Microwave photonic de-chirp receiver for breaking the detection range swath limitation

Zhenwei Mo; Zhenwei Mo; Chenyu Liu; Chenyu Liu; Jiyao Yang; Jiyao Yang; Yuchuang Sun; Yuchuang Sun; Ruixuan Wang; Ruixuan Wang; Jingwen Dong; Wangzhe Li; Wangzhe Li

doi:10.1364/OE.422262

1. Introduction

In recent years, applying microwave photonics (MWP) technologies in a radar becomes a research hotspot because inherent advantages of photonics, such as low transmission loss, electron-magnetic interference resistance, and ultrawide bandwidth, etc., address many drawbacks of the radio frequency (RF) link in the conventional radars [1–4]. A great variety of MWP radars, such as MWP synthetic aperture radars (SARs) [5–8], MWP dual-band radars [9,10], MWP multiple-input multiple-output (MIMO) radars [11–13], have been proposed and demonstrated, showing superiorities over conventional electronic radars in terms of bandwidth, multi-functional and reconfigurability. A large number of MWP radar receivers are based on de-chirp reception which is capable of converting large bandwidth linear frequency modulated (LFM) echoes into intermediate frequency (IF) signals with narrower bandwidth to relieve the pressure of an analog-to-digital converter (ADC) [14] and suitable for the MWP radar with large bandwidth. However, the detection range swath of the de-chirp receiver is limited by two factors. One factor is the difficulty of generating a large pulse width reference signal with the same chirp rate as a transmitted signal, which is necessary to obtain a long de-chirp processing time window without resolution deterioration [15]. The other factor is that the limited sample rate of the ADC restricts the bandwidth of the de-chirped signal which is proportional to the detection range swath [16]. These problems become more serious in the MWP de-chirp radar due to the significantly increased operation bandwidth.

To bypass these issues, tunable time delays are added to the reference signal by inserting optical fibers with different lengths [17,18] or changing the generation time of the reference signal directly [19]. In these methods, extra efforts are needed to obtain a coarse target distance for further focusing, resulting in the difficulty of simultaneously surveillance targets that are continuously distributed in a wide range swath scene. In [20], a channelized photonic-assisted de-chirp receiver has been proposed to extend the detection range swath. However, the number of channels needs to be increased along with the increment of the detection range swath, resulting in a great system complexity in the application demanding an extremely wide range swath. Hence, the MWP de-chirp receivers in previous works cannot meet the requirement of the radar with high-resolution wide-swath (HRWS) imaging ability [21–23].

In this paper, a novel MWP radar de-chirp receiver that breaks the detection range swath limitation is proposed and experimentally demonstrated. In the proposed receiver, an LFM pulse train is modulated on a continuous-wave (CW) light and routed into multiple reception channels. Optical reference signal replicas, which are LFM pulses with different time delays, are obtained by optical delaying the LFM pulse train modulated light through fibers with specific lengths in the different reception channels. By properly setting the parameters of the LFM pulse train and the fiber induced time delay in each reception channel, an echo reflected from any target within a wide range swath always temporally overlaps with one of the optical reference signal replicas in the multiple reception channels so that the echo can be fully de-chirped without bandwidth loss and resolution deterioration. After digital processing the de-chirped signals in all the reception channels, range information of the targets in a wide swath scene can be recovered correctly.

The detection range swath of the proposed de-chirp receiver can be extended enormously because the duration of the single reference pulse and the bandwidth of the de-chirped signal are constants that independent from the detection range swath. To the best of our knowledge, it is the first time to break the detection range swath limitation of the MWP radar de-chirp receiver. To investigate the performance of the proposed receiver, simulated echoes reflected from the targets distributed in a wide range swath are received, showing the high-resolution detection ability in a wide range swath scene. High-resolution 2D imaging ability in a wide range swath is also verified by an ISAR imaging experiment.

2. Principle

Figure 1 shows the schematic diagram of the setup of the proposed MWP de-chirp receiver, which consists of an optical reference signal generator, N reception channels, and a digital signal processor (DSP). The optical reference signal generator generates the master copy of an optical reference signal. In the optical reference signal generator, a CW light from a laser is sent to a Mach-Zehnder modulator (MZM) and modulated by a reference LFM pulse train which is generated by an arbitrary waveform generator (AWG) and expressed as:

(1)$${S_{ref}}(t )= \sum\limits_{m = 1} {\textrm{rect}\left( {\frac{t}{{{T_{ref}}}} - m} \right){V_{ref}}\cos [{{\omega_{ref}}({t - m{T_{ref}}} )+ \pi k{{({t - m{T_{ref}}} )}^2}} ]}$$

where S_ref (t) is the electric field of the reference signal, V_ref is the amplitude, ω_ref is the center angular frequency, T_ref is the reference pulse duration, k is the chirp rate, and m=0, 1, 2, 3, …, M−1, denotes the (m+1)th LFM pulse in the reference pulse train.

Fig. 1. The schematic diagram of the proposed receiver. CW-Laser: continuous-wave laser; MZM: Mach-Zehnder modulator; OC: optical coupler; AWG: arbitrary waveform generator; EC: electrical coupler; ODL: optical delay line; PM: phase modulator; OBPF: optical bandpass filter; PD: photodetector; BPF: bandpass filter; LNA: low-noise amplifier; ADC: analog-to-digital converter; DSP: digital signal processor.

Download Full Size | PDF

The MZM is biased at a minimum transmission point (MITP) and the obtained carrier suppressed double sideband (CS-DSB) modulated optical reference signal is written as:

(2)$$\begin{array}{{c}} {{E_{MZM}}(t )= \sum\limits_{m = 1} { - \sqrt 2 \textrm{rect}\left( {\frac{t}{{{T_{ref}}}} - m} \right){A_0}j\textrm{exp} ({j{\omega_0}t} ){J_1}({{\beta_{ref}}} )\times } }\\ {\textrm{ }\left\{ \begin{array}{l} \textrm{exp} [{j({ - {\omega_{ref}}({t - m{T_{ref}}} )- \pi k{{({t - m{T_{ref}}} )}^2}} )} ]+ \\ \textrm{exp} [{j({{\omega_{ref}}({t - m{T_{ref}}} )+ \pi k{{({t - m{T_{ref}}} )}^2}} )} ]\end{array} \right\}} \end{array}$$

where A₀ is the amplitude of the electric field of the incident light wave, ω₀ is the angular frequency of the optical carrier, J_i denotes the ith-order Bessel function of the first kind. β_ref =πV_ref /V_π is the modulation index, V_π is the half-wave voltage of the MZM.

To generate the optical reference signal replicas with different time delays for successfully de-chirping with echoes reflected from the targets in a wide range swath scene, the master copy of the optical reference signal is split into several copies by an optical coupler (OC). The copies in the different reception channels are delayed by optical delay lines (ODLs) with specifically designed lengths. The delayed optical reference signal copy in the nth reception channel is expressed as:

(3)$$\begin{array}{{l}} {{E_{refn}}(t )= \sum\limits_{m = 1} { - \sqrt 2 \textrm{rect}\left[ {\frac{{t - {\varepsilon_n}}}{{{T_{ref}}}} - m} \right]{A_0}j\textrm{exp} [{j{\omega_0}({t - {\varepsilon_n}} )} ]{J_1}({{\beta_{ref}}} )\times } }\\ {\textrm{ }\left\{ \begin{array}{l} \textrm{exp} \left[ {j\left( \begin{array}{l} - {\omega_{ref}}({t - {\varepsilon_n} - m{T_{ref}}} )- \\ \pi k{({t - {\varepsilon_n} - m{T_{ref}}} )^2} \end{array} \right)} \right] + \\ \textrm{exp} \left[ {j\left( \begin{array}{l} {\omega_{ref}}({t - {\varepsilon_n} - m{T_{ref}}} )+ \\ \pi k{({t - {\varepsilon_n} - m{T_{ref}}} )^2} \end{array} \right)} \right] \end{array} \right\}} \end{array}$$

where ɛ_n is the time delay introduced by the ODL in the nth channel, and ɛ_n₊₁ > ɛ_n. From (3), the optical reference signal replicas which are optical LFM pulses with time delays of ɛ_n+mT_ref are obtained as shown in Fig. 2.

Fig. 2. Time-frequency maps of the echoes and the optical reference signals at the PMs in the reception channels.

Download Full Size | PDF

The delayed optical reference signal replicas are sent to de-chirp processing units in the reception channels to de-chirp with the echoes. As shown in Fig. 1, the de-chirp processing unit is composed of a phase modulator (PM), an optical bandpass filter (OBPF), a photodetector (PD), and a microwave bandpass filter (BPF). The delayed optical reference signal replicas in each channel are modulated by the echoes in the PM. Assuming the echoes is:

(4)$${S_{echo}}(t )= \sum\limits_{p = 1} {\textrm{rect}\left( {\frac{{t - {\tau_p}}}{{{T_e}}}} \right){V_{echo}}\cos [{{\omega_e}({t - {\tau_p}} )+ 2\pi k{{({t - {\tau_p}} )}^2}} ]}$$

where V_echo is the amplitude of the echo from the pth target, T_e and ω_e are the pulse width and the angular frequency of the radar transmitted signal carrier respectively, and τ_p is the roundtrip time between the radar and the pth targets.

The OBPF is set after the PM to select the +1st order sideband of the optical reference signal and the +1st order sideband generated by modulating the −1st order sideband of the optical reference signal by the echoes, as shown in Fig. 3. The output of the OBPF in the nth reception channel is expressed as:

(5)$$\begin{array}{{l}} {{S_{OBPFn}} = \sum\limits_{m = 1} {\sum\limits_{p = 1} {\sqrt 2 {H_1}(t )\textrm{exp} \left[ {j\left( \begin{array}{l} {\omega_0}({t - {\varepsilon_n}} )+ {\omega_e}({t - {\tau_p}} )+ 2\pi k{({t - {\tau_p}} )^2} - \\ {\omega_{ref}}({t - {\varepsilon_n} - m{T_{ref}}} )- \pi k{({t - {\varepsilon_n} - m{T_{ref}}} )^2} \end{array} \right)} \right]} - } }\\ {\textrm{ }\sum\limits_{m = 1} {\sqrt 2 {H_2}(t )\textrm{exp} \left[ {j\left( \begin{array}{l} {\omega_0}({t - {\varepsilon_n}} )+ {\omega_{ref}}({t - {\varepsilon_n} - m{T_{ref}}} )\textrm{ + }\\ \pi k{({t - {\varepsilon_n} - m{T_{ref}}} )^2}\textrm{ + }{\pi / 2} \end{array} \right)} \right]} } \end{array}$$

where H₁(t)=rect[(t−ɛ_n)/T_ref−m]rect[(t−τ_p)/T_e]A₀J₁(β_ref)J₁(β_echo) and H₂(t)=rect[(t−ɛ_n)/T_ref−m] rect[(t−τ_p)/T_e]A₀J₁(β_ref)J₀(β_echo) are the amplitudes of the envelopes of the light fields. β_echo= πV_echo/V_πPM is the modulation index of the PM, and V_πPM is the half-wave voltage of the PM.

Fig. 3. Time-frequency maps of the optical reference signals and the echo-modulated optical signals at the output of the OBPFs in the reception channels.

Download Full Size | PDF

After the OBPF, the optical signal is detected by the PD to perform the optic to electronic conversion and accomplish the de-chirp processing. The de-chirped signal in the nth reception channel can be written as:

(6)$$\begin{aligned} {I_n} &= \eta {|{{S_{OBPF\_n}}} |^2}\\ &= \eta \left\{ \begin{array}{l} 2{H_1}^2(t )+ 2{H_2}^2(t )- 4{H_1}(t ){H_2}(t )\times \\ \sum\limits_{p = 1} {\textrm{Re} [{\textrm{exp} [{j({({{\omega_e} - 2{\omega_{ref}}} )t + 4\pi k({{\varepsilon_n} + m{T_{ref}} - {\tau_p}} )t - {\pi / 2} + {\varphi_d} + {\varphi_{RVP}}} )} ]} ]} \end{array} \right\}\\ &\textrm{ = }B(t )+ \sum\limits_{p = 1} {U(t )\cos [{({{\omega_e} - 2{\omega_{ref}}} )t + 4\pi k({{\varepsilon_n} + m{T_{ref}} - {\tau_p}} )t - {\pi / 2} + {\varphi_d} + {\varphi_{RVP}}} ]} \end{aligned}$$

where η is the responsivity of the PD, B(t)=η[2H₁²(t) + 2H₂²(t)] is the direct current component of the PD output, U(t) = 8ηH₁(t)H₂(t) is the amplitude of the alternating current component of the PD output, φ_d=2ω_ref(ɛ_n+mT_ref)−ω_eτ_p is the Doppler item which is necessary for azimuth focusing, φ_RVP=2πk[τ_p²−(ɛ_n+mT_ref)²] is the residual video phase (RVP) item which can be removed by post-processing [24].

To obtain a detection result without resolution deterioration, a de-chirped signal should be generated by a fully de-chirp processing which requires an echo is completely overlapped by a single reference pulse in the time domain [15]. Thus, the pulse duration of the reference pulse in the proposed receiver is designed larger than the pulse duration of the echo, and the difference between them is defined as the de-chirp processing time window which is denoted as T_w=T_ref−T_e. As shown in Fig. 2, when the time delay τ_p of an echo satisfies the condition of ɛ_n+mT_ref≤τ_p≤ɛ_n+mT_ref+T_w, the echo is completely overlapped by the (m+1)th reference replica in the nth reception channel. After de-chirping the echo with the reference replica, a de-chirped signal with an angular frequency in a range from ω_e−2ω_ref−4πkT_w to ω_e−2ω_ref is generated according to (6). A BPF with a bandwidth of B_dechirp=4πkT_w is set after the PD to select the de-chirped signal with an angular frequency within the range. As for the de-chirped signal with an angular frequency out of the range from ω_e−2ω_ref−4πkT_w to ω_e−2ω_ref, it is eliminated by the BPF. According to (6), the eliminated de-chirped signal is generated by the echo that has a time delay τ_p does not meet the condition of ɛ_n+mT_ref≤τ_p≤ɛ_n+mT_ref+T_w in the nth channel. Such an echo is partially overlapped by two of the adjacent reference replicas in the nth channel, which will cause the deterioration of the detection resolution. Hence, only the de-chirped signal generated by the fully de-chirp processing can provide the detection result without resolution deterioration and pass the BPF, as shown in the left side of Fig. 4.

Fig. 4. Left: time-frequency maps of the de-chirped signals at the outputs of BPFs in the reception channels; Right: the schematic diagram of the range information of targets recovered from the de-chirped signals.

Download Full Size | PDF

After the BPF, the de-chirped signal is sampled by an analog-to-digital converter (ADC) in the reception channel and then sent to the DSP for digital processing. In the DSP, the data captured from different reception channels is processed separately. The data from the nth reception channel is sliced into several data segments according to the time delays of the reference replicas in the nth reception channel, such as (ɛ_n, ɛ_n+T_ref), (ɛ_n+T_ref, ɛ_n+2T_ref), …, (ɛ_n+(m−1)T_ref, ɛ_n+mT_ref). By processing fast Fourier transform (FFT) to each data segment, the frequency information of the de-chirped signal is obtained. Since the channel number n and the reference replica number m of the processed data segment are known, the range information of targets distributed in a wide range swath scene is recovered by processing and time sorting the digital de-chirped signals according to:

(7)$$D = \frac{c}{2}{\tau _p} = \frac{c}{2}\left[ {\frac{{{\omega_e} - 2{\omega_{ref}} - {\omega_{dechirp}}}}{{4\pi k}} + {\varepsilon_n} + m{T_{ref}}} \right]$$

where ω_dechirp is the angular frequency of the de-chirped signal from (6), c is the light speed in free space. As mentioned above, the angular frequency of the de-chirped signal being processed in the DSP must in the range from ω_e−2ω_ref−4πkT_w to ω_e−2ω_ref. By substituting ω_e−2ω_ref−4πkT_w≤ω_dechirp≤ω_e−2ω_ref into (7), a range segment of ɛ_n+mT_ref≤D(m,n)≤ɛ_n+mT_ref+T_w is obtained. The range segments are sorted in the sequence of D(0, 1), D(0, 2), …, D(0, N), D(1, 1), D(1, 2), …, as shown in the right side of Fig. 4. To achieve detection without omission, the range segments should be distributed continuously. Thus, two designing restrictions of ɛ_n₊₁≤ɛ_n+T_w and ɛ_N+T_w≥ɛ₁+T_ref are derived from the continuity requirement of the range segments.

Due to the restrictions of ɛ_n₊₁≤ɛ_n+T_w and ɛ_N+T_w≥ɛ₁+T_ref, there is a minimum value requirement of the total channel number N. Assuming the fiber introducing time delay ɛ_n is uniformly increased by Δɛ along with the channel number n, which means ɛ_n=ɛ₁+(n−1)Δɛ. By substituting ɛ_n=ɛ₁+(n−1)Δɛ and B_dechirp=4πkT_w into the mentioned two designing restrictions, the relationship between the total channel number N and the channel bandwidth B_dechirp is obtained:

(8)$$N \ge \left\lceil {\frac{{{B_{echo}}}}{{{B_{dechirp}}}}} \right\rceil + 1$$

where B_echo is the bandwidth of the echo, ⌈x⌉ is a symbol of mapping x to the least integer greater than or equal to x. According to (8), there is a minimum total channel number when B_echo and B_dechirp are determined.

Thus, the echoes reflected from the targets distributed in a wide range swath scene are time-division de-chirped with the optical reference signal replicas in the proposed multi-channel de-chirp receiver. The bandwidths of the reference signal and the de-chirped signal are not increased along with the detection range swath because the de-chirp processing time window T_w of each optical reference signal replica is a constant. The allowed detection range swath of the proposed receiver is calculated as:

(9)$$Swath = {D_{\max }} - {D_{\min }} = \frac{c}{2}\left[ {\frac{{{B_{dechirp}}}}{{4\pi k}} + {\varepsilon_N} - {\varepsilon_1} + ({M - 1} ){T_{ref}}} \right]$$

From (9), when the de-chirped signal bandwidth B_dechirp and the number of reception channels N are fixed, the detection range swath is able to be extended by increasing the total pulse number M in the LFM pulse train. It means the allowed detection range swath of the proposed receiver can be flexibly extended without changing the hardware setup.

3. Experiment and results

To investigate the performance of the proposed de-chirp receiver, a hardware-in-the-loop experimental setup is implemented based on Fig. 5. The experimental setup consists of a receiver with four reception channels, and an echo simulator which is used to simulate the echoes from a wide swath detection scene under the circumstance of the laboratory. Due to the hardware constraint, in the receiver, channel 1 and channel 3 share one set of hardware while channel 2 and channel 4 share another hardware suite by replacing the ODLs as shown in Fig. 5(a). Because the information of targets is recovered from the de-chirped signal and every channel can implement the de-chirp processing independently, the time-division test method can prove the performance of the receiver that all the channels operate together.

Fig. 5. (a)The schematic diagram of the experimental setup. (b)The schematic diagram of the echo simulator operates in mode A and mode B.

Download Full Size | PDF

As shown in Fig. 5(a), the echo simulator and the receiver share one laser source (RIO, Orion), which emits CW light at 1550 nm with the power of 13dBm, by a 50:50 polarization-maintaining optical coupler (PM-OC).

Details of the echo simulator are shown in Fig. 5(b). In the echo simulator, an LFM pulse signal with a center frequency of 6.5 GHz, a bandwidth of 1 GHz, a pulse width of 10us, and a pulse repetition frequency (PRF) of 10kHz is generated by one of the signal generation channels of a dual-channel AWG (Keysight M8190A) and fed into an MZM (EOspace) in an MWP frequency doubling link to modulate the CW light. By biasing the MZM at a MITP, a frequency doubling LFM signal can be obtained at the output of a customized PD [6]. After being amplified by a low noise amplifier (LNA, Sample) and a power amplifier (PA, Keysight 83020A), and filtered by a BPF, the simulated echo pulse with a center frequency of 13 GHz and a bandwidth of 2 GHz is obtained, whose frequency spectrum is measured by an electrical spectrum analyzer (ESA, Keysight N9030A) and shown in Fig. 6(a). Because echoes can be regarded as the time-delayed copies of a radar transmitted pulse, the echoes reflected from any distance are simulated by software adding corresponding time delays to the generated LFM pulses in the AWG.

Fig. 6. (a) The spectrum of simulated echo pulse. (b)The spectrum of the reference LFM pulse train.

Download Full Size | PDF

In the receiver, the other channel of the dual-channel AWG generates a reference LFM pulse train which has six continuously repeated LFM pulses with a frequency range from 2.3 GHz to 3.8 GHz and a pulse width of 15µs. The spectrum of the reference LFM pulse train is shown in Fig. 6(b). The CW light from the shared laser is CS-DSB modulated by the reference LFM pulse train in a MITP biased MZM to generate ±1st order optical sidebands, and its optical spectrum is measured by an optical spectrum analyzer (OSA, YOKOGAWA AQ6307D) and shown in Fig. 7(a). The modulated light is used as the reference optical signal and split into two reception channels by a 50:50 OC.

Fig. 7. (a) The optical spectrum of the optical reference signal. (b) The transmission spectrum of the OBPF and the echoes modulated optical signal before and after passing the OBPF.

Download Full Size | PDF

To obtain optical reference replicas with different time delays, in each reception channel, the optical reference signal first passes through a section of length specific single mode fiber to introduce a time delay. As shown in Fig. 5, the relative time relations between the simulated echo pulse without delay and the optical reference replicas in each channel are measured at the test points (TPs) A, B, and C which include the consideration of time delays brought by an EDFA and a polarization controller (PC). In the test, the optical signals at TP_B and TP_C are converted to electrical signals by two PDs and the beat signals of the ±1st optical sidebands are recorded by an oscilloscope (Keysight DSO-X 92004A). As shown in Fig. 8, the relative delays between the simulated echo pulse and the optical reference pulse trains in channel 1 to channel 4 are 0.28µs, 4.25µs, 8.12µs, and 12.24µs respectively and the pulse duration of each optical reference replica is 15µs.

Fig. 8. The time-frequency analysis and the relative time relation of the zero-delay simulated echo pulse and the reference pulse train from Ch1 to Ch4. (a) The waveform of the zero-delay simulated echo pulse captured at test point A. (b)–(e) The waveform of the reference pulse train from Ch1 to Ch4 respectively captured at test point B and test point C.

Download Full Size | PDF

The EDFA is applied to enhance the power of the optical reference replicas and the PC is set to maximize the echo modulation efficiency in the following de-chirp processing. After passing the PC, the optical reference replicas are modulated by the simulated echoes in a PM (EOspace) and the optical spectrum of the echoes modulated optical signal at the output of the PM is shown as a green curve in Fig. 7(b). The +1st optical sideband generated by modulating the −1st sideband of the optical reference replica by the echoes and the +1st sideband of the optical reference replica are selected by an OBPF (EXFO, XTM-50) for further de-chirp processing. The transmission spectrum of the OBPF and the optical spectrum of the filtered optical signal are shown as an orange dotted curve and a purple curve respectively in Fig. 7(b). The selected optical signal is detected by a PD (Finisar, XPDV2120RA) to obtain the de-chirped signal. After amplifying by an LNA and passing through an anti-aliasing filter with a center frequency of 6.9 GHz and a bandwidth of 900 MHz, the de-chirped signal is sampled by the oscilloscope with a sample rate of 2.5GSa/s.

In the experiment, the de-chirp processing window of a single reference replica is 4.5µs because the chirp rate of the echo pulse is 200THz/s and the anti-aliasing filter has a bandwidth of 900 MHz, although the pulse width difference between a single optical reference replica and an echo pulse is 5µs. Thus, the time relations of the optical reference replicas in all the channels and the echo pulses satisfy the conditions of ɛ_N+T_w≥ɛ₁+T_ref and ɛ_n₊₁≤ɛ_n+T_w. The overall reception time window of the demonstrated receiver is from 0.53µs to 91.99µs, corresponding to a range swath of about 13.6 km. Compared to the conventional de-chirp receiver with a 4.5µs de-chirp processing window as set in the experimental setup, the overall reception time window as well as the detection range swath of the proposed receiver is 20 times larger.

To validate the experimental radar setup with the proposed receiver can keep high range resolution in a wide range swath detection scene, a detection test along the range direction based on simulated echoes is first demonstrated. In this experiment, 30 samples of the simulated echoes are generated in the echo simulator and directly fed into the receiver after power attenuation, as Mode A showed in Fig. 5(b). The delays of the simulated echoes are randomly distributed in the range from 0.53µs to 91.99µs. The de-chirped results of the simulated echoes are shown in Fig. 9 and Table 1. As shown in Fig. 9(a), the measurement results are recovered by stitching the de-chirped results in different de-chirp processing time windows according to (7). The de-chirped results lay in the same background noted by m are obtained by the echoes de-chirping with the (m+1)th reference pulse in different channels. There is an overlap of the de-chirped results between two adjacent de-chirp processing time windows. Figures 9(b)–9(e) show the zoom-in picture of the overlapping de-chirped results in the adjacent de-chirp processing time windows which are marked by the red dotted frames in Fig. 9(a). The overlapping results show targets can be correctly detected at the edge of the de-chirp processing windows and ensure high-resolution detection without a blind zone. Table 1 shows the setting time delays of the simulated echoes, which also in proportion to the setting distances of the simulated point targets, and the radar measured distances of the simulated targets. The 3 dB width of the main lobe of the de-chirped signal shown in Table 1 is an evaluation index of the range resolution [25]. As can be seen, the measured 3 dB width is 6.9 cm, which is close to the theoretical resolution of a 2 GHz bandwidth signal. The test result proves that the radar setup with the proposed de-chirp receiver can simultaneously detect targets in a wide swath without resolution deterioration.

Fig. 9. (a) The de-chirped results of simulated echoes. (b)–(e) The zoom in pictures of part of the overlapping de-chirped results in adjacent de-chirp processing time windows from channel 1 to channel 4.

Download Full Size | PDF

Table 1. Thirty samples of the simulated echoes setting and the measured results.

View Table

An ISAR imaging experiment based on moving trihedral corner reflectors (TCRs) is also demonstrated to test the 2D imaging ability of the experimental setup. In this experiment, the Doppler frequency information is modulated to the simulated echo pulses by illuminating the TCRs on a rotating platform, as Mode B showed in Fig. 5(b). The experimental scene and the placement of the TCRs are shown in Fig. 10. Three TCRs are used as the imaging targets in the experiment. As shown in Figs. 10(b) and 10(c), the range distances between them are 27 cm and 12 cm respectively, and the cross-range distances are 23 cm and 24 cm respectively. Four LFM pulses with time delays of 3µs, 21µs, 55µs, and 75µs are generated and illuminate the TCRs. The TCRs are placed on a rotating platform with a rotational speed of 60.36degree/s and the coherent integration time of an ISAR imaging processing is 0.1s so that the cross-range resolution is about 11 cm. The overall ISAR image after processing is shown in Fig. 11(a). Four small points can be found in the overall ISAR image which has a range swath of 12 km, and they are noted by numbers 1 to 4 in Fig. 11(a). The measured relative distances between point 2 to point 4 and point 1 are 2700 m, 7999.99 m, and 10799.99 m respectively, and they match well with the distance differences calculated by the delay differences of the four simulated echoes. The zoom-in images of these four points are shown in Figs. 11(b)–11(e). Three points can be distinguished clearly in the zoom-in images, and the relative position of the three points match well with the actual placement of the TCRs. The ISAR imaging experiment results show that the targets distributed in a wide range swath can be focused well simultaneously by using the proposed de-chirp receiver.

Fig. 10. (a) The ISAR imaging experimental scene; (b) The range distance of the TCRs; (c) The cross-range distance of the TCRs.

Download Full Size | PDF

Fig. 11. (a) The overall ISAR image; (b)–(e) The zoom-in images of the ISAR image of the TCRs distributed at different distances corresponding to the time delay setting of the simulated echoes.

Download Full Size | PDF

The spurious-free dynamic range (SFDR) of the reception channel is also tested. A two-tone signal with frequencies of 13 GHz and 13.02 GHz is generated by a signal generator (Agilent, E8267D). The reference signal is a 3.05 GHz single tone signal generated by the AWG. The spectrum of the mixed output is measured by the ESA. The frequencies of the ideal mixed output are 6.9 GHz and 6.92 GHz and the frequencies of the third-order intermodulation distortion (IMD3) are 6.88 GHz and 6.94 GHz. The measured result is shown in Fig. 12. The SFDR of the reception channel in the proposed receiver is 83.91dBc·Hz^2/3.

Fig. 12. Measured fundamental and IMD3 terms at the output of the proposed receiver when a two-tone signal is received.

Download Full Size | PDF

4. Conclusion

In this paper, a novel multi-channel time-division photonics de-chirp receiver that achieves high-resolution wide range swath detection is proposed and demonstrated. In the proposed receiver, multiple de-chirp processing time windows are generated by using fibers in different channels to introduce different time delays to the replicated optical reference signals. Echoes reflected from a wide range swath scene can be de-chirped in corresponding de-chirp processing time windows without resolution deterioration. The experimental radar setup with the proposed de-chirp receiver is evaluated that the detection range swath achieves 13 km with a range resolution of 6.9 cm. The ISAR imaging experiment results prove that the proposed receiver has great potential in large scale radar imaging application. Last but not least, the bandwidths of the reference signal and the de-chirped signal are de-coupled with the detection range swath, so that the detection range swath is able to extend by increasing the number of the optical reference replicas in a software approach instead of upgrading the hardware.

Funding

National Key Research and Development Program of China (2018YFA0701900, 2018YFA0701901); National Natural Science Foundation of China (61690191, 61701476).

Disclosures

The authors declare no conflicts of interest.

References

1. M. I. Skolnik, Introduction to Radar System3rd (McGraw-Hill Education, 2002).

2. P. Ghelfi, F. Laghezza, F. Scotti, G. Serafino, A. Capria, S. Pinna, D. Onori, C. Porzi, M. Scaffardi, A. Malacarne, V. Vercesi, E. Lazzeri, F. Berizzi, and A. Bogoni, “A fully photonics-based coherent radar system,” Nature 507(7492), 341–345 (2014). [CrossRef]

3. X. Zou, B. Lu, W. Pan, L. Yan, A. Stohr, and J. Yao, “Photonics for microwave measurements,” Laser Photonics Rev. 10(5), 711–734 (2016). [CrossRef]

4. A. Bogoni, P. Ghelfi, and F. Laghezza, Photonics for Radar Networks and Electronic Warfare Systems (SciTech Publishing, 2019).

5. Y. Xu, S. Li, X. Xue, X. Xiao, X. Zheng, and B. Zhou, “An interleaved broadband photonic ADC immune to channel mismatches capable for high-speed radar imaging,” IEEE Photonics J. 11(4), 1–9 (2019). [CrossRef]

6. R. Li, W. Li, M. Ding, Z. Wen, Y. Li, L. Zhou, S. Yu, T. Xing, B. Gao, Y. Luan, Y. Zhu, P. Guo, Y. Tian, and X. Liang, “Demonstration of a microwave photonic synthetic aperture radar based on photonic-assisted signal generation and stretch processing,” Opt. Express 25(13), 14334–14340 (2017). [CrossRef]

7. F. Zhang, Q. Guo, Z. Wang, P. Zhou, G. Zhang, J. Sun, and S. Pan, “Photonics-based broadband radar for high-resolution and real-time inverse synthetic aperture imaging,” Opt. Express 25(14), 16274–16281 (2017). [CrossRef]

8. X. Zhang, H. Zeng, J. Yang, Z. Yin, Q. Sun, and W. Li, “Novel RF-source-free reconfigurable microwave photonic radar,” Opt. Express 28(9), 13650–13661 (2020). [CrossRef]

9. P. Ghelfi, F. Laghezza, F. Scotti, D. Onori, and A. Bogoni, “Photonics of radars operating on multiple coherent bands,” J. Lightwave Technol. 34(2), 500–507 (2016). [CrossRef]

10. J. Cao, R. Li, J. Yang, Z. Mo, J. Dong, X. Zhang, W. Jiang, and W. Li, “Photonic deramp receiver for dual-band LFM-CW radar,” J. Lightwave Technol. 37(10), 2403–2408 (2019). [CrossRef]

11. L. Lembo, S. Maresca, G. Serafino, F. Scotti, F. Amato, P. Ghelfi, and A. Bogoni, “In-field demonstration of a photonic coherent MIMO distributed radar network,” in Proceeding of IEEE Radar Conf., (IEEE, 2019), pp.1–6.

12. F. Berland, T. Fromenteze, D. Boudsoque, P. D. Bin, H. H. Elwan, C. Aupetit-Berthelemot, and C. Decroze, “Microwave photonic MIMO radar for short-range 3D imaging,” IEEE Access 8, 107326–107334 (2020). [CrossRef]

13. F. Zhang, B. Gao, and S. Pan, “Photonics-based MIMO radar with high-resolution and fast detection capability,” Opt. Express 26(13), 17529–17540 (2018). [CrossRef]

14. M. A. Richards, Fundamentals of Radar Signal Processing2nd (McGraw-Hill Education, 2014).

15. T. Long, Y. Wang, and T. Zeng, “Signal-to-noise ratio in stretch processing,” Electron. Lett. 46(10), 720–721 (2010). [CrossRef]

16. W. J. Caputi, “Stretch: a time-transformation technique,” IEEE Trans. Aerosp. Electron. Syst. AES-7(2), 269–278 (1971). [CrossRef]

17. A. Wang, J. Wo, X. Luo, Y. Wang, W. Cong, P. Du, J. Zhang, B. Zhao, J. Zhang, Y. Zhu, J. Lan, and L. Yu, “Ka-band microwave photonic ultra-wideband imaging radar for capturing quantitative target information,” Opt. Express 26(16), 20708–20717 (2018). [CrossRef]

18. X. Luo, A. Wang, J. Wo, Y. Wang, J. Fu, Y. Zhu, J. Zhang, W. Cong, R. Liu, H. Yang, and L. Yu, “Microwave photonic video imaging radar with widely tunable bandwidth for monitoring diverse airspace targets,” Opt. Commun. 451, 296–300 (2019). [CrossRef]

19. X. Xiao, S. Li, S. Peng, L. Xing, X. Xue, X. Zheng, and B. Zhou, “A large-range autofocus microwave photonic radar based on adaptive spatial filtering along the range direction,” Opt. Commun. 477, 126354 (2020). [CrossRef]

20. J. Yang, R. Li, Z. Mo, J. Dong, and W. Li, “Channelized photonic-assisted deramp receiver with an extended detection distance along the range direction for LFM-CW radars,” Opt. Express 28(5), 7576–7584 (2020). [CrossRef]

21. N. Gebert, G. Krieger, and A. Moreira, “Digital beamforming for HRWS-SAR imaging: system design, performance and optimization strategies,” in Proceeding of International Symposium on Geoscience and Remote Sensing, (IEEE, 2006), pp.1836–1839.

22. Y. Okada, S. Nakamura, K. Iribe, Y. Yokota, M. Tsuji, M. Tsuchida, K. Hariu, Y. Kankaku, S. Suzuki, Y. Osawa, and M. Shimada, “System design of wide swath, high resolution, full polarimietoric L-band SAR onboard ALOS-2,” in Proceeding of International Geoscience and Remote Sensing Symposium - IGARSS, (IEEE, 2013), pp.2408–2411.

23. I. Sikaneta, C. H. Gierull, and D. Cerutti-Maori, “Optimum Signal Processing for Multichannel SAR: With Application to High-Resolution Wide-Swath Imaging,” IEEE Trans. Geosci. Remote Sensing 52(10), 6095–6109 (2014). [CrossRef]

24. W. G. Carrara, R. S. Goodman, and R. M. Majewski, Spotlight Synthetic Aperture Radar: Signal Processing Algorithms, (Artech House, 1995).

25. I. G. Cumming and F. H. Wong, Digital Processing of Synthetic Aperture Radar Data: Algorithms and Implementation, (Artech House, 2004).

Sample	Simulated Echoes Setting		Measured		Sample	Simulated Echoes Setting		Measured
Sample	Delay (µs)	Distance (m)	Distance (m)	3 dB Width (cm)	Sample	Delay (µs)	Distance (m)	Distance (m)	3 dB Width (cm)
1	1.5	225.00	225.01	6.9	16	46.5	6975.00	6975.01	6.9
2	2.8	420.00	420.01	6.9	17	49.0	7350.00	7350.01	6.9
3	7.0	1050.00	1050.01	6.9	18	51.7	7755.00	7755.01	6.9
4	9.0	1350.00	1350.01	6.9	19	52.6	7890.00	7890.01	6.9
5	15.7	2355.00	2355.01	6.9	20	53.8	8070.00	8070.01	6.9
6	16.5	2475.00	2475.01	6.9	21	61.5	9225.00	9225.01	6.9
7	19.8	2970.00	2970.01	6.9	22	63.6	9540.00	9540.01	6.9
8	26.0	3900.00	3900.01	6.9	23	63.7	9555.00	9555.01	6.9
9	27.7	4155.00	4155.01	6.9	24	64.0	9600.00	9600.01	6.9
10	29.0	4350.00	4350.01	6.9	25	69.0	10350.00	10350.01	6.9
11	31.5	4725.00	4725.01	6.9	26	70.8	10620.00	10620.01	6.9
12	36.8	5520.00	5520.01	6.9	27	74.6	11190.00	11190.01	6.9
13	39.7	5955.00	5955.01	6.9	28	83.0	12450.00	12450.01	6.9
14	41.6	6240.00	6240.01	6.9	29	85.6	12840.00	12840.01	6.9
15	45.0	6750.00	6750.01	6.9	30	89.0	13350.00	13350.01	6.9

Microwave photonic de-chirp receiver for breaking the detection range swath limitation

Abstract

1. Introduction

2. Principle

3. Experiment and results

4. Conclusion

Funding

Disclosures

References

Cited By

Figures (12)

Tables (1)

Equations (9)

Optics Express