Effectiveness of Linear FM Interference Signal on Tracking Performance of PLL in Monopulse Radar Receivers

Transcription

1 202 Effectiveness of Linear FM Interference Signal on Tracking Performance of PLL in Monopulse Radar Receivers Harikrishna Paik*, Dr.N.N.Sastry, Dr.I.SantiPrabha Assoc.Professor, Dept. of E&I Engg, VRSEC, Vijayawada, India Professor & Dean, R&D Wing, VRSEC, Vijayawada, India Professor, Dept. of ECE, JNT University, Kakinada, India Abstract Monopulse radar receivers employing phase locked loop (PLL) invariably track the target in range and angle (frequency) domains. In this paper, the tracking performance of PLL for a highly application specific airborne tracking radar in frequency domain is presented. Two different signals such as the linear frequency modulated (LFM) interference signal from a repeater source and the radar echo signal are injected into the PLL simultaneously with an assumption that initially, the PLL locks onto the echo signal frequency. The effectiveness of key parameters of LFM signal such as frequency deviation, modulation rate and LFM signal power on break-lock are demonstrated. The simulation result shows that break-lock is achieved at a frequency deviation of 0.36 MHz for a typical LFM signal power of -14 dbm and 200 khz modulation rate when the radar echo power at the PLL input is -14 dbm. The break-lock is studied for a typical loop bandwidth of 200 khz and different values of radar echo signal power at the PLL input. The computer simulation is carried out using visual system simulator (VSS) AWR software and potential conclusions are demonstrated. Index Terms frequency modulation, jamming, monopulse, radar receiver, radar echo, tracking I. INTRODUCTION Several advanced techniques have been employed in modern tracking radar against electronic attack (EA) threat in electronic warfare scenario. These include pulse compression, Pulse Doppler, Monopulse, Ultralow sidelobe antennas and Coherent sidelobe cancellation etc. All these techniques make modern radars difficult to jam and require special EA techniques for effective jamming. Of these techniques, monopulse technique is invariably used in all the tracking radars and modern missile seekers [1]. However, monopulse angle (frequency) tracking systems are difficult to jam because this technique provides an inherent resistance to amplitudemodulated jamming waveforms from point target sources [2]. Several jamming techniques have been employed against monopulse tracking systems which include noise jamming, deception jamming, transponder jamming and support jamming. In these techniques, the radar receiver is jammed either by introducing imperfections in the monopulse design or by using multiple repeater sources in order to distort the angle of arrival of the echo signal. Thus, the monopulse tracker is caused to move away from the target and results break-lock in the missile radar [3]. In this paper, the jamming of monopulse radar receiver using deception jamming is illustrated. The basic objective of the deception jamming scheme is to cause frequency (angular) breaklock by injecting suitable modified replica of radar echo signal into the tracking radar receiver. The linear frequency modulated (LFM) signal from a repeater source is used as an interference signal which is injected into the phase locked loop (PLL) along with radar echo signal and break-lock in the PLL is analyzed. The effects of LFM interference signal on tracking performance of PLL in monopulse radar receiver have been presented in several studies [4-7]. The statistical parameters of the monopulse receiver such as steady state gain; tracking index and range Doppler coupling coefficients in the presence of linear frequency modulation signal have been derived in [4]. The time-frequency characteristics of non-stationary LFM signal is estimated more precisely using Wigner Ville

2 203 Transform to detect the presence of wrong frequency component [5]. The maximum likelihood angle estimation technique is analyzed [6-7] for the detection of two closely unresolved targets in the sea clutter environment by implementing modified generalized likelihood ratio. Here, we propose the generation of the LFM signal by frequency modulating the sinusoidal carrier by sawtooth waveform using a frequency modulator. The PLL is assumed to be operating at an intermediate frequency (IF) of 50 MHz with a typical bandwidth of 200 khz. The radar echo and LFM jamming signal are applied to the PLL simultaneously. Initially, it is assumed that the PLL locks onto the radar echo signal frequency. The frequency deviation of LFM signal is increased such that the frequency separation between these two signals increases. Thus, the PLL loses the frequency lock from the echo signal and locks on to the jamming signal frequency. The frequency deviation required for break-lock is estimated as a function of LFM jammer power for different modulation rates such as 200, 300, 400 khz, and so. The break-lock is studied for different values of jammer signal power in the range between -14 and -2 dbm and the echo signal powers of -14 and -10 dbm. II. SYSTEM CONFIGURATION AND MODELLING A. Monopulse Radar Receiver The basic block diagram of two channel monopulse receiver is shown in Fig.1. Fig.1. Block diagram of Monopulse Receiver As shown in Fig. 1, the antenna feeds at the receiver front end receive the echo signal reflected from the target and the repeater interference signal. The outputs of antenna feeds are then given to two inputs of hybrid junction, which is a four port microwave device with two input and two output ports. When the signals from two antenna feeds are applied at the input ports, the sum and difference of the two are obtained at the output ports. The resulting signals in each sum and difference channel are heterodyned to an intermediate frequency (IF) and then amplified by an IF amplifier as necessary. The amplitude of the difference signal indicates the magnitude of the angular error, and its phase indicates the direction of the off-axis error. The sum channel signal is applied to the PLL to predict the break-lock behavior of the receiver. The monopulse radar receiver implemented using visual system simulation (VSS) AWR software is shown in Fig.2.

3 204 Fig.2. Monopulse receiver implemented using VSS As shown in Fig.2, the sinusoidal CW radar echo and the LFM interference signal are applied at the receiver input simultaneously. The quadrature hybrid coupler at the receiver front end divides the input signal into sum and difference channel signals which are 90 0 out of phase to each other. The signals in each channel are then amplified by an RF amplifier and heterodyned to an IF frequency centered at 50 MHz. The sum channel signal is applied to the PLL and break-lock is predicted. B. Phase locked loop The monopulse radar receiver invariably employs PLL as a frequency tracking subsystem. The PLL mainly includes a phase detector (PD), loop filter and a voltage controlled oscillator (VCO). In a classical PLL, when the reference signal and repeat jamming signal are simultaneously applied at its input, the PLL acquires lock onto the reference signal if the strength of the reference signal is larger than the interference signal strength. When strength of interference signal exceeds the echo signal strength, the PLL loses the frequency lock from the reference signal and locks onto certain other frequency [8]. Thus, break-lock is said to occur in the PLL. For our simulation, the PLL with charge pump PD and a third order passive loop filter is designed and implemented VSS software. The charge pump PD is chosen due to the fact that the charge pump PD provides infinite pull-in range and zero steady state error [9]. Furthermore, a third order filer is generally recommended for the most of RF applications and it is rare that a PLL is constructed with a filter higher than third order. In addition, the passive loop filter has the advantage over active filter that there is no active device to add noise into the PLL. The loop filter is designed using exact method. The exact method of filter design involves with solving the time constants and then determining the loop filter components from these time constants. The different key parameters considered for design of the loop filter are phase margin (), loop bandwidth (f c ), phase detector gain (K in ma), VCO gain (K vco in MHz/volt) and pole ratio (T 31 ). The pole ratio is the ratio of third order pole and reference pole of the loop filter. The phase margin determines the loop stability which is typically chosen between 48 and 55 degree. The loop bandwidth is the crucial parameter in filter design which determines the spur rejection and lock time of the loop. The selection of pole ratio has an impact on reference spur in the loop. The time constants of filter are determined from phase margin () of the loop forward gain [G(s)] given by [10]. tan 1 ( T tan 1 ( tan 1 2 ) T 1) ( T 1 T 31 ) (1) where, T 1,T 2 and T 3 are the filter time constants. The loop forward gain G(s) is given by: k (2) k vco G ( s ) Z ( s ) S where, Z(s) is loop impedance, K is phase detector gain, K vco is gain of VCO. In equation (1), the value of phase margin () and the pole ratio (T 31 ) are known, so an equation containing T 1 and T 2 can be obtained. Another equation of T 1 and T 2 can be obtained by finding the maximum value of phase margin at a frequency equal to the loop bandwidth. It is seen that the loop maximizes the phase margin at a frequency equal to loop bandwidth [11]. So, we can write

4 205 d 0 d c Solving the above equation, we can express (3) c T ( T ) c 2 c T 1 c T ( T ) 1 ( ) c 1 c T 3 (4) Now, solving equation (1) and (4) for two unknowns, the time constants T 1 and T 2 can be determined. The time constant T 3 can be obtained by using the relation T 3 (5) T 1 T 31 Once the time constants are determined, the loop filter components are obtained by defining the constants k 1, k 2, k 3 and k 4 given as: k 1 c tot k 2 ( T 1 T 3 ) k 1 (6) (7) k 3 T 1 T 3 k 1 (8) T 2 Fig.3. HMC design window k c 3 4 (9) c1 By solving the above four equations, the filter components R 1, R 2, C 1, C 2 and C 3 are determined. The typical parameters chosen for design of the loop filter are: phase margin () = 55 deg., input frequency (f comp ) = 50MHz, loop bandwidth (f c ) = 200 khz, VCO output frequency (f out ) = 4 GHz, phase detector gain (K ) = 2.5 ma and VCO gain (K vco ) = 40 MHz/volt. The filter is designed using Hittite Microwave Corporation PLL Design and Analysis Tool. The design window which shows the filter component values is shown in Fig. 3. The filter components are found to be R 2 = 1 kω, R 3 = 7.5 kω, C 1 = 100 pf, C 2 =2.2 pf and C 3 = 15 pf. The third order PLL implemented using VSS software is shown in Fig. 4. Fig.4. Third order PLL C. LFM signal generation Several methods have been used for the generation of LFM signal. The LFM signal is generated by fractional synthesizer [12-13] in which the fractional logic controls the division ratio (N) of the programmable frequency divider to generate highly linear LFM signal. The main limitation of this method is quantization noise caused by division ratio as it takes only integer value. The proposed method generates the LFM signal by frequency modulating the sinusoidal carrier signal by a sawtooth waveform using an FM modulator. For computer simulation, the LFM

5 206 signal is generated with a sinusoidal carrier centred at an IF frequency of 50 MHz and modulating sawtooth waveform of 2 V (peak) as shown in Fig. 5. Fig.6. (a). Modulating signal Fig.5. LFM signal generation As shown in Fig. 5, the FM modulator block frequency modulates the sinusoidal carrier signal by a low frequency sawtooth modulating signal and produces baseband LFM signal. Thus, the instantaneous frequency of carrier signal is shifted by an amount equal to the modulating signal times the frequency sensitivity (MHz/volt) of the FM modulator. Specifically, if the modulating input signal is v(t), and carrier has amplitude (A) with center frequency (f c ), then the output signal, y (t) is given by [14]: A f y ( t ) exp[ j ( 2 t 2 v ( ) d )] off (10) where, f off is offset frequency from the carrier and k f is the frequency sensitivity(mhz/volt). The LFM signal is generated with following key parameters: modulating voltage (v m ) = 2 V, modulating frequency (f m ) = 200 khz, carrier center frequency = 50 MHz, carrier power = -14 dbm. The modulating signal and LFM spectrum are shown in Fig. 6. (a) and (b). k f t 0 Fig.6. (b). LFM spectrum The spectrum of the LFM signal shows that the signal is centred at 50 MHz and signal bandwidth is about 5 MHz. III. COMPUTER SIMULATION The designed configuration of the PLL is modeled using VSS simulator AWR software. The elements are modeled using the elements available in VSS library. The computer simulation is carried out at the IF stage of receiver with the sum channel signal at PLL input. The radar echo signal with a typical power of -14 dbm at 50 MHz is applied at the PLL

6 207 input along with the LFM interference signal centered at 50 MHz (200 khz modulation rate). Initially, the power of the LFM signal is assumed to be relatively less compared to echo signal power as a condition for good tracking. Initially, it is seen that the PLL locks onto the radar echo signal frequency. Then the frequency deviation of LFM signal is adjusted such that the frequency separation between these two signals increases. At a certain value of frequency deviation, the PLL loses the frequency lock from the echo signal and locks onto the jamming signal at certain other frequency. The break-lock is observed through the frequency spectrum of the PLL. The deviation required for break-lock is measured as a function of jammer power and different modulation rates such as 200, 300, 400 khz and so. The simulation is also carried out with the echo signal power of -10 dbm applied at the PLL input. Fig.7. (b) PLL spectrum at 0.36 MHz frequency deviation with echo power of -14 dbm IV. RESULTS AND DISCUSSION The simulation results of break-lock measured through frequency spectrum of the PLL are shown in Fig. 7(a), 7(b) and Fig. 8(a), 8(b). The results are presented for a typical jammer power of -14 dbm, modulation rate of 200 khz when the echo powers at the PLL input are -14 and -10 dbm. Fig.8. (a) PLL spectrum at 0.40 MHz frequency deviation with echo power of -10 dbm Fig.7. (a) PLL spectrum at 0.25 MHz frequency deviation with echo power of -14 dbm

7 208 Fig.8. (b) PLL spectrum at 0.51 MHz frequency deviation with echo power of -10 dbm It is clear from Fig. 7(a) that the radar echo signal power at 50 MHz is -14 dbm and the PLL output signal power at 4 GHz is 10 dbm when the frequency deviation is 0.25 MHz. This shows that the PLL is locked onto the echo signal frequency as the PLL output frequency is equal to the VCO frequency. From the PLL spectrum as shown in Fig. 7(b), it is seen that the PLL output signal is at 4.08 GHz (different from 4 GHz) when the frequency deviation is 0.36 MHz. So, it demonstrates that break-lock in the PLL is achieved at a frequency deviation of 0.36 MHz when jammer power is -14 dbm and the echo signal power at the PLL input is -14 dbm. Fig. 8(a) and (b) show the PLL spectrum at frequency deviations of 0.40 MHz and 0.51 MHz, respectively when the echo power at PLL input is -10 dbm. It is clear from Fig. 8(b) that the PLL output signal is at 4.08 GHz demonstrating that break-lock is achieved at a frequency deviation of 0.51 MHz. From the above results, it can be estimated that larger value of frequency deviation is required for break-lock when the echo signal power is large. The simulated results of frequency deviation required for break-lock as a function of jammer power for different modulation rates are shown in Fig. 9(a) and Fig. 9(b). Fig.9. (a). Frequency deviation Vs jammer power with echo power of -14 dbm Fig.9. (b). Frequency deviation Vs jammer power with echo power of -10 dbm From Fig. 9(a), it is clear that for a typical modulation rate of 200 khz, the frequency deviation required for break-lock is 0.36 MHz at -14 dbm jammer power and it is 0.19 MHz at -2 dbm jammer power demonstrating that the PLL loses the frequency lock at higher value of frequency deviation when the jammer power is

8 209 less. Similarly, when the jammer is high, the break-lock is achieved at lower value of frequency deviation. It is also estimated that when the modulation rate of LFM signal is low, break-lock is achieved at lower value of frequency deviation. Similarly, larger value of deviation is required for break-lock when modulation rate is high. From Fig. 9(a), it is seen that the break-lock is achieved at 0.36 MHz when modulation rate is 200 khz and it is 0.91 MHz when modulation rate is 500 khz. So, from the above results it can be demonstrated that the jammer power and frequency deviation are the key parameters for breaking the frequency lock in the PLL. Thus, for effective jamming of the monopulse receiver, the LFM jamming signal with suitable power and frequency deviation is to be injected into the receiver along with the radar echo signal. V. CONCLUSIONS The break-lock of phase locked loop in monopulse receiver in the presence of LFM jamming signal has been presented. It is demonstrated that the break-lock is achieved at a frequency deviation of 0.36 MHz for a typical modulation rate of 200 khz when both the echo and LFM signal powers are -14 dbm. Furthermore, it is estimated that at jammer power of -14 dbm, the frequency deviations required for break-lock are 0.36 and 0.91 MHz when the modulation rates of LFM signal are 200 and 500 khz respectively. It is also verified that breaklock is achieved at lower value of frequency deviation when the modulation rate is less and at higher modulation rate, larger value of frequency deviation is required for break-lock. So, from the above simulation results, conclusions can be drawn that for effective jamming of monopulse receiver using LFM jamming signal, the jammer power and frequency deviation are to be selected suitably. ACKNOWLEDGMENT The authors would thanks to Siddhartha Academy of General & Technical Education who provided the research facilities and technical support. We are also indebted to Director and Principal of the institute who extended support in many useful technical discussions during the progress of this work. REFERENCES [1] D. C. Schleher, Introduction to Electronic Warfare, 1 st ed., Norwood, MA: Artech House, [2] Samuel M Sherman, Monopulse Principles and Techniques, 2 nd ed., Artech House, [3] D. C. Schleher, Electronic Warfare in the Information Age, 2 nd ed., Norwood, MA: Artech House, [4] Winnie Wong, W. D. Blair, Steady-State Tracking with LFM Waveforms, IEEE Transactions on Aerospace and Electronic Systems, Vol. 36, no. 2, pp , April [5] Christophe De Luigi, Eric Moreau, An Iterative Algorithm for Estimation of Linear Frequency Modulated Signal Parameters, IEEE Signal Processing Letters, Vol. 9, no. 4, pp , April [6] A. Sinha, T. Kirubarajan, Y. Bar-Shalom, Tracker and Signal Processing for the Benchmark Problem with Unresolved Targets, IEEE Transactions on Aerospace and Electronic Systems, Vol. 42, no. 1 pp , Jan [7] Blair W.D., Brandt-Pearce M., Monopulse DOA Estimation of Two unresolved Rayleigh Targets, IEEE Transactions on AES, Vol.37, no.2, pp , April [8] F M. Gardner, Phase lock Techniques, 2 nd ed., John Wiley, [9] Gardner F.M., Charge-Pump Phase-Lock Loops, IEEE Trans. Comm., Vol. COM-28, pp , [10] Keese, William O., An Analysis and Performance Evaluation of a Passive Filter Design Technique for Charge Pump Phased Locked Loops, Application Note, National Semiconductor, [11] Franklin F., Powell D., and Emami-Naeini A., Feedback Control of Dynamic Systems, 3 rd ed., Addison-Wesley, [12] T. Musch and B.Schiek, A highly linear frequency ramp generator based on a fractional divider phase locked loop, IEEE Trans. Instrum. Meas., Vol. 48, pp , April [13] T. Musch, I. Rolfes and B. Schiek, Fractional divider concepts with phase locked control for the generation of precise linear frequency ramp, in 28 th EUMC Proc.., Amsterdam, Netherlands, pp , Oct [14] S. Haykin, Communication Systems, 3 rd ed., John Wiley & Sons, New York, 1994.