniques for tracking multiple extended
objects using both Random Finite Set (RFS) and non-RFS multi-object trackers [21].

Introducing an efficient and accurate approach is focused for multiple-vehicle tracking, a

critical task in video surveillance and traffic monitoring. The proposed approach uses an
appearance and motion model based on features extracted from each object using two
different methods: a deep neural network and traditional features. Results indicate that the
proposed methods achieve acceptable accuracy and can be further improved by extracting
more distinctive features [24].

A new tracking algorithm for multiple vehicles is introduced for advanced driver assistance
systems, a crucial component of high-level tasks. The algorithm uses a monocular camera in
the driver's view and follows the tracking-by-detection approach, integrating detection and
appearance descriptors into a single network. The algorithm addresses challenges related to
object illumination variations and deformations while maintaining real-time performance. A
two-step matching strategy and a Kalman filter are employed for data association.
Experimental results demonstrate the algorithm's superiority over state-of-the-art methods
and its ability to maintain a frame rate of 16 FPS on the test dataset while effectively solving
tracking problems in driving scenarios [25].

Multiple radar systems offer an notably solution for target tracking by extending coverage
and improving trajectory detection probability and localization accuracy. However, the
presence of multiple extended or weak targets in cluttered environments poses challenges.
The article introduces an algorithm for detecting and tracking multiple targets, both extended
and weak, using measurements from multiple radars. The method is designed to handle dense
clutter and complex target scenarios, including extended targets, weak targets and clutter
[27].

1.3 Contribution
The dedication of this thesis is as follows:
  To implement Extended object tracking model scenario for detecting highway vehicle.
 To implement multipath radar reflection method scenario for tracking passing vehicle.
 To enhance radar range, false tracking reduction, resolution enhancement etc.
 To propose Simulink model for Extended object tracking and multipath radar
reflection.
 To analyze the performance improvement using Simulink model.

Conclusion and Future Work

7.1 Conclusion

The performance of extended object tracking with multipath radar reflection model to track
highway vehicles has been evaluated. In extended object tracking and multipath radar
reflection method; the detection number has been enhanced. Hence the radar systems false
alarm has been removed, clutter has been removed, tracking range has been enhanced, error
has been reduced. For measuring the vehicles speed, the video capturing from ESP 32 camera
module has been suggested. The accuracy is found 94% that is helpful for reduce horrible
accidents. Rest 6% is not found due to dust in weather. But it is achievable and respective
research is still going on. The existing models have been implemented in Simulink platform
to enhance performance. To enhance the existing model performance that has become the
main concern. This concern is disentangled finally by improving the model performance
using Simulink platform.

7.2 Future Work

 Various trackers may be used to track the vehicles and reduce OSPA metric.
 To observe the various tracker effects on performance improvement.
 Detection number can be increased more.
  Scanning range of the radar can be upgraded from 3850m.
 Vehicle tracking and speed measurement can be done by using Machine
learning.
 Accuracy upgradation more than 94%.

References

[1] M. Rath, "Smart traffic management system for traffic control using automated
mechanical and electronic devices," in International conference on mechanical,
materials and renewable energy, India, 2018.

[2] S. Gupte, O. Masoud, R. Martin and N. Papanikolopoulos, "Detection and classification

of vehicles," in IEEE transactions on intelligent transportation systems, Minneapolis,
2002.

[3] M. G. Moazzam, M. R. Haque and M. S. Uddin, "Image-based vehicle speed

estimation," Journal of computer and communications, vol. 7, no. 1-5, p. 1, 2019.

[4] C. M. Safwan and N. C. Manan, "Vehicle speed estimation system by using matlab,"
Evolution in electrical and electronic engineering, vol. 2, no. 2, p. 704, 2021.

[5] F. Bafghi and B. Shoushtarian, "Multiple-vehicle tracking in the highway using

appearance model and visual object tracking," in The 11th iranian and the first
international conference on machine vision and image processing, Isfahan, Iran, 2020.

[6] M. A. Adnan, N. I. Zainuddin, N. Sulaiman and T. B. H. T. Besar, "Vehicle speed

measurement technique using various speed detection instrumentation," in IEEE
Business Engineering and Industrial Applications Colloquium (BEIAC), Malaysia,
2013.

[7] M. Abdelsalam and T. Bonny, "IoV road safety: vehicle speed limiting system," in
Talal Bonny, Sarjah, UAE, 2019.

[8] P. Jiwattanakulpaisarn, K. Kanitpong and P. Suriyawongpaisal, "Effectiveness of speed

enforcement in Thailand: Current issues,need for changes and new approaches,"
Transport and communications bulletin for Asia and the Pacific, vol. 1, no. 79, p. 41,
 2009.

[9] U. Gupta, U. Kumar, S. Kumar, M. Shariq and R. Kumar, "Vehicle speed detection
system in highway," International research journal of modernization in engineering
technology and science, vol. 4, no. 5, p. 406, 2022.

[10] C. M. S. N. C. Manan and W. M. H. W. Mahmud, Vehicle speed estimation system by

using Matlab , vol. 2, no. 2, p. 704, 2021.

[11] G. D. Sullivan, K. D. Bake, A. D. Worrall, C. I. Attwood and P. R. Remagnino, "Model-

based vehicle detection and classification using orthographic approximations," in Image
and vision computing, United Kingdom, 1996.

[12] S. Gupte, O. Masoud, R. Martin and N. Papanikolopoulos, "Detection and classification

of vehicles," in IEEE transactions on intelligent tansportation systems, Minneapolis,
2002.

[13] R. Mobus and U. Kolbe, "Multi-target multi object tracking, sensor fusion of radar and
infrared," in IEEE intelligemt vehicles sympoelum, Parma, Italy, 2004.

[14] R. A. MacLachlan and C. Mertz, "Tracking of moving objects from a moving vehicle
using a scanning laser range finder," in IEEE intelligent transportation systems
conference, Canada, 2006.

[15] G. Jun, J. K. Aggarwal and M. G¨okmen, "Tracking and segmentation of highway

vehicles in cluttered and crowded scenes," in IEEE Workshop on Applications of
Computer Vision Computing, (WACV), Copper, Colorado, 2008.

[16] G. D. Sullivan, K. D. Bake, A. D. Worrall, C. I. Attwood and P. R. Remagnino, "Model-

based vehicle detection and classification using orthographic approximation," in Image
and vision computing, United Kingdom, 1996.

[17] A. Petrovskaya and S. Thrun, "Model based vehicle detection and tracking for
autonomous urban driving," in Autonomous robots 26, California, 2009.

[18] U. Habiba, Divyaprabha and M. Z. Kurian, "Multi object tracking using kalman filter,"
International journal of science academic research, vol. 2, no. 3, pp. 1217-1219, 2021.

[19] K. Granström, C. Lundquis and U. Orguner, "Tracking rectangular and elliptical

extended targets using laser measurements," in 14th international conference on
information fusion, Chicago, USA, 2011.

[20] K. Granström, C. Lundquist and U. Orguner, "Extended target tracking using a Gaussian
Mixture PHD filter," IEEE transactions on aerospace and electronic systems, vol. 48,
no. 4, pp. 3268-3286, 2012.

[21] T. Kumar and D. S. Kushwaha, "An efficient approach for detection and speed
estimation of moving vehicles," in Twelfth international multi-conference on
information processing-2016 (IMCIP-2016), India, 2016.

[22] K. Granstrom, M. Baum and S. Reuter, "Extended object tracking: introduction,

 overview and applications," Journal of advances in information fusion, vol. 12, no. 2, p.
139, 2017.

[23] P. Prasad and A. Gupta, "Moving object tracking and detection based on kalman filter
and saliency mapping," Advances in intelligent systems and computing, vol. 542, pp.
639-646, 2018.

[24] R. Prophet, J. Martinez, J. F. Michel, R. Ebelt, I. Weber and M. Vossiek, "Instantaneous

ghost detection identification in automotive scenarios," in IEEE radar conferemce
(RadarConf), Boston, USA, 2019.

[25] F. Bafghi and B. Shoushtarian, "Multiple-vehicle tracking in the highway using

appearance model and visual object tracking," in International Conference on Machine
Vision and Image Processing (MVIP), Iran, 2020.

[26] P. Lyu, M. Wei and Y. Wu, "Multi-vehicle tracking based on monocular camera in
driving View," in MDPI, Nanjing, China, 2022.

[27] BoYan, E. Paolini, LupingXu and H. Lu, "A target detection and tracking method for
multiple radar systems," in IEEE transactions on geoscience and remote sensing, China,
2022.

[28] A. Srivastav and S. Mandal, "Radars for autonomous driving: A review of deep learning
methods and challenges," in Digital object identifier, Upton, New York, 2023.

[29] MathWorks, "Extended object tracking of highway vehicles with radar and camera,"
[Online]. Available: https://www.mathworks.com/help/driving/ug/extended-object-
tracking.html. [Accessed 15 July 2022].

[30] Mathworks, "Highway vehicle tracking with multipath radar reflections," [Online].
Available: https://www.mathworks.com/help/driving/ug/multipath-radar-detection-and-
tracking.html. [Accessed 22 Agust 2022].

[31] MathWorks, "Introduction to tracking scenario and simulating sensor detections,"

[Online]. Available: https://www.mathworks.com/help/fusion/ug/introduction-to-
tracking-scenario-and-simulating-sensor-detections.html. [Accessed 5 September 2022].

[32] A. R. Gadah, "Vehicle speed measurement using image processing," in The graduate
school of applied sciences of the near east university, Cyprus, 2018.

[33] C. Manan and W. Mahmud, "Evolution in Electrical and Electronic Engineering Vol. 2
No. 2 (2021) 703-708," Vehicle speed estimation system by using MATLAB , vol. 2, no.
2, pp. 706-707, 2021.

[34] A. Gadah, "Vehicle speed measurement using image processing," Cyprus, The graduate
school of applied sciences of the near east university, 2018, pp. 59-60.