Citation
Abdul Aziz, Noor Hafizah
(2018)
Passive forward scatter radar based on LTE signal for vehicle detection and classification.
Doctoral thesis, Universiti Putra Malaysia.
Abstract
Passive bistatic radar (PBR) system can utilise suitable signal from illuminator of
opportunity to improve radar proficiency. In conventional PBR, a heterodyne
receiver and concept is employed. This is due to the PBR requiring reference signal
from a direct transmission for synchronization, and then the signal is down converted
to the base band where target detection is evaluated by analyzing the ambiguity
function of the received signal. On the other hand, by using the forward scattering
phenomena and technique in specific mode of PBR, the operation can be further
enhanced and improved especially in target detection and classification. This specific
system is identified as the passive forward scattering radar (FSR). One of the main
condition of passive FSR is the operational mode limited to within the radar
baseline. In this scenario, the desired received signal is formed through the
shadowing of a direct signal by the target shape rather than back scattering signal
from the target such as in conventional radar system. Passive forward scattering
radar system offers a number of advantages including enhanced target cross section,
long coherent intervals of the receiving signal, absence of signal fluctuations,
reasonably simple hardware, economical as it does not need a transmitter system and
a spectrum allocation, practically unseen to surveillance receivers, portable due to its
smaller size, no synchronization needed for reference signal from direct signal of
illuminator, straightforward signal pre-processing for target detection and enhanced
classification capability.
Thus, the aim of this thesis is to develop and prove the concept of passive forward
scattering radar especially for ground detection and classification. The objective is to
implement the experimental analysis and results for vehicle detection and
classification by exploiting the latest 4G LTE technology signal. This thesis clarifies
in details the LTE based passive forward scatter radar receiver circuit, the detection
scheme and the classification algorithm. In addition, the proposed passive forward scatter radar circuit employed the self-mixing technique at the receiver, hence it did
not require a synchronization signal from the LTE base station. The classification
capability in passive forward scattering radar increases the effectiveness of the
system.
Three vehicle categories with different sizes and shapes were tested, namely
Compact, Saloon and Small Sport Utility Vehicle (SUV). The passive FSR
experiment results show that the vehicles were successfully detected, even by raw
received signals without any complicated signal processing techniques and the
proposed classification system provided outcomes of satisfactory classification
performance. The Doppler spectrum scattered by the vehicle is used as the features
for input to the classification system. The classification algorithm was developed
based on Principal Component Analysis (PCA) and k-Nearest Neighbours (k-NN).
In addition, the baseline crossing range distance from the radar receiver, as well as
vehicle’s speed, was successfully predicted.
In general this thesis presents the first classification analysis and performance for the
passive forward scatter radar system. The great potential of the passive forward
scatter radar system provides a new research area in passive radar that can be used
for diverse applications. The thesis also proved that if the FSR mode can be
integrated or hybrid with the conventional passive radar system and still have similar
performance competency, the system can achieve tremendous improvement. Hence,
it opens up a new frontier in passive radar that can be used for many applications
including border protection, microwave fences, building monitoring, traffic
surveillance, and so forth.
Download File
Additional Metadata
Actions (login required)
![View Item View Item](/style/images/action_view.png) |
View Item |