Open Access
Add to BibSonomy
Abstract
The 3∼5 µm mid-infrared (mid-IR) light has several exceptional benefits in the case of adverse atmospheric conditions compared to the 1.5 µm band, so it is a promising candidate for optical carriers for free-space communication (FSO) through atmospheric channels. However, the transmission capacity in the mid-IR band is constrained in the lower range due to the immaturity of its devices. In this work, to replicate the 1.5 µm band dense wavelength division multiplexing (DWDM) technology to the 3 µm band for high-capacity transmission, we demonstrate a 12-channel 150 Gbps FSO transmission in the 3 µm band based on our developed mid-IR transmitter and receiver modules. These modules enable wavelength conversion between the 1.5 µm and 3 µm bands based on the effect of difference-frequency generation (DFG). The mid-IR transmitter effectively generates up to 12 optical channels ranging from 3.5768 µm to 3.5885 µm with a power of 6.6 dBm, and each channel carries 12.5 Gbps binary phase shift keying (BPSK) modulated data. The mid-IR receiver regenerates the 1.5 µm band DWDM signal with a power of -32.1 dBm. Relevant results of regenerated signal demodulation have been collected in detail, including bit error ratio (BER), constellation diagram, and eye diagram. The power penalties of the 6th to 8th channels selected from the regenerated signal are lower than 2.2 dB compared with back-to-back (BTB) DWDM signal at a bit error ratio (BER) of 1E-6, and other channels can also achieve good transmission quality. It is expected to further push the data capacity to the terabit-per-second level by adding more 1.5 µm band laser sources and using wider-bandwidth chirped nonlinear crystals.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The transmission capacity of wireless interconnection has increased exponentially over the last few decades, due to the emergence of various real-time high-quality services. To provide global information access, 6 G wireless communication networks will expand connectivity from terrestrial areas to near-Earth space, deep space, and ocean regions [1–3], of which free space optical (FSO) communication is an indispensable part. To improve availability and data rates in the FSO communication system, many new spectral bands are being exploited for specific scenarios, such as mid-infrared (mid-IR) region (3∼5 µm), terahertz (THz), and millimeter-wave (mmWave) bands [4–7]. Among these spectral bands, mid-IR communication is expected to play a vital role in improving and enhancing the transmission performance of optical signals in atmospheric channels, such as low absorption loss [8], low waveform distortion, high penetration through bad weather, and small beam spread over atmospheric turbulence channels [9,10].
Up to now, many mid-IR communication studies on the FSO link have been carried out. There are two significantly different approaches in mid-IR signals generation and detection, one is the direct modulation (DM) and direct detection (DD) method, based on quantum-cascade lasers (QCLs) or inter-band cascade lasers (ICLs) and quantum-well infrared photodetectors (QWIPs), and the other is the nonlinear frequency conversion (NFC) method [11–13] based on the effect of difference-frequency generation (DFG).
For DM applications, its simple transmission configurations and commercial off-the-shelf devices are more favored on some specific occasions. In the 80s, results of a 100 Mbps optical communication experiment using a PbCdS diode laser emitting at 3.5 µm were presented [14], and the laser diode needs to operate in a cryogenic environment. With the gradual maturity of photonic chip design, growth, and packaging technology, the operation of QCLs at room temperature emerged due to the strengthening of wall-plug efficiency (WPE) [15]. Under an uncooled environment, a continuous-wave distributed-feedback (DFB) QCL modulated by analog signals with a carrier frequency of 30 MHz was realized [16]. Recently, with many efforts to reduce the influence of parasitic effects, the bandwidth of QCLs can be extended to the maximum multi-gigahertz level [17,18]. Benefiting from the improvements mentioned above, several latest studies combined advanced intensity modulation and digital signal processing (DSP) techniques to promote further the data rate and spectrum utilization of signals in the 3∼5 µm band. An encoded and shaped eight-level pulse amplitude modulation (PAM-8) signal transmission with a wavelength of 4.5 µm and a data rate of up to 6 Gbps was experimentally carried out by Pang et al. [19,20], which was the highest data rate available for unidirectional mid-IR transmission using the DM approach. With the improvement of the ICL performance [21], the DM method using high-bandwidth ICL is an important alternative for mid-IR FSO communication, which has achieved rapid development in terms of optical power and data rate. An ICL laser with an emission wavelength of 4 µm was modulated by non-return-to-zero (NRZ) and return-to-zero (RZ) formats by Spitz et al. [22], and the results showed that the highest data rates were 110 Mbps and 300 Mbps for NRZ and RZ formats, respectively. Moreover, a full interband cascade system for high-speed transmission at 4.18 µm was demonstrated by Didier et al. [23], which uses a radio-frequency optimized ICL and an interband cascade infrared photodetector to achieve 12 Gbps on-off keying (OOK) and 14 Gbps 4-level PAM-4 schemes, which were the highest data rate achieved by ICL-based system. For the long-wave infrared (LWIR) in a range of 8 to 12 µm, some studies on high data rate FSO transmission were also reported. A 10.6 µm FSO communication testbed employing directly-modulated QCL up to 1 Gbps using commercial off-the-shelf components was reported by Liu et al. [24]. A 2 Gbps transmission at room temperature with a QCL emitting near 8.1 µm was reported by Spitz et al. [25]. By using a unipolar quantum optoelectronic system at room temperature, composed of a QCL, a modulator, and a quantum cascade detector, free space optics data transmission with a bit rate in excess of 10 Gbps was demonstrated at 9 µm by Dely et al. [26]. An 11 Gbps LWIR FSO transmission demonstration with a 9.6 µm directly-modulated QCL and a fully passive quantum cascade detectors (QCDs) without any active cooling or bias voltage was reported by Joharifar et al. [27,28]. Moreover, a record bit rate of 30 Gbps for both OOK and PAM-4 modulation schemes for a 31-m propagation link at 9 µm was achieved by Didier et al. [29], which involved the external modulation for LMIR waves. However, applications relying on DM for mid-IR communication still have some tricky limitations in replicating the characteristics of the 1.5 µm band signals. For example, except for the external modulation, the current data rates based on the DM method are still quite a challenge for single-wavelength transmission exceeding ∼10 Gbps even with the promotion of high-order modulation, and the modulation formats are limited to several intensity formats such as OOK, PAM, and pulse position modulation (PPM). Therefore, NFC is another promising technical route supporting high data rates, arbitrary modulation formats, and massive wavelength transmission capabilities.
For NFC applications, data-format-independent all-optical transmitter and receiver modules operated in the 3.8 µm region have been developed by Buchter et al. [30], which exploits DFG [31] in precision-fabricated Ti-indiffused periodically poled LiNbO3 waveguides (Ti:PPLN) [32], and has successfully demonstrated a short-distance transmission modulated by phase transition with a data rate of 2.488 Gbps. Moreover, low signal-to-noise ratio (SNR) penalty on mid-IR signals generation and inverse conversion based on bulk PPLN crystals were experimentally accomplished by Hao et al. [33], which obtained 10 mW average power at 3.6 µm with less than 10 dB SNR loss. Recently, a total spectral bandwidth of 1.6 nm carrying three wavelengths in the 1.5 µm band was converted to the 3.8 µm band, and then reconverted to the 1.5 µm band after propagation through a 0.5 m FSO link, which has brought a breakthrough in transmission capacity beyond 100 Gbps based on the NFC effect [34]. However, the reported number of wavelengths is still minimal, and more than 95% of the International Telecommunication Union-Telecommunication Standardization Bureau (ITU-T) grid wavelengths in the 1.5 µm band are not yet utilized for wavelength conversion, so there remains much room for improvement to realize a large-scale mid-IR dense wavelength division multiplexing (DWDM) transmission system.
In this paper, we have achieved a 150 Gbps mid-IR DWDM FSO transmission over a 5 m long free-space link in the 3 µm band, based on our previous research on 10 Gbps single-wavelength transmission at 3.6 µm [35,36]. Twelve optical channels with a frequency grid of 25 GHz in the 1.5 µm band are modulated by 12.5 Gbps binary phase-shift keying (BPSK) signals, and then converted to mid-IR band ranging from 3.5768 µm to 3.5885 µm, by using DFG effect in a MgO-doped PPLN (MgO: PPLN) crystal. The obtained power of the mid-IR DWDM signal is 6.6 dBm under an optimum temperature of 65.6 °C at the mid-IR transmitter. The mid-IR receiver converts the 12-channel mid-IR DWDM signals back to the 1.5 µm band, and the power of the regenerated 1.5 µm band signal is -32.1 dBm, which includes a fiber coupling loss of 4.4 dB. The power penalties of the selected 6th to 8th channels are all less than 2.2 dB, compared with back-to-back (BTB) DWDM signal at a bit error ratio (BER) of 1E-6. To the best of our knowledge, this is the closest proof of concept for large-scale mid-IR DWDM transmission with a telecom band as a reference.
2. Experiment setup
Figure 1 shows our proposed mid-IR DWDM transmission system over a 5 m FSO link. The experimental setup mainly includes a 1.5 µm band DWDM signal transmitter, a mid-IR DWDM signal transmitter, a mid-IR DWDM signal receiver, and a 1.5 µm band DWDM signal receiver.
2.1 1.5 µm band DWDM signal generation
At the 1.5 µm band DWDM signal transmitter, a total of 12 narrow linewidth (∼20 kHz linewidth) integrated tunable laser assembly (ITLA) modules are used to generate continuous-wave light with wavelengths that match the ITU-T grid. These light waves are coupled into a polarization-maintaining (PM) 16-port dense wavelength division multiplexing (DWDM) module with a frequency spacing of 25 GHz. The coupled DWDM signal is then fed into a Mach-Zehnder modulator (MZM) with a 3-dB bandwidth of 20 GHz, twelve channels are modulated by the same electrical signal with a data rate of 12.5 Gbps. The electrical signal is output from an arbitrary waveform generator (AWG, Tektronix AWG70002A) and amplified to a peak-to-peak voltage of 8.2 V (22.3 dBm) via a broadband electrical diver. Moreover, an automatic bias control (ABC) module locks the bias point of the MZM at the null point for long-term stable BPSK modulation.
2.2 Mid-IR DWDM signal transmitter
To produce a significant DFG process, the signal and pump lights need to be amplified to the watt level. As shown in Fig. 1, two stages of amplification provide a power boost for 1.5 µm band DWDM signal light and 1.083 µm pump light, respectively. The 1.5 µm band signal can be amplified up to 34.2 dBm by sequentially passing through an Er-doped fiber amplifier (EDFA) and a double-cladding Er/Yb-doped fiber amplifier (DC-EYDFA) with fully PM components. A 10 kHz linewidth Yb-doped fiber laser is used as a pumping seed laser, which is divided into two paths through a 50:50 beam splitter to provide pump light for the mid-IR DWDM signal transmitter and receiver, respectively. By configuring a two-stage amplification operation similar to the 1.5 µm band, the 1.083 µm pump light can be amplified up to 37 dBm by cascading a PM Yb-doped fiber amplifier (YDFA) and a DC-PM-YDFA. Two fiber isolators with power handling higher than 5 W and isolation higher than 35 dB are placed in the middle of the two stages of amplification to suppress backward amplified spontaneous emission (ASE) noise.
A specially customized 2 × 2 port high-power WDM is employed to couple the amplified 1.5 µm band DWDM signal and 1.083 µm pump light. The combined beam is collimated through an achromatic aspheric collimator Col1 with an adjustable focal length. The beam exits the collimator Col1 and vertically enters the center of a 5% MgO-doped PPLN crystal (HC Photonics) with a length of 50 mm. For fulfillment of the type-0 (eee) quasi-phase-matching (QPM) in the PPLN crystal, an achromatic half-wave plate (HWP) is placed in front of the PPLN crystal to adjust the polarization states of the two collimated beams. To remove the residual power at 1.083 and 1.5 µm, the mixed DFG light is reflected by a dichromatic mirror M1 and then filtered through a 0.5 µm bandwidth flat silicon window F1. In addition, the light reflected by mirror M1 is guided into an optical trap, which can greatly reduce the danger of the high-power beam, as well as at the receiver. Two achromatic doublets lenses L1 and L2, with the same focal length of 75 mm and an anti-reflection (AR) coating optimized for 3∼5 µm, are used to collimate the mid-IR light to a slight divergence angle. Two high reflection (> 99.5%) enhanced golden mirrors M2 and M3, are employed to build a 5-m long mid-IR FSO propagation path.
2.3 Mid-IR DWDM signal receiver
At the mid-IR DWDM signal receiver, after the beam spot of mid-IR DWDM light is focused by passing through the Si-Ge lens L2, the focused light enters the center of another PPLN crystal with the same parameters as the transmitter. Meanwhile, the amplified 1.083 µm pump light collimated by a fiber collimator Col2 is emitted into free space and then reflected by a high reflection (> 99.5%) mirror M4 and propagates to a flat-concave dielectric mirror M5. By precisely adjusting the positions of the M4 and M5 mirrors, the optical axes of mid-IR DWDM light and 1.083 µm pump light are aligned to a fine spatial coaxially to ensure the maximum overlap of the two beam spots over a long spatial distance for optimal conversion efficiency. A dichroic mirror M6 with a high reflection (> 97%) at 1.083 µm and a window filter F2 with a cut-on wavelength at 1.5 µm are used to remove the residual 1.083 µm and mid-IR DWDM light. In the end, a single mode fiber (SMF) is used to couple the regenerated signal for subsequent 1.5 µm band signal demodulation.
2.4 Regenerated 1.5 µm band DWDM signal demodulation
In the 1.5 µm band demodulation processing unit, the power of the regenerated 1.5 µm DWDM signal can be manually attenuated by a variable optical attenuator (VOA) for performance comparison. A low-noise EDFA with a noise figure of 3.6 dB is used to amplify the DWDM signal, and the amplified signal is routed into a programmable optical bandpass filter (OBPF, Finisar Waveshaper 4000S) with a minimum filter bandwidth of 0.08 nm, which can extract any wavelength from the DWDM signal. The subsequent experiment settings are designed to perform single-channel BPSK demodulation, constellation observation, and BER measurement. Here, a typical BPSK intradyne coherent detection module is applied, in which a tunable narrow-linewidth local oscillator (LO) laser (Teraxion, PS-TNL), a 2 × 4 port 90° optical hybrid (Kylia, COH24), and two balanced detectors (BPD, Finisar BPDV2120R) are used to realize phase signal to intensity signal conversion in the optical domain, and the intensity electrical signal is sampled by a 4-channel digital sampling oscilloscope (DSO, Tektronix DPO72304X) with a sample rate of 25 GSa/s. Finally, the original data is recovered by the offline digital signal processing (DSP) algorithms.
3. Results and discussion
3.1 DWDM signal generation in the 1.5 µm band
For the 1.5 µm band DWDM signal generation, we set up a group of ITLA lasers to generate 12 channels ranging from 1551.12 to 1553.33 nm, which meets the ITU-T standard grid for 25 GHz channel spacing. The modulation format is initially set to BPSK with a data rate of 12.5 Gbps, and a pseudo-random bit sequence (PRBS) with a length of 215-1 is repeatedly sent as the original input data. The optical spectrum of the modulated DWDM signal is measured after the MZM at the transmitter and shown in Fig. 2(a). The measured output power of the modulated 12-ch DWDM signal is 1.4 dBm. The optical signal-to-noise ratio (OSNR) and 20-dB optical spectral width of the WDM signal are 54.5 dB and 2.5 nm respectively, measured by an optical spectrum analyzer (OSA, YOKOGAWA AQ6370D). Figure 2(b) is the measured eye diagram trace of the single-channel BPSK signal.
Fig. 2. Characteristics of the generated 12-ch 1.5 µm band DWDM signal. (a) The measured optical spectrum with an OSNR of 54.5 dB in the 1.5 µm band. (b) Eye diagram trace of single-channel 12.5 Gbps BPSK signal.
3.2 DWDM signal conversion to mid-IR band
To implement the DFG process in the MgO: PPLN crystal, the 1.5 µm band DWDM signal is amplified through two-stage amplification, and the built-in gain-flattening filter in the DC-PM-EYDFA enables high gain flatness at these wavelengths. Under the condition of laser diodes (LDs) power of 42 dBm (16W), we obtain the maximum output powers of 34.2 dBm at 1.5 µm and 37.3 dBm at 1.083 µm, respectively. Figure 3(a) depicts the measured power flatness between the amplified 12-ch at different output power levels. The measurement results show a power flatness of less than 1.3 dB at all measured output power levels. Considering the safety of the two high-power fiber amplifiers, we adopt thermal management for two devices with the largest heat generation, the double-cladding Er/Yb-doped fiber and the 974 nm high-power pump LDs. The double-cladding Er/Yb-doped fiber is tightly attached to the surface of the aluminum alloy by aluminum foil tape, and the mounting surface of the 974 nm high-power pump LDs is coated with thermal grease and fixed on the aluminum alloy for large area heat dissipation. Figure 3(b) shows the collected output power of two amplifiers in 1.5 hours. Both amplifiers enter a thermal equilibrium state within 12 minutes, after which the power fluctuation of the two amplifiers is less than 1.4% in steady-state.
Fig. 3. Performance measurements of two high-power amplifiers. (a) The measured power flatness of 1.5 µm amplifier at different output power levels. (b) Output power fluctuation of the 1.5 µm and 1.083 µm amplifier in 1.5 hours;
The two amplified lights are coupled by the high-power WDM and incident vertically to the center of the PPLN crystal for the DFG process. DFG results from a three electromagnetic waves nonlinear interaction process that occurs within an optical material exposed to very intense lights, where the magnitude of the nonlinear response of the crystal is characterized by the χ(2). The nonlinear polarization generated in the DFG process is related to χ(2), which can be expressed as [12]:
where the P is the nonlinear polarization, ${\omega _p}$ is the angular frequency of pump light, ${\omega _s}$ is the angular frequency of signal light, ${\omega _i}$ is the angular frequency of idler light, ${E_p}$ is the electrical field of pump light, ${E_s}$ is the electrical field of signal light, and ${\varepsilon _0}$ is the permittivity of vacuum. For a continuous wave (CW) pump, the coupled differential equations for the field amplitudes in propagation direction can be expressed as [12]:To achieve a quasi-phase matching (QPM) status for high conversion efficiency, the PPLN crystal is placed in an oven composed of four-sided copper blocks. The oven temperature can be controlled from 0 0C to 150 0C with an accuracy of 0.1 0C. Under these pre-prepared operations, we obtain a mixed DFG light by injecting 34.2 dBm signal power and 37.3 dBm pump power into the PPLN crystal. To extract a pure mid-IR light, the residual 1.5 µm signal light and 1.083 µm pump light are filtered out from the spectral of the mixed light by cascading the mirrors M1 and F1, and the obtained power of the pure mid-IR light is 6.6 dBm measured by a thermal power meter (Thorlabs, S401C). Figure 4(a) measured after the achromatic doublets lens Si-Ge lens L1 illustrates the optical spectrum ranging from 1 to 5.5 µm measured by a Fourier transform OSA (Thorlabs, OSA205). As shown in Fig. 4(a), the mid-IR DWDM signal light can be clearly observed, and the measured OSNR is about 32.2 dB. In addition, on the left side of the optical spectrum, it can be seen that the vast majority of the pump and signal light power is significantly suppressed, and the calculated optical power ratios relative to the mid-IR DWDM signal light are -26.3 dB and -28.4 dB, respectively. Furthermore, a light in the range 1.788 to 1.794 um is generated by frequency doubling of the mid-IR DWDM signal light in the PPLN crystal. Figure 4(b) depicts the zoomed-in detail view of the optical spectrum of the 12-ch mid-IR DWDM signal, where the measured leftmost and rightmost center wavelengths are 3.5768 µm and 3.5885 µm, respectively. In particular, when the oven temperature is adjusted to 65.6 0C, the optimum power flatness is achieved with a maximum power difference of 3.9 dB for 12 channels, which originates from two aspects: one is due to the gain difference of 1.5 µm band amplifier for 12 channels at around 1 dB, and the other is the different phase mismatch factor $\Delta {\beta _{DFG}}$ for each channel at a specific temperature, resulting in different conversion efficiencies. The measured 20-dB optical spectral width of the 12-ch mid-IR DWDM signal is 12.9 nm.
Fig. 4. (a) Optical spectrum of the DFG light filtered by mirrors M1 and F1. (b) Zoomed-in optical spectrum of the 12-ch mid-IR DWDM signal.
Additionally, due to the lack of a charge-coupled device (CCD) camera in the mid-IR band, we use a mid-IR liquid crystal detector card (Thorlabs, VRC6S) to measure the parameters of the mid-IR beam. The area of detector card is a thin layer of thermochromic liquid crystals (TLC), which are temperature-sensitive organic chemicals with twisted helical molecular structures. The measured divergence angle and beam waist of the mid-IR beam are roughly 0.8 mrad and 0.6 mm, respectively. Since the conversion efficiency of the DFG process is closely related to the phase mismatch in a QPM structure. Furthermore, the refractive index of the three beams changes with the temperature of the PPLN crystal, which in turn changes the phase mismatch $\Delta {\beta _{DFG}}$. The relationship between phase mismatch $\Delta {\beta _{DFG}}$ and temperature T can be expressed as [37,38]:
Moreover, we select six odd-numbered channels marked as λ1∼λ11 and measure the power value of each channel from 53 to 78 °C at intervals of 0.2 °C, as shown in Fig. 6. For these wavelengths, the profile of the curve is very similar. We take the λ1 channel as an analysis example, when the temperature gradually rises from 50 to 61 °C, the normalized power of the λ1 channel increases from 0.06 to 0.95. Then, as the temperature continues to rise to 78 °C, the power starts to decrease until it reaches 0.14.
Fig. 5. Optical spectrum of 12-ch mid-IR DWDM signal measured at four temperatures, and the measured maximum power differences are (a) 9.2 dB at 69.6 °C; (b) 6.2 dB at 67.6 °C; (c) 5.7 dB at 63.6 °C; (d) 8.6 dB at 61.6 °C.
3.3 Regeneration of 1.5 µm DWDM signal
The divergence angle of the generated mid-IR light is narrowed after passing through the Si-Ge lens L1, and a mid-IR FSO link with a length of approximately 5 m is established through two golden mirrors M1 and M2. At the mid-IR DWDM signal receiver, the mid-IR beam spot is focused to the center of the PPLN crystal through the Si-Ge lens L2, and the focused spot size is less than 0.5 mm measured by the mid-IR liquid crystal detector card. Meanwhile, the optical axis of the 1.083 µm pump light is ensured to be coaxial with the mid-IR light through fine joint adjustment of mirrors M4 and M5, as shown in Fig. 1. By injecting mid-IR DWDM signal light and 1.083 µm pump light with powers of 2.2 dBm and 37 dBm, we observe regeneration of the 1.5 µm DWDM signal from the exit facet of the PPLN crystal. The dichroic mirror M6 reflects more than 99.5% of the residual pump light, and the remaining light continues to be filtered by mirror F2 to improve the OSNR of 1.5 µm DWDM signal, and the total out-of-band spectral rejection can exceed 30 dB. The collimator Col3 with an effective focal length of 11 mm, couples the 1.5 µm DWDM signal into the SMF, and the measured coupling power is -32.1 dBm, which includes a coupling loss of 4.4 dB.
Figure 7(a) measured after the VOA depicts the measured optical spectrum of the regenerated signal at an optimum temperature of 62.5 °C. The optical spectrum shown in Fig. 7(a) is similar to the original spectrum shown in Fig. 2(a), except for the power imbalance between the 12 channels. The center wavelength of the measured 12 channels deviates from the original wavelength by less than 0.02 nm, which may be limited by the maximum OSA resolution of 0.02 nm. In addition, the maximum power difference occurs between the 1st and 6th channel, and the calculated result is 8.1 dB. The main reason for the larger power difference is the limited gain bandwidth of the single-grating period PPLN crystal used in the experiment. If an apodized chirped PPLN crystal is used instead, it is expected to obtain tens of times the current gain bandwidth and a flatter gain curve [39,40]. Figure 7(b) depicts two curves of the performance of the regenerated signal affected by temperature, the black curve is the measured regenerated signal power as a function of the temperature, and the red curve is the calculated maximum power difference between 12 channels at different temperatures. As depicted in Fig. 7(b), when the temperature of PPLN crystal is heated up from 55 to 63 °C at a rate of 0.1 °C/min, the power of the regenerated signal increases from -36.4 to -32.4 dBm, while the maximum power difference decreases from 14.1 to 8.4 dB. As the temperature continues to increase to 69 °C, the regenerated signal power begins to decrease monotonically and finally stabilizes at -37.1 dBm, while the value of maximum power difference keeps growing and reaches 16.2 dB.
Fig. 7. (a) Measured optical spectrum of the regenerated 1.5 µm DWDM signal. (b) Regenerated signal power and maximum power difference at different temperatures.
After establishing the mid-IR FSO link, we compare the conversion efficiency. At the mid-IR DWDM signal transmitter, the conversion efficiency from the 1.5 µm DWDM signal light to the mid-IR DWDM signal light is -27.6 dB (0.17%). At the mid-IR WDM signal receiver, the conversion efficiency from the mid-IR DWDM signal light to the 1.5 µm DWDM signal light is -29.9 dB (0.1%). Moreover, considering the coupling loss of the SMF at the receiver, the total fiber-to-fiber loss of the mid-IR transmission system is 61.9 dB.
3.4 Demodulation of 1.5 µm DWDM signal
As shown in Fig. 1, the weak regenerated signal is amplified to a fixed power of 10 dBm by low noise EDFA operating in automatic power control (APC) mode. The 3 dB filter bandwidth of the programmable OBFP is configured to 0.16 nm (∼20 GHz) with a rectangular filtering shape, and the wavelength can be set arbitrarily. Here, we extract the 6th channel (1552.125 nm) for demodulation. Figure 8(a) measured after the OBPF at the receiver shows the optical spectrum of the 6th channel with a clear profile. The power of the LO laser is always kept at 12 dBm, and the extracted single-channel signal is fed into the BPSK intradyne coherent detection module, whose detailed structure is described in the previous subsection 2.4. The final offline processing part is used to recover the original data, mainly including the serial execution of timing recovery, frequency offset compensation, carrier phase recovery, and symbol decoding processes. By manually rotating the attenuation knob of the VOA device, the demodulation performance under different input powers can be measured. Figures 8(b)–8(d) are the recovery of two phase-states constellation diagrams of the 6th channel, which correspond to powers at the input of the EDFA of -33, -35, and -37 dBm, respectively. From the distribution of constellation points, it can be seen that as the power decreases, the constellation points start to spread around randomly. The root mean square (RMS) error vector magnitude (EVM) calculated from the recovered binary data corresponding to the three constellation diagrams are 15.4%, 18.7%, and 24.8%, respectively. Additionally, the long-term cumulative eye diagrams corresponding to the three received power levels are also shown in Figs. 8(e)–8(g), when the power is reduced, the additional fluctuation noise leads to a wider width of the traces of bits ‘1’ and ‘0’ in the eye diagram, causing the eye diagram tend to close.
Fig. 8. Demodulation of the regenerated 1.5 µm DWDM signal. (a) Optical spectrum of the 6th channel extracted by the narrow bandwidth OBPF. (b)-(d) Measured constellation diagrams at received power of (b) -33 dBm; (c) -35 dBm; (d) -37 dBm. (e)-(g) Measured eye diagrams at received power of (e) -33 dBm; (f) -35 dBm; (g) -37 dBm.
The BERs of the regenerated signal are measured to evaluate the transmission performance of the mid-IR FSO system. Figure 9 presents the BERs versus the received power of the selected three channels. Here, the 6th, 7th, and 8th channels with a measured wavelength of 1552.125, 1552.321, and 1552.518 nm are chosen for the BER measurement, labeled as C6, C7, and C8 in Fig. 9, respectively, and the same three channels of BTB DWDM signal for comparison. As shown in Fig. 9, for mid-IR DWDM transmission, the required powers for the 6th, 7th, and 8th channels are -40.4, -40.2, and -40.8 dBm at BER of 1E-6, respectively, while the required powers for the same three channels of BTB DWDM transmission are -42, -42.3 and -42.6 dBm, respectively. Hence, the power penalties of these three channels are 1.6, 2.1, and 1.8 dB, respectively. The power penalty could be caused by ASE noise and waveform distortion during wavelength conversion. In addition, the BERs of the two edge channels, the 1st and 12th channels, are also evaluated. The measured center wavelength of the 1st and 12th channels are 1551.118 nm and 1553.321 nm, respectively. From the regenerated optical spectrum shown in Fig. 7, it can be observed that the optical power of the 1st and 12th channels are 8.1 dB and 7.9 dB lower than the 6th channel, respectively. Therefore, the calculated single-channel powers of the 1st and 12th channels are -44.8 dBm and -44.6 dBm respectively, which correspond to the BERs are 5.2E-4 and 3.4E-4, respectively. Due to the BERs of two edge channels are above the enhanced forward error correction (EFEC) limit of 2E-3, so the quality of communication can still be at a good level. Moreover, Fig. 9 depicts the measured BERs for the 1st and 12th channels (labeled as C1 and C12) in the power range from -45 to -49 dBm by adjusting the attenuation of the VOA.
Fig. 9. Measured BERs versus the received power of the 1st, 6th, 7th, 8th, and 12th channels, and the same six channels of BTB WDM signal for comparison.
In general, this work explores the feasibility of large-scale DWDM transmission in the 3∼5 µm band. It is worth noting that the PPLN crystals are single-grating period crystals in the experiment, which have a more limited gain bandwidth. However, nonlinear crystals for wider conversion bandwidth have been reported, such as using chirped nonlinear crystals or dispersion-engineering thin films [41], which can theoretically achieve 1.5 µm band to mid-IR band conversion with a bandwidth of more than 100 nm for terabit level mid-IR transmission.
4. Conclusion
In conclusion, we have established a 12-ch mid-IR FSO link with a total capacity of 150 Gbps, by successfully utilizing wavelength conversion between 1.5 and 3 µm bands based on our developed mid-IR transmitter and receiver. The power of the generated 12-ch mid-IR DWDM signal is 6.6 dBm under an optimum temperature of 65.6 °C at the mid-IR transmitter. The power of the regenerated signal is -32.1 dBm, which includes a fiber coupling loss of 4.4 dB. The power penalties of the 6th to 8th channels selected from the regenerated signal are all less than 2.2 dB, compared with back-to-back (BTB) DWDM signal at a BER of 1E-6. Since the PPLN crystals used at the transmitter and receiver in the current experiment are single-grating period crystals, its 3-dB conversion bandwidth is several nanometers obtained from the experiment. This also reminds us that if we want to achieve more channels of wavelength conversion, a new structure of PPLN crystals needs to be developed, such as using chirped nonlinear crystals or dispersion-engineering thin films. Meanwhile, the conversion efficiency of the DFG process still needs to be further improved to support longer-distance transmission, such as using high-efficiency PPLN waveguides [42,43] to replace bulk PPLN crystals. We hope that by focusing on the above key difficulties, the DFG-based mid-IR FSO communication will be more practical and promising.
Funding
National Natural Science Foundation of China (62205261); Natural Science Basic Research Program of Shaanxi Province (2022JQ-709); State Key Laboratory of Transient Optics and Photonics (SKLST202106); Fundamental Research Funds for the Central Universities (XJS222801).
Disclosures
The authors declare no conflicts of interest.
Data availability
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.
References
1. A. Kumar, M. Gupta, P. Pitchappa, N. Wang, P. Szriftgiser, G. Ducournau, and R. Singh, “Phototunable chip-scale topological photonics: 160 Gbps waveguide and demultiplexer for THz 6 G communication,” Nat. Commun. 13(1), 5404 (2022). [CrossRef]
2. W. Shao, Y. Wang, S. Jia, Z. Xie, D. Gao, W. Wang, D. Zhang, P. Liao, B. E. Little, S. T. Chu, W. Zhao, W. Zhang, W. Wang, and X. Xie, “Terabit FSO communication based on a soliton microcomb,” Photon. Res. 10(12), 2802–2808 (2022). [CrossRef]
3. L. Xu and M. Rahmani, “Programmable structured surfaces can change the future of wireless communications,” Light-Sci. Appl. 11(1), 196 (2022). [CrossRef]
4. T. Nagatsuma, G. Ducournau, and C. Renaud, “Advances in terahertz communications accelerated by photonics,” Nat. Photonics 10(6), 371–379 (2016). [CrossRef]
5. S. Pirotta, N. L. Tran, A. Jollivet, G. Biasiol, P. Crozat, J. M. Manceau, A. Bousseksou, and R. Colombelli, “Fast amplitude modulation up to 1.5 GHz of mid-IR free-space beams at room-temperature,” Nat. Commun. 12(1), 799 (2021). [CrossRef]
6. X. Pang, O. Ozolins, S. Jia, L. Zhang, R. Schatz, A. Udalcovs, V. Bobrovs, H. Hu, T. Morioka, Y. Sun, J. Chen, S. Lourdudoss, L. K. Oxenløwe, S. Popov, and X. Yu, “Bridging the Terahertz Gap: Photonics-Assisted Free-Space Communications From the Submillimeter-Wave to the Mid-Infrared,” J. Lightwave Technol. 40(10), 3149–3162 (2022). [CrossRef]
7. H. Zhang, L. Zhang, S. Wang, Z. Lu, Z. Yang, S. Liu, M. Qiao, Y. He, X. Pang, X. Zhang, and X. Yu, “Tbit/s Multi-Dimensional Multiplexing THz-Over-Fiber for 6 G Wireless Communication,” J. Lightwave Technol. 39(18), 5783–5790 (2021). [CrossRef]
8. M. Grabner and V. Kvicera, “Multiple Scattering in Rain and Fog on Free-Space Optical Links,” J. Lightwave Technol. 32(3), 513–520 (2014). [CrossRef]
9. T. Plank, E. Leitgeb, P. Pezzei, and Z. Ghassemlooy, “Wavelength-selection for high data rate Free Space Optics (FSO) in next generation wireless communications,” 2012 17th European Conference on Networks and Optical Communications, Spain, 1–5 (2012).
10. A. Arnulf and J. Bricard, “Transmission by haze and fog in the spectral region 0.35 to 10 microns,” J Opt. Soc. Am. 47(6), 491–498 (1957). [CrossRef]
11. E. Sahin, B. Zabelich, O. Yakar, E. Nitiss, J. Liu, R. N. Wang, T. J. Kippenberg, and C. Brès, “Difference-frequency generation in optically poled silicon nitride waveguide,” Nanophotonics 10(7), 1923–1930 (2021). [CrossRef]
12. R. L. Sutherland, D. G. Mclean, and S. Kirkpatrick, Handbook of Nonlinear Optics (Marcel Dekker, 2003), Chap. 2.
13. L. Goldberg and W. K. Burns, “Wide acceptance bandwidth difference frequency generation in quas-phase-matched LiNbO3,” Appl. Phys. Lett. 67(20), 2910–2912 (1995). [CrossRef]
14. R. Denti, P. A. Andrekson, H. Ahlberg, and S. T. Eng, “3.5 µm optical communication experiment at 100 Mbits−1 using a PbCdS diode laser,” Opt. Quant. Electron. 21(1), 69–72 (1989). [CrossRef]
15. F. Wang, S. Slivken, D. H. Wu, and M. Razeghi, “Room temperature quantum cascade lasers with 22% wall plug efficiency in continuous-wave operation,” Opt. Express 28(12), 17532–17538 (2020). [CrossRef]
16. C. Liu, S. Zhai, J. Zhang, Y. Zhou, Z. Jia, F. Liu, and Z. Wang, “Free-space communication based on quantum cascade laser,” J. Semicond. 36(9), 094009 (2015). [CrossRef]
17. H. Li, W. J. Wan, Z. Y. Tan, Z. L. Fu, H. X. Wang, T. Zhou, Z. P. Li, C. Wang, X. G. Guo, and J. C. Cao, “6.2-GHz modulated terahertz light detection using fast terahertz quantum well photodetectors,” Sci. Rep. 7(1), 3452 (2017). [CrossRef]
18. A. Hangauer, G. Spinner, M. Nikodem, and G. Wysocki, “High frequency modulation capabilities and quasi single-sideband emission from a quantum cascade laser,” Opt. Express 22(19), 23439 (2014). [CrossRef]
19. X. Pang, O. Ozolins, R. Schatz, J. Storck, A. Udalcovs, J. R. Navarro, A. Kakkar, G. Maisons, M. Carras, G. Jacobsen, S. Popov, and S. Lourdudoss, “Gigabit free-space multi-level signal transmission with a mid-infrared quantum cascade laser operating at room temperature,” Opt. Lett. 42(18), 3646–3649 (2017). [CrossRef]
20. X. Pang, R. Schatz, M. Joharifar, A. Udalcovs, V. Bobrovs, L. Zhang, X. Yu, Y. Sun, G. Maisons, M. Carras, S. Popov, S. Lourdudoss, and O. Ozolins, “Direct Modulation and Free-Space Transmissions of up to 6 Gbps Multilevel Signals With a 4.65- μm Quantum Cascade Laser at Room Temperature,” J. Lightwave Technol. 40(8), 2370–2377 (2022). [CrossRef]
21. H. Lotfi, L. Li, L. Lei, H. Ye, S. M. S. Rassel, Y. Jiang, R. Q. Yang, T. D. Mishima, M. B. Santos, J. A. Gupta, and M. B. Johnson, “High-frequency operation of a mid-infrared interband cascade system at room temperature,” Appl. Phys. Lett. 108(20), 201101 (2016). [CrossRef]
22. O. Spitz, P. Didier, L. Durupt, D. A. Daz-Thomas, A. N. Baranov, L. Cerutti, and F. Grillot, “Free-space communication with directly modulated mid-infrared quantum cascade devices,” IEEE J. Sel. Top. Quantum Electron. 28(1), 1200109 (2021). [CrossRef]
23. P. Didier, H. Knötig, O. Spitz, L. Cerutti, A. Lardschneider, E. Awwad, D. D. Thomas, A. N. Baranov, R. Weih, J. Koeth, B. Schwarz, and F. Grillot, “Interband cascade technology for energy-efficient mid-infrared free-space communication,” Photonics Res. 11(4), 582–590 (2023). [CrossRef]
24. J. J. Liu, B. L. Stann, K. K. Klett, P. S. Cho, and P. M. Pellegrino, “Mid and long-wave infrared free-space optical communication,” Proc. SPIE 11133, 1113302 (2019). [CrossRef]
25. O. Spitz, K. Yang, A. G. Manca, P. Didier, J. Liu, and F. Grillot, “Multi-Gb/s free-space communication with energy-efficient room-temperature quantum cascade laser emitting at 8.1 µm,” 2021 IEEE Photonics Conference (IPC), Canada, 1–2 (2021).
26. H. Dely, T. Bonazzi, O. Spitz, E. Rodriguez, D. Gacemi, Y. Todorov, K. Pantzas, G. Beaudoin, I. Sagnes, L. Li, A. G. Davies, E. H. Linfield, F. Grillot, A. Vasanelli, and C. Sirtori, “10 Gbit s−1 Free Space Data Transmission at 9 µm Wavelength With Unipolar Quantum Optoelectronics,” Laser Photonics Rev. 16(2), 2100414 (2022). [CrossRef]
27. M. Joharifar, H. Dely, X. Pang, R. Schatz, D. Gacemi, T. Salgals, A. Udalcovs, Y. T. Sun, Y. Fan, L. Zhang, E. Rodriguez, S. Spolitis, V. Bobrovs, X. Yu, S. Lourdudoss, S. Popov, A. Vasanelli, O. Ozolins, and C. Sirtori, “High-Speed 9.6-µm Long-Wave Infrared Free-Space Transmission with a Directly-Modulated QCL and a Fully-Passive QCD,” J. Lightwave Technol. 41(4), 1087–1094 (2023). [CrossRef]
28. X. Pang, H. Dely, R. Schatz, D. Gacemi, M. Joharifar, T. Salgals, A. Udalcovs, Y. Sun, Y. Fan, L. Zhang, E. Rodriguez, S. Spolitis, V. Bobrovs, X. Yu, S. Lourdudoss, S. Popov, O. Ozolins, A. Vasanelli, and C. Sirtori, “Gb/s LWIR FSO Transmission at 9.6 µm using a Directly-Modulated Quantum Cascade Laser and an Uncooled Quantum Cascade Detector,” 2022 Optical Fiber Communications Conference (OFC). IEEE, 1–3 (2022).
29. P. Didier, H. Dely, T. Bonazzi, O. Spitz, E. Awwad, E. Rodriguez, A. Vasanelli, C. Sirtori, and F. Grillot, “High-capacity free-space optical link in the midinfrared thermal atmospheric windows using unipolar quantum devices,” Adv. Photonics 4(05), 056004 (2022). [CrossRef]
30. K. F. Buchter, H. Herrmann, C. Langrock, M. M. Fejer, and W. Sohler, “All-optical Ti:PPLN wavelength conversion modules for free-space optical transmission links in the mid-infrared,” Opt. Lett. 34(4), 470–472 (2009). [CrossRef]
31. C. Erny, K. Moutzouris, J. Biegert, D. Kuhlke, F. Adler, A. Leitenstorfer, and U. Keller, “Mid-infrared difference-frequency generation of ultrashort pulses tunable between 3.2 and 4.8 µm from a compact fiber source,” Opt. Lett. 32(9), 1138–1140 (2007). [CrossRef]
32. M. Asobe, O. Tadanaga, H. Miyazawa, Y. Nishida, and H. Suzuki, “Multiple quasi-phase-matched LiNbO3 wavelength converter with a continuously phase-modulated domain structure,” Opt. Lett. 28(7), 558–560 (2003). [CrossRef]
33. Q. Hao, G. S. Zhu, S. Yang, K. W. Yang, T. Duan, X. P. Xie, K. Huang, and H. P. Zeng, “Mid-infrared transmitter and receiver modules for free-space optical communication,” Appl. Opt. 56(8), 2260–2264 (2017). [CrossRef]
34. K. Zou, K. Pang, H. Song, J. Fan, Z. Zhao, H. Song, R. Zhang, H. Zhou, A. Minoofar, C. Liu, X. Su, N. Hu, A. M. Clung, M. Torfeh, A. Arbabi, M. Tur, and A. E. Willner, “High-capacity free-space optical communications using wavelength- and mode-division-multiplexing in the mid-infrared region,” Nat. Commun. 13(1), 7662 (2022). [CrossRef]
35. Y. Su, W. Wang, X. Hu, H. Hu, X. Huang, Y. Wang, J. Si, X. Xie, B. Han, H. Feng, Q. Han, G. Zhu, T. Duan, and W. Zhao, “10 Gbps DPSK transmission over free-space link in the mid-infrared,” Opt. Express 26(26), 34515–34528 (2018). [CrossRef]
36. W. Wang, Y. Zheng, X. Xie, Y. Su, X. Huang, T. Duan, and W. Zhao, “5 Gbaud QPSK coherent transmission in the mid-infrared,” Opt. Commun. 466, 125681 (2020). [CrossRef]
37. O. Gayer, Z. Sacks, E. Galun, and A. Arie, “Temperature and wavelength dependent refractive index equations for MgO-doped congruent and stoichiometric LiNbO3,” Appl. Phys. B. 91(2), 343–348 (2008). [CrossRef]
38. D. H. Jundt, “Temperature dependent sellmeier equation for the index of refraction, ne, in congruent lithium niobate,” Opt. Lett. 22(20), 1553–1555 (1997). [CrossRef]
39. X. Wu, L. Zhang, Z. Hao, R. Zhang, R. Ma, F. Bo, G. Zhang, and J. Xu, “Broadband second-harmonic generation in step-chirped periodically poled lithium niobate waveguides,” Opt. Lett. 47(7), 1574–1577 (2022). [CrossRef]
40. A. Bostani, M. Ahlawat, A. Tehranchi, R. Morandotti, and R. Kashyap, “Design, fabrication and characterization of a specially apodized chirped grating for reciprocal second harmonic generation,” Opt. Express 23(4), 5183–5189 (2015). [CrossRef]
41. J. Mishra, M. Jankowski, A. Y. Hwang, H. S. Stokowski, T. P. McKenna, C. Langrock, E. Ng, D. Heydari, H. Mabuchi, A. H. Safavi-Naeini, and M. M. Fejer, “Ultra-broadband mid-infrared generation in dispersion-engineered thin-film lithium niobate,” Opt. Express 30(18), 32752–32760 (2022). [CrossRef]
42. C. Cho, J. Lai, C. Hsu, Y. Huang, J. Jang, and M. Chou, “Power scaling of continuous-wave second harmonic generation in a MgO:PPLN ridge waveguide and the application to a compact wavelength conversion module,” Opt. Lett. 46(12), 2852–2855 (2021). [CrossRef]
43. C. Cho and Y. Chen, “Compactly efficient CW 3 to 4.5 µm wavelength tunable mid-infrared laser in optically pumped semiconductor laser with intracavity OPO,” IEEE J. Sel. Top. Quantum Electron. 28(1: Semiconductor Lasers), 1–6 (2022). [CrossRef]
