Open AccessCommunication

Generation of Wideband Signals Based on Continuous-Time Photonic Compression

Zhen Zhou

Yukang Zhang

and

Hao Chi

School of Communication Engineering, Hangzhou Dianzi University, Hangzhou 310018, China

Author to whom correspondence should be addressed.

Photonics 2024, 11(11), 1019; https://doi.org/10.3390/photonics11111019

Submission received: 27 September 2024 / Revised: 17 October 2024 / Accepted: 25 October 2024 / Published: 29 October 2024

(This article belongs to the Special Issue New Perspectives in Microwave Photonics)

Download

Browse Figures

Figure 1
A single-channel photonic time compression system. MLL, mode-locked laser. EOM, electro-optical modulator. DE, dispersive element. "> Figure 2
The continuous-time photonic compression (CTPC) system with multiple channels. MLL, mode-locked laser. OC, optical coupler. EOM, electro-optical modulator. DAC, digital-to-analog converter. DL, optical delay line. "> Figure 3
Schematic illustration of the operation principle of a three-channel continuous-time photonic compression (CTPC) system. "> Figure 4
The measured power transfer function of the experimental CTPC system and the theoretically predicted one. "> Figure 5
The modulated pulses prior to DE 2 (a) and the compressed pulses after DE 2 (b) of channel 1. "> Figure 6
The recorded combined waveform (a) and the waveforms after bandpass filtering with a passband of 250 MHz (b) and 150 MHz (c), respectively. "> Figure 7
The spectra of the pulses from two channels (a) and the pulse after combining (b). "> Figure 8
Simulation results of a four-channel CTPC system without predistortion: (a) the output waveform; (b) a zoom-in display on the connection area. "> Figure 9
Simulation results of a four-channel CTPC system with predistortion: (a) the output waveform; (b) a zoom-in display on the connection area given by the dotted frame in (a); (c) the spectrogram of the output chirped signal. ">

Versions Notes

Abstract

A detailed study on continuous-time photonic compression (CTPC) for generating wideband signals is presented in this paper. CTPC enables the conversion of parallel analog waveforms from multiple channels into a time-compressed continuous-time waveform with increased bandwidth. We demonstrate a CTPC system with a compression factor of two in a proof-of-concept experiment. Subsequently, the origin of the distortion in the generated signals is investigated, and we proposed a method based on bandpass filtering to remove the periodic dips observed in the generated waveforms. In addition, a predistortion method is proposed to eliminate the distortion caused by the non-ideal spectral property of the multichannel system. Further simulation results are presented to show the potential of the proposed approach.

Keywords:

wideband signals; continuous-time photonic compression; photonic time stretch; wavelength-to-time mapping

1. Introduction

High-speed analog signals are highly desired in many areas such as wideband optical and wireless communication systems, modern radars, instrument diagnostics, remote sensing, and microwave imaging systems [1]. Conventionally, wideband analog signals can be generated by employing direct digital synthesizers (DDSs) for producing intermediate frequency signals and analog mixers for up-converting to desired carrier frequencies. However, the signal bandwidth of the DDSs is fundamentally constrained by the bandwidth of their internal digital-to-analog converters (DACs), while mixers and other analog electronic circuits also have a limited operating frequency range.

Microwave photonics provides potential solutions to the generation of high-speed analog signals, which has the key advantages of wide instantaneous bandwidth and a large operating frequency range. In the past two decades, a number of photonic approaches have been proposed and demonstrated for generating wideband signals [2]. Typical approaches include line-by-line pulse shaping, spectral shaping and wavelength-to-time mapping, Fourier synthesis, and temporal pulse shaping [3,4,5,6]. There are also some approaches specially designed to generate chirped or phase-coded microwave signals for pulse compression radars or microwave imaging systems, such as the schemes based on interference between chirped optical pulses, Fourier-domain mode-locked optoelectronic oscillators, heterodyne beating between a fixed laser, and a fast frequency-sweeping laser [7,8,9]. While each of these approaches shows promise in wideband signal generation, they often face limitations such as high complexity, limited adaptability to dynamic signals, or significant latency. Photonic time stretch (PTS) and compression techniques offer distinct advantages, providing a practical approach to achieve both high bandwidth and low latency.

Photonic time compression (PTC) is a technique used for signal speedup, which has a similar structure but different dispersion configuration with the technique of PTS. A PTS system slows down input analog signals prior to an analog-to-digital converter (ADC), which can therefore effectively lower the requirement of bandwidth on ADCs [10]. In contrast, a PTC system after a DAC can be utilized to increase the bandwidth and carrier frequency of the generated analog signals [11]. Regarding PTS, a lot of research works have been conducted on different aspects, such as the realization of PTS ADC with an ultra-high sampling rate, the elimination of frequency-dependent RF power fading induced by chromatic dispersion and double sideband modulation, continuous-time operation of PTS, characterization and mitigation of nonlinear distortion, as well as others [12,13,14,15,16,17,18]. Unlike PTS, there have been relatively fewer reports about PTC so far. The concept of PTC was introduced in [19], and its applications in frequency multiplication and chirped signal generation were explored under the name of unbalanced temporal pulse shaping [20,21]. Recently, works utilizing PTC with finite time aperture for waveform generation have been reported [11,22]. In [23], we reported a design of an integrated transceiver based on PTS and PTC with finite time aperture, while a continuous-time photonic compression (CTPC) system based on phase modulation (PM) is briefly introduced and analyzed. However, these efforts have yet to provide a comprehensive report on a complete CTPC system. And one critical challenge in CTPC systems lies in the inability to post-process the generated analog waveforms within the digital domain, as the signals are processed after the DAC to compress the input signals in the time domain. Among these distortions, channel mismatch during signal stitching is a notable issue, which is significant for practical applications.

In this paper, we conduct a detailed study and experimental demonstration of the CTPC technique, exploring sources of distortions in the generated waveforms and investigating solutions to inter-channel distortions. A novel multichannel CTPC structure, combined with a well-suited dispersion configuration, is proposed to facilitate the transformation of parallel analog waveforms from multiple channels into a continuous-time waveform with multiplied bandwidth. The origin of the distortions in the generated waveforms is discussed. A method based on bandpass filtering is presented to mitigate distortions. We further investigate the distortions induced by channel mismatch and propose a digital predistortion method to alleviate this effect.

2. Principle of Operation

A single-channel PTC system is shown in Figure 1. The system comprises a mode-locked laser (MLL), a pair of dispersive elements (DEs) with opposite dispersion signs, and an electro-optical modulator (EOM) between the two DEs. The EOM is used to modulate an analog waveform onto to the stretched pulses, which are to be compressed in time. The optical pulse from the MLL is assumed to be transform-limited. The electric field of the pulse in the time domain is denoted by

a (t)

. Its spectrum is denoted by

A (ω)

, which has a spectral width of

Δ ω

. The dispersion values

Φ_{1}

and

Φ_{2}

(in ps²) of the two DEs satisfy

sgn \{Φ_{1}\} = - sgn \{Φ_{2}\}

and

|Φ_{1}| > |Φ_{2}|

. The dispersion value is defined as the first-order derivative of the group delay with respect to the angular frequency [24]. The total dispersion is denoted by

Δ Φ = Φ_{1} + Φ_{2}

. We assume that the far-field condition is satisfied, i.e.,

|Δ Φ| \cdot {|Δ ω|}^{2} / (2 π) ≫ 1

[10]. Note that this condition is easy to satisfy for sub-picosecond pulses. The temporal envelope of the pulse after dispersive propagation through

Φ_{1}

can be approximately modeled by

A (t / Φ_{1})

according to the theory of real-time Fourier transform [24]. The duration of the stretched pulse

A (t / Φ_{1})

can be calculated according to the spectral width

Δ ω

and the correspondence between time and frequency, which is

Δ t_{1} = |Φ_{1}| Δ ω

. Note that this pulse width is also the time aperture for signal modulation. The modulating analog signal is assumed to be sinusoidal with a frequency of

ω_{m}

. A small signal modulation is assumed. Similar to the theoretical results in PTS [10,25], the intensity signal detected by a photodiode, after dc removal, can be written as

I (t) = C \cdot A^{2} (\frac{t}{Δ Φ}) \cdot \cos (\frac{M Φ_{2}}{2} ω_{m}^{2}) \cdot \cos (M ω_{m} t),

(1)

where

C

is a time-independent constant, depending on the signal modulation index, dispersion values, and other parameters such as the optical power and the responsivity of the photodiode,

M = |Φ_{1} / Δ Φ| > 1

is the compression factor,

A^{2} (t / Δ Φ)

represents the envelope of the output signal with a time width of

Δ t_{2} = |Δ Φ| Δ ω

, the first cosine term means the frequency response of the PTC system, and the second cosine term represents the compressed sinusoidal signal with a multiplied frequency of

M ω_{m}

. By comparing the time width of the pulse after and before DE 2, it can be found that the pulse width is also compressed by a factor of

M

The single-channel PTC system above is capable of compressing an input waveform into a bandwidth-multiplied waveform with periodic time apertures. This period is decided by the repetition period of the applied MLL, and the width of the time aperture is equal to the pulse width after DE 1. By using multiple single-channel PTC systems and parallel analog signals, we may design a CTPC system for converting multiple analog signals into a continuous-time high-speed signal. However, this configuration would be quite complicated due to the involvement of numerous MLLs and DEs. Figure 2 shows the CTPC system with simplified architecture, in which an MLL and two DEs are shared by all channels. The optical pulse from the MLL becomes stretched in time and chirped after propagating through DE 1. The chirped optical pulses after DE 1 are split into multiple channels via an optical coupler (OC 1). Each channel is equipped with an EOM for signal modulation. The parallel analog signals from DACs are modulated on the chirped optical pulses in all channels through the EOMs. The modulated pulses are then combined by OC 2, which is followed by DE 2 for pulse compression. Finally, the photodetector (PD) performs O/E conversion. The electric signal then passes through a bandpass filter (BPF), yielding the desired RF output signal. By utilizing periodical short pulses from the MLL, input parallel analog signals can be compressed in time and combined into a continuous-time signal with multiplied bandwidth. Note that optical delay lines can be placed in the channels for delay adjustment and time alignment. In essence, this system optically realizes the parallel-to-serial conversion for analog signals with signal bandwidth multiplication. If the pulse repetition period is

T

and the compression factor is

M

, the time aperture seen by each EOM is

T

, and that of each pulse after DE 2 is

T / M

. The number of channels

N

should be equal to the compression factor

M

. The configuration of the dispersion values and spectral width

Δ ω

should match the repetition period and the compression factor, i.e.,

T = |Φ_{1}| Δ ω

and

M = |Φ_{1} / Δ Φ| = N

. In order to show the process of the generation of continuous-time signals more clearly, the sequence diagram of a three-channel CTPC system is given in Figure 3.

3. Experiment

Proof-of-concept experiments were implemented. We utilized two push–pull Mach–Zehnder modulators (MZMs, Avanex 8F042D04, Avanex Corp, Richardson, TX, USA) for signal modulation. A spool of a standard single-mode fiber, with a length of 40 km, and a dispersion compensating module (for compensating the accumulated dispersion of 20 km standard single-mode fiber) were employed as the DEs, with dispersion values of

Φ_{1} = - 846.7 {ps}^{2}

and

Φ_{2} = 424.6 {ps}^{2}

, respectively, corresponding to a compression factor of

M \approx 2

. An MLL (MenloSystems ELMO, Planegg, Germany) with a repetition rate of 100 MHz and a bandwidth over 40 nm in the 1550 nm waveband was employed as the pulsed source. A programmable optical filter (Finisar WS 4000A, Finisar, Menlo Park, CA, USA) with multiple output ports was used as the demultiplexer to truncate the spectrum to be within 1534.9–1550.22 nm and 1550.22–1564.78 nm, corresponding to two different output ports, respectively, such that the temporal duration of the stretched pulses in each channel is 10 ns, matching the repetition period of the MLL. A polarization controller was placed prior to each MZM to minimize the polarization dependent loss. And a tunable delay line and an optical attenuator were placed in one channel for time alignment and power equalization. Then, an RF signal from a signal source (R&S SMB100A, Mühlhausen, Germany), with a frequency that is an integer multiple of the MLL’s repetition rate, was applied to both of the EOMs, ensuring that the modulating signals on the optical pulses in the two channels were the same and could be connected end to end. This arrangement is equivalent to the EOMs being modulated by the interleaved segments from a sinusoidal signal. Two optical channels were combined by a coupler, which was then followed by DE 2. A PD (CONQUER KG-PT-10G-A-FA, CONQUER, Chongqing, China) with a bandwidth of 10 GHz was applied for O/E conversion.

Firstly, we measured the power transfer function of the system. The frequency of the applied RF signal was gradually increased from 1 GHz to 7 GHz with a step of 0.5 GHz, while the injected RF power into each MZM was fixed at 5 dBm. The RF power of the output signal was measured by a spectrum analyzer (R&S FSV30, Munich, Germany). The recorded normalized RF power is shown in Figure 4, where the horizontal axis denotes the frequency of the input signal. A theoretically predicted curve according to the power transfer function

H (ω_{m}) = \cos^{2} [Φ_{2} ω_{m}^{2} / (2 M)]

is also given, which is in good agreement with the measurement results. The 3 dB bandwidth of the experimental CTPC system is around 6.5 GHz.

Here, we compare the power transfer functions of the PTC and PTS. For comparison, the power transfer function of the PTC can be expressed as

H^{c} = \cos^{2} [(M^{c} - 1) Φ_{1} ω_{m}^{2} / 2]

, where

M^{c} > 1

denotes the compression factor. The power transfer function of the PTS can be written as

H^{s} = \cos^{2} [(1 - 1 / M^{s}) Φ_{1} ω_{m}^{2} / 2]

, where

M^{s} > 1

denotes the stretch factor. When a PTC and a PTS systems have the same compression/stretch factor, i.e.,

M^{c} = M^{s}

, and the same dispersion value

Φ_{1}

, the PTC system will suffer a more severe power penalty than the PTS, as

M^{c} - 1 > 1 - 1 / M^{s}

for

M^{c} = M^{s} > 1

. For the experimental PTC system, the first power null occurs at around 9.7 GHz, but for a PTS system with

M^{s} = 2

and the same

Φ_{1}

, the first power null occurs at around 13.7 GHz. If the compression/stretch factor is increased to 6, these values are 4.3 GHz for PTC and 10.6 GHz for PTS. In order to overcome this more severe power penalty in PTC induced by chromatic dispersion and double sideband modulation, the technique of single sideband (SSB) modulation is a potential solution, as in the case of PTS [13].

Next, we chose an RF signal with a frequency of 2 GHz and a power of 5 dBm to apply to the EOMs, which falls within the 3 dB bandwidth of the system. The modulated pulses prior to DE 2 of channel 1 are shown in Figure 5a, which were recorded by a real-time oscilloscope (R&S RPT084, Munich, Germany). It is seen that the pulse width is equal to the pulse repetition period, and the pulses are connected end to end. The compressed pulses of channel 1 after DE 2 are shown in Figure 5b, where the channel was measured individually with the other channel being turned off. Then, both channels were turned on and the combined signal was recorded, as shown in Figure 6a. It is shown there are periodic dips in the combined waveform, which are located in the connected areas between the pulses from different channels. This unwanted distortion was mainly caused by the spectral edge properties of the applied optical filters, such as falling edges and overlapping between adjacent channels. The transmission spectra of the two channels set within the programmable filter are shown in Figure 7a, and the combined spectrum is given in Figure 7b. Apparently, there is a dip in between the two channels, which well explains the dips in the combined waveforms according to the principle of frequency to time mapping. We utilized bandpass filtering to remove these dips in the compressed waveform. Figure 6b,c show the filtered waveforms using a bandpass filter with a passband of 250 MHz and 150 MHz, respectively. The results indicate that the unwanted dips can be thoroughly removed given that the filter passband is less than twice the repetition rate of the applied MLL. Therefore, distortion removal based on bandpass filtering is suitable for generating the narrowband signals with high carrier frequencies.

4. Predistortion Algorithm

Transition and overlapping in the channels’ spectral edges in a CTPC system is inevitable. This channel mismatch leads to periodic dips in the combined signals. The distortion removal method based on bandpass filtering discussed above can be applied in the generation of continuous-time signals with a bandwidth less than twice the repetition rate of the applied MLL, which typically falls within the tens of megahertz. However, it is not suitable for the generation of wideband continuous-time signals.

Fortunately, the channels’ spectral property in a CTPC system is static and can be measured in advance. According to the principle of frequency to time mapping, the shape of the optical spectrum determines the envelope of the time-domain signal after dispersive propagation. This characteristic can be effectively utilized in designing a predistortion algorithm aimed at correcting waveforms and eliminating unwanted distortions. Referring to Equation (1), the envelope of a compressed signal from a channel is as

A^{2} (t / Δ Φ)

, with a shape corresponding to the power spectrum

{|A (ω)|}^{2}

of the channel.

The modulated signal in its electrical field prior to DE 2 of a channel can be modeled as

E_{M} (t) = A (t / Δ Φ) \cdot [1 + m \cdot x (t)],

(2)

where

m

is the modulation index. The signal distortion is induced by the spectral edge property in

A (t / Φ_{1})

. In order to cancel the edge effect of the channel, an analog signal

x (t)

within a time duration injected into this channel can be predistorted as

x' (t) = \frac{x (t)}{A (t / Φ_{1})} + \frac{1 - A (t / Φ_{1})}{m \cdot A (t / Φ_{1})} .

(3)

Therefore, Equation (2) can be rewritten as

\begin{array}{l} E_{M} (t) = A (t / Φ_{1}) \cdot [1 + m \cdot x' (t)] \\ = 1 + m \cdot x (t) \end{array} .

(4)

It is seen clearly that the impact of the spectral shape, including ripples and falling edges in the passband of the channel, is avoided. According to Equation (1), the output signal can be rewritten as

I (t) = C \cdot x (M t)

, where we omitted the power fading term induced by double sideband modulation and chromatic dispersion. The output is a temporally compressed signal with a compression factor of

M

without any spectral-shape-induced envelope.

We performed computer simulation to test the above predistortion method in a four-channel CTPC system. The parameters of the system were set as follows. The total spectral width is 8 THz, which is the same for each channel. The dispersion values of two dispersive elements were

Φ_{1} = - 198 {ps}^{2}

and

Φ_{2} = 148.5 {ps}^{2}

, respectively, corresponding to a compression factor of

M = 4

. The repetition rate of the applied MLL is 100 MHz. The optical spectrum was shaped with a flat top and spectral edges as the experimentally recorded spectrum in Figure 7. The target signal was a periodic chirped signal with frequencies ranging from 4 GHz to 12 GHz and with a period of 25 ns. For this, time-interleaved segments from a relatively slow chirped waveform were prepared in advance for injection into the EOMs across the four channels. In the first simulation, input waveforms without predistortion were applied. The output waveform is depicted in Figure 8a. There are periodic dips in the connection area, as anticipated by theory and in accordance with the experimental results in Figure 6a. A zoom-in display on the connection area is shown in Figure 8b. We also conducted tests using various spectral shapes with different roll-off properties. It was found that the profile of the output waveform is almost always consistent with the combined spectral shape of the four channels. This confirms that the periodic dips in the generated waveforms are indeed caused by the spectral edge properties.

In the second simulation, we studied the performance of the presented predistortion method. The interleaved segments were predistorted according to the method shown in Equation (3) before being applied to the EOMs. The output signal in the time domain, after compression and combination, is shown in Figure 9a. A zoom-in view on the connection area is shown in Figure 9b. As expected, the unwanted periodic dips almost disappear, and the corrected waveform is relatively smooth, with only slight distortions, which verifies the effectiveness of the given predistortion method. The residual slight distortion at the connection areas can be attributed to the spectral edge extension effect. Due to this effect, pulse waveforms at the edges interfere with the neighboring pulses, and this spectral extension effect has not been taken into account in the predistortion algorithm. The time–frequency plot of the generated chirped waveform is given in Figure 9c, which shows a good linear frequency modulation in the expected spectral range of 4–12 GHz. The given results demonstrate that the presented predistortion method can be effectively applied in a CTPC system for generating wideband continuous-time signals.

5. Discussions

Compared to the continuous-time photonic compression system presented in [23], this study utilizes OCs to construct a multichannel structure instead of relying on spectral splitting methods. This approach enables full use of the optical spectrum, reducing the dispersion requirements typically imposed on the system, and effectively minimizes power loss within the system link. These improvements not only enhance system performance but also make the setup more adaptable for practical implementation, where managing optical spectrum resources is critical. In [15], a redundancy-based distortion compensation scheme was introduced for continuous-time photonic time-stretch (PTS) ADCs, where redundant modulation was applied to signal segments within the overlapping regions of continuous optical carriers to avoid inter-pulse mismatch distortion. While effective for TPS-ADCs, this technique is limited in scenarios for signal generation, where post-processing options are inherently constrained. Since signal generation in a CTPC setup occurs after the DACs, our proposed bandpass filtering and predistortion techniques address this gap by minimizing spectral edge effects at the analog stage, as confirmed by our experimental and simulation results.

Beyond the channel mismatch-related signal distortion previously discussed, other factors merit attention. Dispersion non-flatness and higher-order dispersion in dispersive elements may lead to significant variations in phase delay across different wavelengths, which can cause severe distortion in the generated waveforms. Furthermore, while SSB modulation can help address frequency-dependent power fading caused by dispersion, the output signal bandwidth remains limited by the operating range of the applied 90° hybrid, as it may not provide an accurate 90-degree phase shift at certain frequencies.

Additionally, phase mismatch among multiple input signals may occur due to possible phase discrepancies in the input signals generated by DACs. This challenge is particularly significant in multichannel CTPC applications, where precise timing and phase alignment across channels are critical for maintaining the coherency and stability of the output signal. Potential solutions include synchronizing the DACs to a common phase reference or implementing active phase correction circuits to compensate for these discrepancies. Optimizing these parameters could considerably broaden the application range of CTPC systems, enabling more complex waveform generation and advancing their utility in areas such as wideband radar systems, next-generation 5G communications, and electronic warfare.

6. Conclusions

In summary, we conducted an in-depth study on the technique of CTPC and proposed solutions to mitigate inter-channel distortions. A novel system design of CTPC was proposed, and a detailed analysis of the CTPC theory and characteristics has been presented. A proof-of-concept experiment of a CTPC system has been implemented. Meanwhile, we focused on the signal distortion issue, i.e., unwanted periodic dips in the generated waveforms. A method based on bandpass filtering was proposed to remove unwanted dips in the generated waveforms, which is suitable for the generation of narrowband signals. The origin of the signal distortions was discussed. We presented a digital predistortion method to solve the waveform distortion problem induced by the edge effect of the channels’ spectra. The given CTPC architecture, as well as the related distortion removal methods, provides a potential solution for generating continuous-time signals with ultra-high bandwidth.

Author Contributions

Conceptualization, H.C.; methodology, Z.Z. and H.C.; software, Z.Z.; validation, H.C., Y.Z. and Z.Z.; formal analysis, Z.Z. and H.C.; investigation, Z.Z. and H.C.; data curation, Y.Z. and Z.Z.; writing—original draft preparation, Y.Z. and Z.Z.; writing—review and editing Z.Z. and H.C.; supervision, H.C.; funding acquisition, H.C. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported in part by the National Natural Science Foundation of China, grant number 62375071.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

Skolnik, M. Role of radar in microwaves. IEEE Trans. Microw. Theory Tech. 2002, 50, 625–632. [Google Scholar] [CrossRef]
Chi, H.; Wang, C.; Yao, J. Photonic generation of wideband chirped microwave waveforms. IEEE J. Microw. 2021, 1, 787–803. [Google Scholar] [CrossRef]
Jiang, Z.; Leaird, D.E.; Weiner, A.M. Line-by-line pulse shaping control for optical arbitrary waveform generation. Opt. Exp. 2005, 13, 10431–10439. [Google Scholar] [CrossRef]
Chen, L.R. Photonic generation of chirped microwave and millimeter wave pulses based on optical spectral shaping and wavelength-to-time mapping in silicon photonics. Opt. Commun. 2016, 373, 70–81. [Google Scholar] [CrossRef]
Zhang, F.; Ge, X.; Pan, S. Triangular pulse generation using a dual-parallel Mach–Zehnder modulator driven by a single-frequency radio frequency signal. Opt. Lett. 2013, 38, 4491–4493. [Google Scholar] [CrossRef]
Saperstein, R.E.; Fainman, Y. Information processing with longitudinal spectral decomposition of ultrafast pulses. Appl. Opt. 2008, 47, A21–A31. [Google Scholar] [CrossRef]
Jolly, S.W.; Matlis, N.H.; Ahr, F.; Leroux, V.; Eichner, T.; Calendron, A.; Ishizuki, H.; Taira, T.; Kärtner, F.X.; Maier, A.R. Spectral phase control of interfering chirped pulses for high-energy narrowband terahertz generation. Nat. Commun. 2019, 10, 2591–2598. [Google Scholar] [CrossRef]
Hao, T.; Tang, J.; Li, W.; Zhu, N.; Li, M. Fourier domain mode locked optoelectronic oscillator based on the deamplification of stimulated Brillouin scattering. OSA Continuum 2018, 1, 408–415. [Google Scholar] [CrossRef]
Shi, J.W.; Kuo, F.M.; Chen, N.W.; Set, S.Y.; Huang, C.B.; Bowers, J.E. Photonic generation and wireless transmission of linearly/nonlinearly continuously tunable chirped millimeter-wave waveforms with high time-bandwidth product at W-band. IEEE Photon. J. 2012, 4, 215–223. [Google Scholar] [CrossRef]
Coppinger, F.; Bhushan, A.S.; Jalali, B. Photonic time stretch and its application to analog-to-digital conversion. IEEE Trans. Microw. Theory 1999, 47, 1309–1314. [Google Scholar] [CrossRef]
Dong, X.; Zhou, Y.; Kong, J.; Zhang, F.; Pan, S. High-speed digital-to-analog conversion by fiber-dispersion-based temporal compression. In Proceedings of the 2022 20th International Conference on Optical Communications and Networks (ICOCN), Shenzhen, China, 12–15 August 2022. [Google Scholar]
Zhang, Y.; Jin, R.; Peng, D.; Lyu, W.; Fu, Z.; Zhang, Z.; Zhang, S.; Li, H.; Liu, Y. Broadband transient waveform digitizer based on photonic time stretch. J. Lightw. Technol. 2021, 39, 2880–2887. [Google Scholar] [CrossRef]
Fuster, J.M.; Novak, D.; Nirmalathas, A.; Marti, J. Single-sideband modulation in photonic time-stretch analogue-to-digital conversion. Electron. Lett. 2001, 37, 67–68. [Google Scholar] [CrossRef]
Han, J.; Seo, B.; Han, Y.; Jalali, B.; Fetterman, H.R. Reduction of fiber chromatic dispersion effects in fiber-wireless and photonic time-stretching system using polymer modulators. J. Lightw. Technol. 2003, 21, 1504–1509. [Google Scholar]
Yang, S.; Wang, J.; Chi, H.; Yang, B. Distortion compensation in continuous-time photonic time-stretched ADC based on redundancy detection. Appl. Opt. 2021, 60, 1646–1652. [Google Scholar] [CrossRef] [PubMed]
Fard, A.M.; DeVore, P.T.; Solli, D.R.; Jalali, B. Impact of optical nonlinearity on performance of photonic time-stretch analog-to-digital converter. J. Lightw. Technol. 2011, 29, 2025–2030. [Google Scholar] [CrossRef]
Fard, A.M.; Gupta, S.; Jalali, B. Photonic time-stretch digitizer and its extension to real-time spectroscopy and imaging. Laser Photon. Rev. 2013, 7, 207–263. [Google Scholar] [CrossRef]
Qian, N.; Zou, W.; Zhang, S.; Chen, J. Signal-to-noise ratio improvement of photonic time-stretch coherent radar enabling high-sensitivity ultrabroad W-band operation. Opt. Lett. 2018, 43, 5869–5872. [Google Scholar] [CrossRef]
Coppinger, F.; Bhushan, A.S.; Jalali, B. Time reversal of broadband microwave signals. Electron. Lett. 1999, 35, 1230–1232. [Google Scholar] [CrossRef]
Wang, C.; Li, M.; Yao, J. Continuously tunable photonic microwave frequency multiplication by use of an unbalanced temporal pulse shaping system. IEEE Photon. Technol. Lett. 2010, 22, 1285–1287. [Google Scholar] [CrossRef]
Li, M.; Wang, C.; Li, W.; Yao, J. An unbalanced temporal pulse-shaping system for chirped microwave waveform generation. IEEE Trans. Microw. Theory 2010, 58, 2968–2975. [Google Scholar] [CrossRef]
Konishi, T.; Kaihori, Y.; Makino, M. Photonic-assisted arbitrary waveform generation for uplink applications in beyond 5G taking advantage of low frequency technology. J. Lightw. Technol. 2022, 40, 6608–6615. [Google Scholar] [CrossRef]
Zhang, Y.; Chi, H. An optical front-end for wideband transceivers based on photonic time compression and stretch. Photonics 2022, 9, 658. [Google Scholar] [CrossRef]
Muriel, M.A.; Azaña, J.; Carballar, A. Real-time Fourier transformer based on fiber gratings. Opt. Lett. 1999, 24, 1–3. [Google Scholar] [CrossRef]
Mei, Y.; Xu, B.; Chi, H.; Jin, T.; Zheng, S.; Jin, X.; Zhang, X. Harmonics analysis of the photonic time stretch system. Appl. Opt. 2016, 55, 7222–7228. [Google Scholar] [CrossRef]

Figure 1. A single-channel photonic time compression system. MLL, mode-locked laser. EOM, electro-optical modulator. DE, dispersive element.

Figure 2. The continuous-time photonic compression (CTPC) system with multiple channels. MLL, mode-locked laser. OC, optical coupler. EOM, electro-optical modulator. DAC, digital-to-analog converter. DL, optical delay line.

Figure 3. Schematic illustration of the operation principle of a three-channel continuous-time photonic compression (CTPC) system.

Figure 4. The measured power transfer function of the experimental CTPC system and the theoretically predicted one.

Figure 5. The modulated pulses prior to DE 2 (a) and the compressed pulses after DE 2 (b) of channel 1.

Figure 6. The recorded combined waveform (a) and the waveforms after bandpass filtering with a passband of 250 MHz (b) and 150 MHz (c), respectively.

Figure 7. The spectra of the pulses from two channels (a) and the pulse after combining (b).

Figure 8. Simulation results of a four-channel CTPC system without predistortion: (a) the output waveform; (b) a zoom-in display on the connection area.

Figure 9. Simulation results of a four-channel CTPC system with predistortion: (a) the output waveform; (b) a zoom-in display on the connection area given by the dotted frame in (a); (c) the spectrogram of the output chirped signal.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Zhou, Z.; Zhang, Y.; Chi, H. Generation of Wideband Signals Based on Continuous-Time Photonic Compression. Photonics 2024, 11, 1019. https://doi.org/10.3390/photonics11111019

AMA Style

Zhou Z, Zhang Y, Chi H. Generation of Wideband Signals Based on Continuous-Time Photonic Compression. Photonics. 2024; 11(11):1019. https://doi.org/10.3390/photonics11111019

Chicago/Turabian Style

Zhou, Zhen, Yukang Zhang, and Hao Chi. 2024. "Generation of Wideband Signals Based on Continuous-Time Photonic Compression" Photonics 11, no. 11: 1019. https://doi.org/10.3390/photonics11111019

APA Style

Zhou, Z., Zhang, Y., & Chi, H. (2024). Generation of Wideband Signals Based on Continuous-Time Photonic Compression. Photonics, 11(11), 1019. https://doi.org/10.3390/photonics11111019

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu