Advanced
LTE-Based Passive Bistatic Radar System for Detection of Ground-Moving Targets
LTE-Based Passive Bistatic Radar System for Detection of Ground-Moving Targets
ETRI Journal. 2016. Apr, 38(2): 302-313
Copyright © 2016, Electronics and Telecommunications Research Institute (ETRI)
  • Received : March 11, 2015
  • Accepted : December 07, 2015
  • Published : April 01, 2016
Download
PDF
e-PUB
PubReader
PPT
Export by style
Share
Article
Author
Metrics
Cited by
TagCloud
About the Authors
Raja Syamsul Azmir Raja Abdullah
Asem Ahmad Salah
Alyani Ismail
Fazirulhisyam Hashim
Nur Emileen Abdul Rashid
Noor Hafizah Abdul Aziz

Abstract
Use of a passive bistatic radar (PBR) system in the surveillance or monitoring of an area has its advantages. For example, a PBR system is able to utilize any available signal of opportunity (for example, broadcasting, communication, or radio navigation signals) for the purposes of surveillance. With this in mind, there are potentially many research areas to be explored; in particular, the capability of signals from existing and future communication systems, such as 4G and 5G. Long-Term Evolution (LTE) is the world’s most current communication system. Given this fact, this paper presents the latest feasibility studies and experimental results from using LTE signals in PBR applications. Details are provided about aspects such as signal characteristics, experimental configurations, and SNR studies. Six experimental scenarios are carried out to investigate the detection performance of our proposed system on ground-moving targets. The ability to detect is demonstrated through use of the cross-ambiguity function. The detection results suggest that LTE signals are suitable as a source signal for PBR.
Keywords
I. Introduction
A passive radar system comprises a receiver without a co-located transmitter. It uses non-cooperated illumination sources for target detection. Thus, it has numerous advantages over a conventional radar system: (a) it is practically invisible to surveillance receivers using conventional radio direction finding techniques, (b) it is easily transported due to its smaller size, (c) it is cheaper, as it does not send out a signal, and (d) it requires no spectrum allocation. Consequently, it has no environmental impact.
Recently, the use of illuminators of opportunity by passive radar systems has gained the interest of radar engineers and researchers. As a result, illuminators of opportunity are employed in many areas, such as Global System for Mobile Communications [1] , [2] and Worldwide Interoperability for Microwave Access [3] , [4] for medium range applications. Long-Term Evolution (LTE) is one of the latest wireless communication technologies to provide last-mile broadband wireless access with anticipated widespread accessibility. A typical LTE signal has certain characteristics that make it attractive for use in applications related to passive radar. The parameters and characteristics of interest (of a typical LTE signal) are as follows:
  • ▪ A typical LTE signal has a broad bandwidth, within the range of 1.4 MHz to 20 MHz; thus, it can have a high resolution range.
  • ▪ The use of a special downlink frame structure for LTE with orthogonal frequency-division multiple access guarantees the presence of low side-lobes with regard to the ambiguity function.
  • ▪ The LTE standard covers a range of many different frequency bands ranging from 800 MHz to 3500 MHz, and an LTE signal has the ability to support both frequency division duplex and time division duplex[5],[6], which means that the opportunity for LTE to be deployed worldwide is enhanced.
  • ▪ The number of commercial LTE networks is increasing year on year. Therefore, increasing the availability of LTE signals serves to enhance opportunities to deploy LTE-based passive radar systems.
The aforementioned LTE signal characteristics motivated us to study the feasibility of LTE signals as the new illuminator of opportunity for passive-radar applications for the first time in [7] , where we fully analysed the LTE signal based on range, Doppler ambiguities, and resolutions. The results showed that LTE signals can be used in passive-radar applications, since such signals are able to achieve a good range and Doppler resolution (8.6 m and 0.132 m/s, respectively). These preliminary results motivated us to conduct a further investigation [8] into the feasibility of LTE-based passive radars for detecting a ground-moving target, where a theoretical analysis was conducted on a captured LTE signal in the atmosphere, received from a real LTE Evolved Node B (eNB). This was then followed by a field experiment [9] . Despite the positive results shown, there is still the need for more experimental studies to investigate the extent to which an LTE-based passive radar system is capable of detecting a ground-moving target. Therefore, this paper intends to investigate the extent to which an LTE-based passive radar system is capable of detecting the following:
  • ▪ Ground-moving targets with different speeds
  • ▪ Ground-moving targets with changeable trajectories
  • ▪ Multiple ground-moving targets in the same scene.
The results complement the gaps within the previously mentioned studies on passive radar systems and can be useful for more advanced, practical passive radar systems with high resolution requirements.
Passive bistatic radar (PBR) could be utilized but is not limited to the following potential applications:
  • ▪ Border protection: since passive radar does not involve the transmission of any kind of signals, it will be invisible to surveillance.
  • ▪ Detection of targets flying at low altitudes: PBR can be used since most of the illuminators of opportunity used in PBR emit their signals into the air.
  • ▪ Monitoring of intruders: passive radar can be used to detect the movements of a human.
  • ▪ Traffic surveillance: passive radar can be used to detect the movements of vehicles.
Figure 1 illustrates some of the potential applications of LTE-based PBR.
PPT Slide
Lager Image
Potential applications of LTE-based passive radar.
This paper is organised as follows. Section II analyses the characteristics of LTE waveforms for radar applications in terms of range and Doppler resolution. The SNR calculation for the propagation of an LTE signal is presented in Section III. The architecture for the proposed LTE-based passive radar system is illustrated and explained in Section IV. Experimental results and discussions for the six conducted scenarios are presented in Section V. Section VI concludes this paper.
II. Resolutions in LTE PBR
- 1. Range Resolution
In bistatic radar applications, the minimum required range separation between two targets is known as range resolution, Δ R , where the two targets are assumed to be collinear with the bistatic bisector. Range resolution is defined as follows [10] , [11] :
(1) ΔR=c/2B cos(β/2),
where B and c are the signal bandwidth and speed of light, respectively; β is the bistatic angle, which is defined as the angle between the transmitter and the receiver, with the vertex at the target [10] . From (1), one can see that range resolution is a function of the signal bandwidth. The larger the bandwidth of the waveform used in the radar, the greater the range resolution. The bandwidth of an LTE signal may range from 1.4 MHz to 20 MHz; thus, a range resolution of 8.6 m is achievable by using an LTE signal of bandwidth 20 MHz and a bistatic angle of 60º. Similarly, a range resolution of 17.3 m is achievable by using an LTE signal of bandwidth 10 MHz and a bistatic angle of 60º. An LTE signal has a better range resolution compared to other illuminators of opportunity; thus, this fact helps to efficiently identify two targets that are within close proximity to each other.
- 2. Doppler Resolution
Doppler resolution determines how well a radar can observe moving targets of different radial velocities. It can be determined from the receiver’s coherent integration time (CIT), where the adequate degree of Doppler separation between the echoes of two moving targets (T1 and T2) at the receiver is given by [11]
(2) Δ f d = f d T1 − f d T2 =1/T,
where Δ f d is the Doppler resolution and T is the CIT; f dT1 and f dT2 are the received Doppler echoes from the first and second targets, respectively, and they are defined as [11]
(3) f d T1 = 2 v T1  cos( α 1 )cos(β/2)/λ,  
(4) f d T2 = 2 v T2  cos( α 2 )cos(β/2)/λ.
The geometry for the velocities of targets 1 and 2 ( v T1 and v T2 , respectively) is shown in Fig. 2 , where α 1 and α 2 are the velocity radial angle for target 1 and target 2, respectively. The two targets are assumed to be co-located so that they share the same bistatic bisector. By combining equations (2), (3), and (4), we can derive the following equation:
(5) Δ f d =2 cos(β/2)( v T1  cos( α 1 )− v T2  cos( α 2 ) )/λ.
PPT Slide
Lager Image
Geometry for bistatic Doppler resolution.
The required difference between the velocity vectors pertaining to the two targets projected onto the bistatic bisector is known as the velocity resolution, Δ v , and it is given as [11]
(6) Δv= v T1  cos( α 1 )− v T2  cos( α 2 ).
Then, by combining equations (5) and (6), the velocity resolution becomes
(7) Δv=λ/2T cos(β/2).
As an example, in this paper, the adopted CIT is 0.2 s and the Δ f d is calculated to be 5 Hz, corresponding to the velocity resolution of 0.288 m/s if β = 60º and f c = 2.635 GHz are used. Therefore, the LTE-based passive radar system can distinguish two ground targets moving with a velocity difference of 0.332 m/s (1.18 km/h). Figure 3 shows the velocity resolutions of different CIT values for various LTE carrier frequencies, and it is clearly shown that the velocity resolution improves as the carrier frequency increases. Considering this case study, where the LTE carrier frequency is 2.635 GHz, by increasing the CIT to 0.5 s, the LTE signal can achieve a velocity resolution of 0.1322 m/s, which is considered as a good Doppler resolution — one that makes an LTE-based passive radar suitable for ground-moving-target detection applications.
PPT Slide
Lager Image
Velocity resolutions of different CIT values for various LTE carrier frequencies.
III. SNR Calculation for LTE Signal Propagation
Target radar cross-section (RCS) illuminated from an LTE signal is first analyzed to give more understanding in calculating the SNR. Thus, a real target was modeled using Autodesk and then imported to CST software for RCS calculation. Two incident angles, ϕ , were analysed to understand the position that gives a maximum cross-sectional profile (see Fig. 4 ). The RCS results indicate that the value of RCS at ϕ = 90° is higher than that at ϕ = 59°, which increases the SNR. Moreover, the side-lobe level is smoother, which shows that there is more fluctuation due to the high visibility for ϕ = 90°.
PPT Slide
Lager Image
(a) Vehicle model, (b) signal source, and (c) RCS.
To estimate the SNR for the LTE-based passive radar system, the reflected signal from an object with specific RCS at a specified range using a point target is considered (see Fig. 5 ). The equation for the power at the input to the receiver is [12]
(8) P r = P t G t G r λ 2 σ (4π) 3 R t 2 R r 2 L ,
where P t and P r are the transmitted and received power, respectively; G t and G r are the antenna gains for the transmitter and receiver, respectively; λ is the signal wavelength; σ is a target’s non-fluctuating RCS (measured in square meters); L is a general loss factor that accounts for both system and propagation losses; R t is the range from the transmitter to the target; and R r is the range from the receiver to the target. The noise is modeled by assuming the thermal noise at the receiver has white power spectral density; thus, the total noise power at the output of the receiver is given by [12]
(9) N=kT F n /τ,
where k is the Boltzmann constant, T is the effective noise temperature, and F n is the receiver noise factor. We assume that the squared magnitude of the receiver frequency response approximates a rectangular filter with bandwidth equal to the reciprocal of the pulse duration, 1/ τ . The product of the effective noise temperature and the receiver noise factor, which is the system temperature, is denoted by T s . We have T s = TF n . Therefore, the receiver output SNR is
(10) P r N = P t τ G t G r λ 2 σ (4π) 3 k T s R t 2 R r 2 L .
PPT Slide
Lager Image
LTE transmission with ground-moving target reflection diagram.
The estimated SNR for the signal reflected from different targets in the LTE system is shown in Fig. 6 . The SNR is calculated for three different ground-moving targets by using the following typical relative RCS values: (i) “Car” has an RCS of approximately. 2.2 m 2 , (ii) “Motorbike” has an RCS of approximately. 0.5 m 2 , and (iii) “Human body” has an RCS of approximately. 0.2 m 2 . From Fig. 6 , it can be seen that the SNR is at its lowest for “Human body” and that it is at its highest for “Car,” corresponding to the larger reflective area of the car itself. It is also noted that the SNR is reduced as the range between the targets and receiver increases.
PPT Slide
Lager Image
SNR estimation for LTE signal reflected from different ground-moving targets.
IV. LTE-Based Passive Radar System Architecture
The proposed system consists of two parallel co-located channels, Ch 1 and Ch 2, as illustrated in the receiver hardware system architecture in Fig. 7 . Ch1 is dedicated to receive the direct-path reference signal from the LTE eNB, while Ch2 is dedicated to receive the echo signal reflected from the ground-moving targets. The two channels have the same structure, where each channel is started by a horn antenna used to receive the LTE signal carried at the 2.635 GHz carrier frequency. The antenna of Ch1 is directed toward the LTE eNB (LTE base station), while the antenna of Ch2 faces the area where the ground-moving target should be. The antenna is followed by a low-noise amplifier (LNA); the LNA is used to amplify the received RF signal. Then, the desired frequency band of the LTE downlink signal is selected through a band-pass filter. Subsequently, the desired signal is down-converted to the baseband by heterodyning it with a local oscillator signal using a frequency mixer. An amplifier is used to amplify the baseband signals to provide sufficient gains for the LTE signal before filtering out any undesired frequencies using a low-pass filter. Then, the LTE baseband signals are saved into PC hard drives at a sampling rate of 25 MS/s. An exemplar channel of the implemented experimental LTE-based passive radar system for detection is shown in Fig. 8 .
PPT Slide
Lager Image
Architecture of experimental LTE-based passive bistatic radar system for ground-moving target detection.
PPT Slide
Lager Image
Hardware implemented for LTE-based passive radar.
Both of the received signals from the direct-path channel Ch1 and reflected-path channel Ch2 are saved in long data sets; thus, huge processers are required to handle such large quantities of data; consequently, it may take a considerable amount of time to process such quantities and therefore data formatting is necessary for both channels before continuing with cross-ambiguity coherent processing. The overall signal processing scheme associated with the LTE-based passive radar is illustrated in Fig. 9 .
PPT Slide
Lager Image
LTE-based passive radar signal processing scheme for ground-moving target detection.
The received LTE signal data are formed in segments; one segment is to be processed at a time. The length of each segment will decide the CIT, and subsequently the Doppler resolution. In this paper, a segment of length 5 MS is adopted, corresponding to a CIT of 0.2 s for a 25 MS/s sampling rate. Therefore, the obtained Doppler resolution is 5 Hz, corresponding to 0.3 m/s (1.1 km/h) velocity resolution with f c = 2.635 GHz and β = 60°.
The two LTE signals received from Ch1 and Ch2 are unequal by two parameters — time delay and Doppler shift. In fact, these two parameters will decide the range and velocity of the detected ground-moving target. Therefore, the cross ambiguity function (CAF) is applied [13] , [14] , which is the matched filter response to the joint time-delay and Doppler-shift versions of the LTE signal it is matched to. It is given by
(11) A(τ,  f d )=∫ s r (t)⋅ s d * (t−τ) e −2π f d t  dt,
where s r ( t ) and s d (t) are the received target echo signal and direct reference signal, respectively. Time-delay τ and Doppler-shift f d are the two parameters to be searched for the values that cause A ( τ , f d ) to peak. This can be achieved by delaying the direct signal, s d ( t ), in time τ and shifting its frequency by some amount, f d , then cross-correlating it with the reflected signal, s r ( t ), followed by searching for the maximum value of A ( τ , f d ) that gives the peak. After obtaining all the A ( τ , f d ) values, the Doppler-range plane is plotted in contours.
In this paper, all the detection results are illustrated in contour plots, which show the isolines of the CAF output matrix A ( τ , f d ). After normalizing the CAF matrix, the contour plots are drawn with contour lines at a specified contour cut-off level. The contour cut-off level is determined to be (the strongest peak detected for f d ≠ 0) − (3 dB). Here, we choose 3 dB because this value allows us to see a peak clearly. This means that the contour lines will be plotted for all levels from the cut-off value to zero.
An illustrative example of complete contour plots is shown in Fig. 10 for normalized CAF under a two-target detection scenario. Two peaks can be clearly seen in Fig. 10(a) , Peak 1 and Peak 2, corresponding to two detected targets. The first target is detected at a range of approximately. 40 m and has a Doppler frequency of −120 Hz (Peak 1). At the same time, a second target is detected at a range of approximately. 20 m having a Doppler frequency of −90 Hz (Peak 2). A contour plot of the CAF is shown in Fig. 10(b) , where the two peaks are more readable.
PPT Slide
Lager Image
Example of CAF plots for two-target detection scenario: (a) mesh plot and (b) contour plot.
V. Experimental Results and Discussion
Field experiments were conducted by utilizing a signal transmitted from a dedicated operational LTE eNB. The experiment site is located in an open car park situated 400 m from the LTE eNB transmitter (see Fig. 11 ). The experiment site is used to illustrate each scenario appearing in the following subsections. Six scenarios were carried out on different types of ground-moving targets, including cars, people, and motorbikes, with the aim of examining the proposed LTE-based PBR system’s capability of detecting diverse types of ground-moving targets with different speeds and different trajectories. The conducted scenarios are summarized in Table 1 .
PPT Slide
Lager Image
Aerial photo of geometrical configurations for experiment site for detection.
Summary of conducted scenarios.
Aim Scenario Scenario description
Different speeds detection A Car move from the receiver to 140 m at different speeds.
Different trajectories detection B One car move away from the receiver and make a turn back after 90 m.
Different targets detection C Motor bike move away from the receiver to 130 m.
Multi-targets detection D Two people runsat different speeds move away from the receiver up to 65 m.
E Two cars follow each other in a straight line from the receiver to a 160 m distance away.
F Two cars and one motor bike move away from the receiver in straight lines with different speeds.
- 1. Scenario A
Figure 12 depicts the geometrical configuration and ground truth for scenario A, whereby a car is to move in a straight line starting from the receiver’s Ch2 antenna until it reaches a point at a distance of 130 m away. The scenario is carried out for four different car speeds: v = 10 km/h, 20 km/h, 30 km/h, and 40 km/h. The channel antenna of the direct path (marked as “Ch1 Antenna”) is shown to be directed toward the LTE eNB transmitter, while the channel antenna of the reflected path (marked as “Ch2 Antenna”) is shown to be directed toward the target area to receive the target echo signals.
PPT Slide
Lager Image
Geometry of scenario A: car moving away in straight line from receiver.
The detection processing results are displayed in the Doppler range contours in Fig. 11 . The speed of the car and its corresponding Doppler frequency can also be calculated from
(12) v= f d c/2 f c  cos(α)cos(β/2),
where c is the speed of light and f c is the LTE signal carrier frequency. The relative velocity angle and bistatic angle are denoted by α and β , respectively.
From Fig. 13(a) , the detected Doppler frequencies for the range 10 m to 130 m fluctuate between −45 Hz and −55 Hz, which corresponds to a velocity of approximately. 10 km/h; the negative sign indicates that the car is moving away from the receiver. The detected Doppler frequencies for the range 20 m to 120 m in Fig. 13(b) fluctuate between −80 Hz and −100 Hz, which corresponds to a velocity of approximately. 20 km/h. In Fig. 13(c) , the detected Doppler frequencies for the range 40 m to 180 m fluctuate between −115 Hz and −130 Hz, which corresponds to a velocity of approximately 30 km/h. Finally, the detected Doppler frequencies for the range 45 m to 160 m in Fig 13(d) fluctuate between −155 Hz to −165 Hz, which corresponds to a velocity of approximately 40 km/h. It is noted from the whole of Fig. 13 that the strength of peaks reduces as the range of a target increases, which is due to the fact that fewer reflected signals will be received from such targets. From these results, it can be concluded that the LTE-based passive radar can detect ground-moving targets with different speeds.
PPT Slide
Lager Image
Doppler-range detection contour for car moving at speed of (a) v = 10 km/h, (b) v = 20 km/h, (c) v =30 km/h, and (d) v = 40 km/h.
- 2. Scenario B
In scenario B, a car is to move in straight line from the bistatic receiver’s Ch2 antenna and then make a U-turn at around 90 m away from the receiver. The geometric configuration for scenario B and Doppler-range detection results of the entire cell ranges are illustrated in Fig. 14 . It can be seen that the car travelled at an average velocity of 10 km/h and made the required U-turn at a range of 90 m.
PPT Slide
Lager Image
Geometrical configurations for scenario B and detection results.
- 3. Scenario C
In scenario C, a motorbike is used instead of a car. The motorbike moves in a straight line starting from the receiver’s Ch2 antenna until it reaches a point at a distance of 130 m away, with increasing speed up to 30 km/h. The geometric configuration for scenario C and the Doppler-range detection results of the entire cell ranges are shown in Fig. 15 . From the figure, it is shown that the motorbike has travelled for approximately. 130 m at a non-uniform speed ranging from 22 km/h to 32 km/h (−95 Hz to −135 Hz). As the motorbike is smaller in size than a car, it has a smaller RCS. However, the LTE-based passive radar shows the capability to detect it at a range of approximately 130 m.
PPT Slide
Lager Image
Geometrical configurations for scenario C and detection results.
- 4. Scenario D
The aim of scenario D is to examine the system’s capability to detect two ground-moving targets in the same scene. Two persons (Person 1 and Person 2) started running from the Ch2 antenna until they reached a point 50 m away. The geometrical configuration for the experiment site of scenario D and the detection processing results for the 4th and 9th range cells are illustrated in Figs. 16(a) and 16(b) , respectively. To examine the LTE-signal range resolution (17.3 m), which was calculated in Section II, the two persons are separated at a distance of 20 m. They then run at different speeds to examine the calculated Doppler (velocity) resolution (1.18 km/h or 5 Hz). It is shown from Fig. 16(a) that both persons are almost at the same range of the receiver (12 m), but the speed of Person 1 is greater than that of Person 2 (approximately 15 km/h (−65 Hz) for Person 1 and approximately. 13 km/h (−55 Hz) for Person 2). The estimated positions for them are illustrated in Fig. 16(b) . Person 1 exceeded Person 2. Person 1 was detected at a range of 48 m with a speed of 15 km/h, while Person 2 was detected at a range of 36 m with a speed of 12 km/h.
PPT Slide
Lager Image
Scenario D: detection results in Doppler-range contour for (a) 4th range cell and (b) 9th range cell.
From the results, it can be deduced that the LTE-based passive radar system can differentiate two moving humans separated by a distance 20 m, which is a bit higher than the range resolution (17.3). The two persons had also moved with different speeds; the speed difference of the two persons is 2 km/h, which is slightly higher than the LTE velocity resolution (1.3 km/h). Therefore, the system is suitable to be used for monitoring intruders.
- 5. Scenario E
In scenario E, two cars drove in straight line from the Ch2 antenna until they reached a distance of approximately 100 m away. The two cars moved sequentially (Car 2 followed Car 1), as shown in Fig. 17 , which illustrates the geometrical configurations for the experiment site and the detection processing results for the 7th and 10th range cells, respectively. The two cars were separated by a distance of 20 m so as to be beyond the range resolution (17.3 m). It is shown from Fig. 17(a) that Car 1 exceeded Car 2 in both range and velocity, where Car 1 was detected at 48 m with a velocity of 28 km/h (−120 Hz) and Car 2 was detected at 24 m with a speed of 21 km/h (−90 Hz). Both cars were detected again in Fig. 17(b) but with a higher range and velocity, where Car 1 was detected at a range of 72 m with a velocity of 32 km/h (−135 Hz) and Car 2 at a range of 50 m with a velocity of 28 km/h (−120 Hz). The result of this scenario shows that the LTE-based passive radar system can detect and differentiate two cars following each other. The two cars were separated by a distance of 20 m and moved with varying speeds. Thus, we can conclude that the LTE-based passive radar system is suitable for use in applications such as border protection in the case where one may wish to monitor intruder vehicles.
PPT Slide
Lager Image
Scenario E: detection results in Doppler-range contour for (a) 7th range cell and (b) 10th range cell.
- 6. Scenario F
In scenario F, two cars and a motorbike were used in the same scene. The three vehicles, starting from the same position (parallel to each other), travelled from Ch2 antenna to a distance of 160 m away, in a straight line. The geometrical configurations for the experiment and detection processing results for the 8th and 14th range cells are illustrated in Fig. 18 . To examine the range and Doppler resolutions, the three vehicles were separated by a distance of 20 m, which is more than the range resolution (17.3 m), and moved with different speeds. It is shown from Fig. 18(a) that the three vehicles were detected at the same range (approximately 35 m) but with different velocities. The estimated positions for the three vehicles are illustrated in Fig. 18(a) , where the speeds were 20 km/h (−85 Hz) for Car 1, 16 km/h (−70 Hz) for Car 2, and 14 km/h (−60 Hz) for the motorbike. In Fig. 18(b) , the motorbike can still be seen at a range of 60 m with a velocity of 28 km/h (−120 Hz), while Car 1 and Car 2 are detected at ranges of 120 m and 100 m, with speeds of 44 km/h (−185 Hz) and 28 km/h (−120 Hz), respectively. It is noted that the detection peak of the motorbike is smaller than those of the cars because the motorbike has less RCS compared to the cars.
PPT Slide
Lager Image
Scenario F: detection results in Doppler-range contour for (a) 8th range cell and (b) 14th range cell.
From the results of this scenario, it can be deduced that the LTE-based passive radar system can detect and differentiate three vehicles separated at a distance slightly higher than the range resolution moving with different speeds; thus, the LTE-based passive radar system can be used in surveillance applications involving varying traffic speeds.
VI. Conclusion
An experimental LTE-based passive bistatic radar (PBR) system for detection was proposed, and realized. The proposed LTE-based PBR system performance was evaluated by conducting an outdoor field experiment using a real LTE eNB transmitter as an illumination source. In the experiment, six scenarios were carried out to investigate the system’s capability to detect diverse ground-moving targets, moving with different speeds and in different directions. In addition, the system’s capability to detect multiple targets moving on the ground in the same scene was examined. The experimental results showed that the LTE-based passive radar system has the capability to detect cars, a motorbike, and a human body, all of which can be moving at various speeds and at different ranges. Therefore, from the experimental results, there is no doubt that an LTE signal can be utilized as a source signal for a PBR system. This paper also demonstrates that both the designed system and the proposed algorithm can work effectively.
In spite of the positive results obtained, it should be pointed out that there is still a need for further studies and improvements. A future study could include implementing advanced signal processing algorithms for improving detection accuracy. Furthermore, more experiments can be conducted to investigate the system’s capability to detect and track ground-moving targets that are moving at very low speeds and have low profiles.
This work was supported by the Geran Putra of Universiti Putra Malaysia, Malaysia.
BIO
Corresponding Author r_syamsul@upm.edu.my
Raja Syamsul Azmir Raja Abdullah received his BEng degree in electronic and electrical engineering and MSc in communication system engineering and PhD in radar and microwave, system from the University of Birmingham, UK in 2000, 2001, and 2005, respectively. Since 2005 he has been with Department of Computer and Communication Systems Engineering, Faculty of Engineering, Universiti Putra Malaysia (UPM), Selangor, Malaysia, where he is now an associate professor. From 2008 to 2010 he worked as head of Wireless and Photonics Research Centre, UPM. And from 2011 to 2013 he worked as general manager for UPM International. His current research areas of interest include the microwave, radar systems, wireless sensor, wireless communications and signal processing.
asemsalah@gmail.com
Asem Ahmad Salah received his BSc degree majoring in telecommunication technology from Arab American University-Jenin (AAUJ), Palestine in 2005, and his MSc and PhD degrees in communications and network engineering from Universiti Putra Malaysia, (UPM) Selangor, Malaysia, in 2011 and 2015, respectively. From 2005 to 2008, he worked as a director of SHAREK Information Technology Center, Jenin, Palestine, at the same time; he was a part-time trainer in Hassib Sabbagh Information Technology Center of Excellence, AAUJ. Currently, he is working as an associate researcher in UPM, his research interest include radar, signal processing, wireless communications and applications of communication systems.
alyani@upm.edu.my
Alyani Ismail received her BEng (Hons) electronic and information engineering from University of Huddersfied, UK in 1999, and her MSc and PhD in engineering from University of Birmingham, UK in 2001 and 2006, respectively. Since 2006 she has been with the Department of Computer and Communication Systems Engineering, Faculty of Engineering, Universiti Putra Malaysia, Selangor, where she is now an associate professor. She is an active member of IEEE and she served as an Executive Committee for IEEE AP/MTT/EMC Chapter. Her area of specialization is micromachining, microwave filters and other components, RF MEMS and millimeterwave. Currently she involved in new research areas such as microwave passive filters and antennas; ultra-wide band components; RF MEMS; micromachined microwave and millimeter-wave components; Terahertz Applications, RF metamaterials; wireless communications, engineering education; outcome based education, delivery methods, project based learning, assessment and evaluation and Islamic approach to engineering education.
fazirul@upm.edu.my
Fazirulhisyam Hashim received his BEng degree from Universiti Putra Malaysia (UPM), Selangor, and holds his MSc degree from Universiti Sains Malaysia, Penang, and his PhD degree from the University of Sydney Australia. He is currently a researcher and a lecturer at the Wireless and Photonic Network Research Center of Excellence at the UPM. His research interests include network security and QoS of next generation mobile networks, green communication systems, and wireless sensor networks. He has published over 70 journal and conference papers. He is the current Chair of IEEE ComSoc/VTS Malaysia Chapter (year 2015), and he was the Vice Chair (2014) and the Secretary (2013). In addition, he was the Chair of IEEE Malaysia Young Professionals (year 2013-2014). He also involved in organizing a number of conferences; Secretary for APCC 2011 (Kota Kinabalu, Sabah), MICC 2013 (Kuala Lumpur) and ISTT 2014 (Langkawi, Kedah), Malaysia. He is also a TPC member/reviewer for many international conferences and journals.
emileen98@salam.uitm.edu.my
Nur Emileen Abdul Rashid received her Bachelor of electrical engineering (telecommunication engineering) from the Universiti Kebangsaan Malaysia, Selangor, in 2001 and subsequently her MSc and PhD in computer, communication and human centered engineering from the University of Birmingham, UK in 2002 and 2011, respectively. Currently she is a senior lecturer at University Technology MARA, Selangor, Malaysia. Her current research interests include radar technology, telecommunication signal processing and clutter modeling.
fizahaziz81@gmail.com
Noor Hafizah Abdul Aziz received her BEng in microelectronic engineering and MSc in electrical, electronic and system engineering from Universiti Kebangsaan Malaysia, Selangor in 2004 and 2007 respectively. Since 2008, she has been with the Centre for Communication Engineering Studies, Faculty of Electrical Engineering, Universiti Teknologi MARA, Selangor, Malaysia, where she is currently works as a lecturer. From 2013 onwards, she is undergoing her PhD program majoring in Communications and Network Engineering at Universiti Putra Malaysia, Selangor. She is undertaking her research on passive forward scattering radar system, classification, signal processing and applications of communication system.
References
Tan D. 2005 “Passive Radar Using Global System for Mobile Communication Signal: Theory, Implementation and Measurements” IEE Proc. Radar, Sonar Navigation 152 (3) 116 - 123    DOI : 10.1049/ip-rsn:20055038
Tan D. , Sun H. , Lu Y. “Sea and Air Moving Target Measurements Using a GSM Based Passive Radar” IEEE Radar Conf. VA, USA May 9–12, 2005 783 - 786
Wang Q. , Hou C. , Lu Y. 2010 “An Experimental Study of Wimax-Based Passive Radar” IEEE Trans. Microw. Theory Techn. 58 (12) 3502 - 3510
Chetty K. “Passive Bistatic WiMAX Radar for Marine Surveillance” IEEE Radar Conf. Washington, DC, USA May 10–14, 2010 188 - 193
Ahmadi S. 2009 New Directions in Wireless Communications Research Springer Hillsboro, OR, USA “An Overview of 3GPP Long-Term Evolution Radio Access Network,” 431 - 465
Chadchan S. , Akki C. 2010 “3GPP LTE/SAE: An Overview,” Int. J. Comput. Elect. Eng. 2 (5) 806 - 814
Salah A. “Feasibility Study of LTE Signal as a New Illuminators Of Opportunity for Passive Radar Applications” IEEE Int. RF Microw. Conf. Penang, Malaysia Dec. 9–11, 2013 258 - 262
Salah A. 2014 “Experimental Study of LTE Signals as Illuminators of Opportunity for Passive Bistatic Radar Applications” Electron. Lett. 50 (7) 545 - 547    DOI : 10.1049/el.2014.0237
Abdullah R.S.A.R. 2016 “Experimental Investigation on Target Detection and Tracking in Passive Radar Using Long-Term Evolution Signal” IET Radar Sonar Navigation 10 (3) 577 - 585    DOI : 10.1049/iet-rsn.2015.0346
Cherniakov M. 2006 “Automatic Ground Target Classification Using Forward Scattering Radar” IEE Proc. Radar, Sonar Navigation 153 (5) 427 - 437    DOI : 10.1049/ip-rsn:20050028
Willis N.J. 2005 “Bistatic Radar” SciTech Publishing Raleigh, NC, USA 131 - 145
Richards M.A. 2005 “Fundamentals of Radar Signal Processing,” Tata Mcgraw-Hill Education NY, USA 295 - 345
Tsao T. 1997 “Ambiguity Function for a Bistatic Radar” IEEE Trans. Aerosp. Electron. Syst. 33 (3) 1041 - 1051    DOI : 10.1109/7.599331
Jan P. 2013 “Computing of Bistatic Cross-Ambiguity Function on GPU” Symp. Ind. Electron. Taipei, Taiwan 1 - 5