Open Access
Add to BibSonomy
Abstract
Transmission of sub-terahertz (sub-THz) signals over a fiber-free-space optical (FSO)-fifth-generation (5 G) new radio (NR) hybrid system is successfully realized. It is a promising system that utilizes multiple media of optical fiber, optical wireless, and 5 G NR wireless to achieve a 227.912-Gb/s record-high aggregate net bit rate. The system concurrently transmits a 59.813-Gb/s net bit rate in the 150-GHz sub-THz frequency, 74.766-Gb/s in the 250-GHz sub-THz frequency, and 93.333-Gb/s in the 325-GHz sub-THz frequency through the fiber-FSO-wireless convergence, including 25-km single-mode fiber, 100-m FSO, and 30-m/25-m/20-m sub-THz-wave transmissions. This system achieves sufficiently low bit error rates (< hard-decision forward error correction (FEC) threshold of 3.8 × 10−3 at 16 and 20 Gbaud symbol rates; < soft-decision FEC threshold of 2 × 10−2 at 28 Gbaud symbol rate) and clear and distinct constellation diagrams, meeting the demands of 5 G NR communications in the sub-THz band. The development of fiber-FSO-5 G NR hybrid system represents a substantial development in the field of advanced communications. It has the ability to enhance the way we communicate in the future.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
With the advent of new technologies such as fifth-generation (5 G) new radio (NR) and beyond, people are moving towards new applications that require higher data rates. With the increasing demand for higher data rates, researchers and engineers are exploring the potential of using sub-terahertz (sub-THz) frequency in 5 G NR communications to enable new applications [1–5]. 5 G NR sub-THz communications utilize carrier frequencies in 90-350 GHz range, which is higher than the carrier frequencies used in 5 G sub-6 GHz and 5 G NR millimeter-wave communications. The use of higher carrier frequencies in sub-THz communications offer a large available bandwidth, which can support extremely high data rates [6–10]. However, the use of sub-THz frequencies faces a great technical challenge of high atmospheric attenuation, which can limit the range of communications. To address this technical challenge, researchers and engineers are exploring new types of communication technologies that are efficient and optimized for 5 G NR sub-THz communications. One promising technology that can compensate for the high signal attenuation of sub-THz frequencies is free-space optical (FSO) communication. FSO is well-suited for high-frequency communications because it operates in the optical spectrum, where atmospheric attenuation is much lower compared to attenuation of guided wave at radio frequencies (RFs) [11–15]. FSO communication remains a promising technology for high-speed and long-distance free-space link in the sub-THz frequencies. High-speed and long-distance communications can be achieved by FSO communications, making them an attractive option for next-generation 5 G NR communications. In result, the use of sub-THz signals transmitted through fiber-FSO-wireless hybrid systems (see Fig. 1) can afford the potential for achieving extremely high bit rates over long-haul wireline-wireless transmissions. This hybrid system can enable wide-ranging applications such as 4 K extended reality, 5 G NR and beyond, ultra-high-speed internet access, and more.
Fig. 1. The use of sub-THz signals transmitted through fiber-FSO-wireless hybrid systems can afford the potential for achieving extremely high bit rates over long-haul wireline-wireless transmissions.
In recent years, some demonstrations of transmission at 100 Gb/s or higher operated at THz frequency have been reported. Constructing a THz-over-fiber transmission at 320-GHz with polarization-division-multiplexing (PDM) technique has been proven to be achievable [16]. Building a 0.4-THz 131-Gb/s single-channel THz photonic-wireless transmission over 10.7 m wireless link utilizing a photonic integrated circuit (PIC)-based THz transmitter has been also proven to be practicable [17]. In a previous study, researchers implemented a THz-over-fiber transmission with 5.12 Tb/s net rate in an 80-channel wavelength-division-multiplexing (WDM) system [18]. Another study enabled a 119.1-Gb/s THz photonic-wireless transmission in 350 GHz frequency through 26.8 m wireless link with probabilistic shaping (PS) technique [19]. However, sophisticated PDM, WDM, or PS technique, or complicated PIC-based THz transmitter is required. Moreover, academics achieved a 160-Gb/s wireless transmission utilizing a single THz emitter and modulating 8-channel with 25 GHz spacing in the 300-500 GHz THz band [20]. In practical THz photonic-wireless transmission, nevertheless, it is essential to transmit different data rates at different frequencies to make the most efficient use of the limited spectral resources available in the wireless domain.
THz-over-fiber (THz photonic-wireless) transmission has been widely explored and applied to meet the growing demand for high data rate applications. However, studies on sub-THz signals through fiber-FSO-5 G NR hybrid systems have not been reported, and delivering extremely high bit rates over long-haul transmissions remains a great challenge. In this demonstration, we present a fiber-FSO-5 G NR hybrid system for sub-THz signals over fiber-FSO-wireless transmission. Adopting 150-, 250-, and 325-GHz frequencies as carrier frequencies, a fiber-FSO-5 G NR hybrid system is built in practice. 150 GHz, 250 GHz, and 325 GHz carriers are adopted for 5 G NR signal transmission to meet 5 G NR sub-THz frequency band (90 GHz-350 GHz) demands. The post-forward error correction (FEC) error-free net bit rates are 59.813 Gb/s at 150 GHz, 74.766 Gb/s at 250 GHz, and 93.333 Gb/s at 325 GHz. The 150-GHz low carrier frequency is designed to afford a 59.813-Gb/s low net bit rate, whereas the 325-GHz high carrier frequency is designed to afford a 93.333-Gb/s high net bit rate. Different bit rates are transmitted at different carrier frequencies and a 227.912-Gb/s record-high aggregate post-FEC error-free net bit rate is achieved. Through a distance of 25-km single-mode fiber (SMF), 100-m FSO, and 30-m/25-m/20-m RF wireless, 5 G sub-THz 16-Gbaud, 20-Gbaud, and 28-Gbaud 16-quadrature amplitude modulation (QAM) signals are transmitted with impressive performance of satisfactorily low bit error rates (BERs) and clear and distinct constellation diagrams. Note that a 100-m FSO communication with a fiber collimator at the wireless transmission site and an optical disk antenna at the wireless reception site is equivalent to a 3500-m FSO communication with doublet lens at both wireless transmission and reception sites. This new established fiber-FSO-5 G NR hybrid system uses multiple media to transport 5 G sub-THz 16-QAM signals over long distances with good performance. It is the highest aggregate net bit rate and the longest transmission distance for sub-THz signals through fiber-FSO-wireless transmission. The successful demonstration of the fiber-FSO-5 G NR hybrid system is a significant step towards the realization of 5 G sub-THz communications.
2. Experimental setup
The optical frequency comb generator (OFCG) source shown in Fig. 2(a) consists of several components that work together to generate multiple optical carriers spaced by 25 GHz. It comprises a 25-GHz RF signal source, a 25-GHz RF amplifier, a 1 × 4 RF splitter, a distributed feedback laser diode with 1548.84 central wavelength, three phase modulators (PMs), one intensity modulator (IM), four phase shifters, a 25-GHz delay interferometer (DI), an erbium-doped fiber amplifier (EDFA), and an optical band-pass filter (OBPF). A laser light with a 100-kHz linewidth is inputted into four cascaded modulators, including three PMs and one IM. Four modulators are modulated by a 25-GHz RF signal, which decides optical carriers’ spacing. Three PMs are used to increase the modulation index, contributing to approximately 20 optical carriers. The IM is utilized to decrease the power difference between optical carriers, improving the spectral flatness [21,22]. The number of optical carriers depends on the phase modulation depth. As for the amplitude flatness of the optical carrier, it primarily depends on MZM’s modulation depth. The OFCG source can be realized using an external modulator (a PM or an MZM). The OFCG source can be also realized using a PM cascaded with an MZM. For these two scenarios, however, the phase modulation depth is too low to obtain multiple optical carriers. The number of optical carriers cannot meet system’s requirement. To obtain multiple optical carriers, a high modulation depth is required. A configuration combines a three-PM with an IM to produce an OFCG source with a flat spectral envelope, a desirable characteristic in an optical source. The generated optical carriers go through a DI to reduce the noise between each pair of optical carriers. Next, an EDFA amplifies the optical power of multiple optical carriers. The multiple optical carriers with amplified power and reduced noise are then directed to an OBPF to select and filter a specific range of optical carriers with 25 GHz spacing and ±0.7 dB power fluctuation, as Fig. 2(b) shows. The wavelengths of optical carriers 1, 7, 11, and 14 are 1548.4 nm, 1549.6 nm, 1550.4 nm, and 1551 nm, respectively. Optical carrier number 1 is selected as an optical local oscillator (LO). Three sub-THz and THz signals are generated by beating this optical LO with optical carriers numbered 7, 11, and 14. Optical carrier number 7 is positioned at 150 GHz from the optical LO (25 GHz × 6). This beating generates a 150-GHz signal. Optical carrier number 11 is positioned at 250 GHz from the optical LO (25 GHz × 10). This beating generates a 250-GHz signal. Further, optical carrier number 14 is positioned at 325 GHz from the optical LO (25 GHz × 13). This beating generates a 325-GHz signal.
Fig. 2. (a) The OFCG source consists of several components to generate multiple optical carriers spaced by 25 GHz. (b) Multiple optical carriers with amplified power and reduced noise are directed to an OBPF to select and filter a specific range of optical carriers with 25 GHz spacing and ±0.7 dB power fluctuation. The wavelengths of optical carriers 1, 7, 11, and 14 are 1548.4 nm, 1549.6 nm, 1550.4 nm and 1551 nm, respectively.
Figure 3 shows the structure of the fiber-FSO-5 G NR hybrid system, using 150-, 250-, and 325-GHz frequencies as carrier frequencies, through a 25-km SMF, 100-m FSO, and 30-m/25-m/20-m sub-THz-wave transmissions (referred to as system I). First, an OFCG source with multiple optical carriers spaced by 25 GHz (0.2 nm) is produced. After being magnified by an EDFA and controlled by a variable optical attenuator (VOA), the output of OFCG source goes through a wave shaper (Finisar 4000A) to select four optical carriers. Optical carriers numbered 1, 7, 11, and 14 are selected from four output ports of the wave shaper. The optical carrier from number 1 output port is selected as an optical LO. Optical carriers from numbers 7, 11, and 14 output ports are modulated with 16-QAM at symbol rates of 16, 20, and 28 Gbaud, respectively. Optical carriers numbered 7, 11, and 14 are spaced 150-, 250-, and 325-GHz, respectively, apart from the optical LO. Three modulated optical carriers are sent through an EDFA. A VOA optimally adjusts the optical powers, and an optical coupler integrates them with the optical LO. The integrated optical signals are subsequently transported through 25 km SMF and 100 m FSO using a fiber collimator at the wireless transmission site and an optical disk antenna at the wireless reception site. A reflective mirror is located at the optical disk antenna’s focal point to focus and direct the laser beam to the receiving areas of the subsequent EDFA. An optical disk antenna is an optical antenna with a parabolic structure to focus the incoming laser beam onto the mirror and EDFA input. Based on two reflective mirrors on one side and one reflective mirror as well as one optical disk antenna on the other side, the FSO link is greatly extended to 100 m (50 m × 4). These amplified optical signals are separated by a 1 × 3 optical splitter, optimally controlled by three separate VOAs, and detected by three separate uni-travelling carrier (UTC)-photodiodes (PDs). UTC-PD operates at different frequencies, UTC-PD-1 detects signals at 110-170 GHz frequencies and holds a responsivity of 0.28 A/W at 1550 nm, UTC-PD-2 detects signals at 220-260 GHz frequencies and holds a responsivity of 0.23 A/W at 1550 nm, and UTC-PD-3 detects signals at 300-340 GHz frequencies and holds a responsivity of 0.21 A/W at 1550 nm. At the outputs of UTC-PD-1, UTC-PD-2, and UTC-PD-3, electrical 150-GHz 16 Gbaud, 250-GHz 20 Gbaud, and 325-GHz 28 Gbaud 16-QAM sub-THz signals are generated from optically beating. Next, these sub-THz signals are amplified by three independent power amplifiers (PAs). Table 1 lists the key parameters of the PAs, including frequency range (GHz), gain (dB), output P1dB (dBm), and maximum RF input power (dBm). They are subsequently wirelessly transmitted using a set of horn antennas (HAs) operating in the D-band at 110-170 GHz, H-band at 170-260 GHz, and J-band at 220-330 GHz frequencies. These sub-THz beams are radiated to free space by a set of TPX (Polymethylpentene) lenses operating at frequencies from 0.1 to 3 THz After a transmission distance of 30 m (150 GHz)/25 m (250 GHz)/20 m (325 GHz), these sub-THz signals are down-converted into the intermediate frequencies using separate sub-THz-wave mixers with 34 GHz/17 GHz/26 GHz electrical LOs and frequency multipliers by factors of 4/14/12. Afterwards, the down-converted signals are boosted by three separate low noise amplifiers at frequencies of 2-22 GHz, and fed into a digital sampling oscilloscope for communication performance analysis.
Table 1. The key parameters of the PAs, containing frequency range, gain, output P1dB, and maximum RF input power
Fig. 3. Structure of the fibe-FSO-5 G NR hybrid system, using 150-, 250-, and 325-GHz frequencies as carrier frequencies, through a 25-km SMF, 100-m FSO, and 30-m/25-m/20-m sub-THz-wave transmissions (referred to system I).
Additionally, we change a 100-m FSO communication realized by using a fiber collimator and an optical disk antenna at the wireless transmission and reception sites (system I) to a 100-m FSO communication implemented by utilizing a set of doublet lens at the wireless transmission and reception sites (system II). Subsequently, we analyze and compare the performance of systems I and II. The free-space transmission distance of both systems I and II is 100 m. This ensures a fair comparison in terms of the physical propagation distance.
3. Experimental results and discussions
Figure 4 depicts the optical spectrum of four different signals: an unmodulated optical carrier and three optical carriers modulated with 16-QAM signals at different symbol rates. The unmodulated optical carrier operates as an optical LO. Obviously, a flat-top region at the center of each modulated signal exists. To alleviate the signal-signal beat interference, the carrier-to-signal power ratio (CSPR) is adapted to 0 dB. Maintaining a low CSPR allows for more electrical signal power to be acquired at a given optical power, resulting in an improved signal-to-noise ratio (SNR) [23–24]. Because the CSPR is set to 0 dB and the three modulated signals have a wider bandwidth, the optical LO appears to be higher in the spectrum than the three modulated signals. It can be seen that the optical LO has a peak power of -6.6 dBm and a total power of -2.4 dBm, and each modulated signal has a peak power of -7.2 dBm and a total power of -2.4 dBm. Using a UTC-PD-1 with 110-170 GHz frequencies to detect, a 150-GHz signal with 16-QAM at symbol rate of 16 Gbaud is generated by optically beating an optical LO and an optical carrier with a frequency separation of 150 GHz. Employing a UTC-PD-2 with 220-260 GHz frequencies to detect, a 250-GHz signal with 16-QAM at symbol rate of 20 Gbaud is produced by optically beating an optical LO and an optical carrier with a frequency separation of 250 GHz. Similarly, utilizing a UTC-PD-3 with 300-340 GHz frequencies to detect, a 325-GHz signal with 16-QAM at symbol rate of 28 Gbaud is yielded by optically beating an optical LO and an optical carrier with a frequency separation of 325 GHz.
Fig. 4. Optical spectrum of four different signals: an unmodulated optical carrier and three optical carriers modulated with 16-QAM signals at different symbol rates.
THz lenses are designed for sub-THz/THz beam collimation, and they can be made of various materials, including polymers, metals, and ceramics. One specific type of THz lens is the TPX lens, which is made of a polymer material. TPX lenses are transparent in the sub-THz/THz frequency ranging from 0.1 to 3 THz and have low absorption properties. Compared to the PTFE (polytetrafluoroethylene) lenses, TPX lenses offer higher transmission in the sub-THz/THz frequencies [25]. TPX lenses have lower attenuation coefficients than PTFE lenses, making them more suitable for transmitting sub-THz/THz signals. Therefore, the combination of HAs and TPX lenses is more suitable for sub-THz/THz-wave transmission than the combination of HAs and PTFE lenses, especially in applications where long-distance wireless transmission is required.
For 250-GHz signal transmission, we measure the transmission power before the H-band HA combined with the TPX lens and the reception power after the TPX lens combined with the H-band HA. A transmission power of 5 dBm and a reception power of -35.3 dBm are obtained. For HAs combined with TPX lenses, the reception power (PR) can be obtained from Friis’s equation [26–28]:
where PT is the transmission power, GT is the combined gain of HA and TPX lens at the wireless transmission site, GR is the combined gain of TPX lens and HA at the wireless reception site, l is the wireless transmission link, f is the carrier frequency, c is the speed of light, and Al is the atmospheric loss. For H-band HAs combined with TPX lenses, PT is around 5 dBm, GT and GR are 42 dBi, and l is 25 m. The atmospheric loss Al is about 15.2 dB for 25 m wireless transmission at 250 GHz. Because the wireless path loss 20·log (4 π l f /c) is calculated as 108.4 [20·log(4π·25·(250 × 109)/(3 × 108))] dB, a reception power of -34.6 dBm (5 + 42 + 42 - 108.4 - 15.2) can be obtained, which approximates the measured reception power of -35.3 dBm. Results show that a pair of H-band HAs with a set of TPX lenses can realize a long-distance wireless transmission to 25 m. Similarly, a pair of D-band HAs with a set of TPX lenses can achieve a long-distance wireless transmission to 30 m, and a pair of J-band HAs with a set of TPX lenses can realize a long-distance wireless transmission to 20 m.A 100-m FSO communication employing a fiber collimator with a divergence angle of 0.05° at the wireless transmission site and an optical disk antenna with a diameter of 120 cm and a focal length of 45.6 cm at the wireless reception site is illustrated in Fig. 5(a). Since the fiber collimator’s divergence angle is 0.05° (θ), over a 100-m FSO communication (r), the laser light’s diameter on the optical disk antenna (S) can be approximately obtained as
Fig. 5. (a) A 100-m FSO communication employing a fiber collimator with a divergence angle of 0.05° at the wireless transmission site and an optical disk antenna at the wireless reception site. (b) An l-m FSO communication employing a set of doublet lenses at the wireless transmission and reception sites.
An l-m FSO communication employing a doublet lens with 75 mm diameter and 150 mm focal length at the wireless transmission site is illustrated in Fig. 5(b). With 150 mm focal length and 0.15 fiber numerical aperture, laser light’s diameter (d) is obtained as 45 mm [2 × (150 × 0.15)]. Moreover, the corresponding beam radius [r (= 2.3/(SFC × 2π); SFC is the spatial frequency cutoff] and the laser light’s divergence angle [$\theta$(= 3.6 µm/150 mm)] are calculated as 3.6 µm and 24 × 10−6, respectively. To fit an l-m FSO communication with a set of doublet lenses into a 100-m FSO communication with a fiber collimator and an optical disk antenna, through an l-m FSO communication laser light’s diameter (dl) should be ≤ 0.174 m (174 mm):
l (maximum) is obtained as 3500 m, indicating that a 100-m FSO communication using a fiber collimator at the wireless transmission site and an optical disk antenna at the wireless reception site is equivalent to a 3500-m FSO communication using a set of doublet lenses at the wireless transmission and reception sites.Figure 6(a) shows the measured BERs as a function of UTC-PD input power of system I (25 km SMF + 100-m FSO communication using a fiber collimator and an optical disk antenna + 30-m/25-m/20-m sub-THz-wave transmissions) and system II (25 km SMF + 100-m FSO communication using a set of doublet lenses + 30-m/25-m/20-m sub-THz-wave transmissions), at three different symbol rates of 16, 20, and 28 Gbaud. Apparently, BER performances below the hard-decision FEC (HD-FEC) threshold or the soft-decision FEC (SD-FEC) threshold are attained at three different symbol rates. In system I, for 150- and 250-GHz sub-THz signal transmission with 16-QAM at symbol rates of 16 and 20 Gbaud, the BERs achieved in both transmissions are below the HD-FEC threshold of 3.8 × 10−3 with 7% overhead (7% overhead HD-FEC). The post-FEC error-free net bit rates achieved are 59.813 [(16 × 4)/(1 + 7%)] Gb/s and 74.766 [(20 × 4)/(1 + 7%)] Gb/s for 16 Gbaud and 20 Gbaud 16-QAM sub-THz-wave signal transmission, respectively. Furthermore, for 325-GHz sub-THz signal transmission with 16-QAM at symbol rate of 28 Gbaud, the BER achieved is below the SD-FEC threshold of 2 × 10−2 with 20% overhead (20% overhead SD-FEC). The post-FEC error-free net bit rate achieved is 93.333 [(28 × 4)/(1 + 20%)] Gb/s for 28 Gbaud 16-QAM sub-THz-wave signal transmission. Further, it is important to note that for different baud rates, there are specific optimal UTC-PD input power that result in the lowest BER. These input powers are 12.2 dBm for 16 Gbaud, 12.4 dBm for 20 Gbaud, and 12.7 dBm for 28 Gbaud. As the UTC-PD input power exceeds the optimal value, the sub-THz signal output power becomes saturated, and the BER performance starts to degrade. Increasing the UTC-PD input power beyond the optimal value does not improve BER performance; instead, it can have a negative impact on the sub-THz signal output power and further degrade BER performance [29–30]. Therefore, to achieve the lowest BER in sub-THz communications, it is crucial to provide the UTC-PD with the optimal input power. With 150-, 250-, and 325-GHz frequencies as carrier frequencies, an aggregate post-FEC error-free net bit rate of 227.912 Gb/s is achieved. Sufficiently low BERs verify the achievability of sub-THz-wave signal transmission with 16-QAM at symbol rates of 16, 20, and 28 Gbaud over fiber-FSO-5 G NR hybrid systems.
Fig. 6. (a) Measured BERs as a function of the UTC-PD input power of system I (25 km SMF + 100-m FSO communication using a fiber collimator and an optical disk antenna + 30-m/25-m/20-m sub-THz-wave transmissions) and system II (25 km SMF + 100-m FSO communication using a set of doublet lenses + 30-m/25-m/20-m sub-THz-wave transmissions), at three different symbol rates of 16, 20, and 28 Gbaud. The associated constellation diagrams of system I obtained at (b) 12.2 dBm for the symbol rate of 16 Gbaud, (c) 12.4 dBm for the symbol rate of 20 Gbaud, and (d) 12.7 dBm for the symbol rate of 28 Gbaud.
In systems I and II, for 16 Gbaud and 20 Gbaud 16-QAM sub-THz-wave signal transmission with 7% overhead HD-FEC threshold, the power penalties are around 3.6 and 3.8 dB, respectively. For 28 Gbaud 16-QAM sub-THz-wave signal transmission with 20% overhead SD-FEC threshold, the power penalty is about 4.1 dB. Since a 100-m FSO communication utilizing a fiber collimator and an optical disk antenna equals a 3500-m FSO communication utilizing a set of doublet lenses, these power penalties are mainly due to the additional 3400 m FSO communication loss of system I. Furthermore, laser beam alignment is a critical issue of FSO communication because it affects the BER performance and communication reliability directly. Misalignment of the laser beam can cause performance degradation, resulting in communication dropouts and reduced BER performance. Therefore, laser beam alignment is vitally important for FSO communication [31]. An optical disk antenna with high reflectivity at 1550 nm is an effective laser beam alignment scheme that can be used in FSO communication. The optical disk antenna is highly directional, allowing it to guide the laser beam accurately to the focal point [32–33]. A reflective mirror is placed at the focal point of the optical disk antenna to reflect the laser beam precisely into the subsequent device’s receiving areas. An optical disk antenna with a reflective mirror provides a reliable FSO communication and a high BER performance. Moreover, the performance of the sub-THz-wave wireless transmission is highly influenced by UTC-PD’s responsivity. Due to poorer optical coupling efficiency, the UTC-PD’s responsivity is lower than the traditional PIN-PD’s responsivity, bringing on signal with lower SNR and system with poorer BER performance. However, the UTC-PD is worth employing for optical sub-THz signal detection in fiber-FSO-5 G NR hybrid systems. The unique carrier transmission characteristic of the UTC-PD makes it well-suited for this application.
Figures 6(b), 6(c), and 6(d) present the associated constellation diagrams of system I obtained at 12.2 dBm, 12.4 dBm, and 12.7 dBm UTC-PD input power at the symbol rates of 16, 20, and 28 Gbaud, respectively. Evidently, clear and distinct constellation diagrams [Figs. 6(b) and 6(c)] and somewhat clear and distinct constellation diagrams [Fig. 6(d)] are acquired. Phase noise is higher at higher carrier frequencies, and it can degrade system’s performance by causing errors in the transmitted data [34]. The 325-GHz signal has a higher carrier frequency than the 150- and 250-GHz signals, indicating that the 325-GHz signal has higher phase noise. Additionally, the 325-GHz signal has a higher baud rate than the 150- and 250-GHz signals, also increasing the phase noise. In result, the constellation diagrams of the 325-GHz signal with 16-QAM at symbol rate of 28 Gbaud are slightly worse than those of the 150- and 250-GHz signals with 16-QAM at symbol rate of 16 and 20 Gbaud.
In system I, the resulting electrical spectra captured by down-converting the 150-GHz 16 Gbaud, 250-GHz 20 Gbaud, and 325-GHz 28 Gbaud 16-QAM sub-THz signals at 12.2 dBm, 12.4 dBm, and 12.7 dBm UTC-PD input power are displayed in Figs. 7(a), 7(b), and 7(c), respectively. As Fig. 7(a) shows, the down-converted signal, centered at 14 GHz [150–34 (LO) × 4 = 14], has a bandwidth of 10 GHz and a flat power of ±3.1 dB. A flat electrical spectrum results in good transmission performance in terms of low BER and clear and distinct constellation diagrams. Furthermore, the down-converted signal in Fig. 7(b) has a center frequency of 12 GHz [250–17 (LO) × 14 = 12] and a bandwidth of 10 GHz. The signal’s power fluctuation is slightly larger than in Fig. 7(a), with a range of ±4 dB which is an increase of ±0.9 dB compared to Fig. 7(a). The increased power fluctuation in the 250-GHz 20 Gbaud 16-QAM signal is primarily due to the higher level of inter-carrier interference. Additionally, Fig. 7(c) shows the electrical spectrum of the 325-GHz 28 Gbaud 16-QAM sub-THz signal after down-converting, with a central frequency of 13 GHz [325–26 (LO) × 12 = 13] and a bandwidth of 10 GHz. It has a tolerable power fluctuation in the range of ±4.8 dB. In comparison with the electrical spectrum presented in Fig. 7(a), it should be noted that the electrical spectra presented in Figs. 7(b) and 7(c) have lower power levels. The power levels of the 250-GHz 20 Gbaud and 325-GHz 28 Gbaud 16-QAM sub-THz signals after down-converting [Figs. 7(b) and 7(c)] are approximately 3 dB and 5 dB lower than that of the 150-GHz 16 Gbaud 16-QAM sub-THz signal after down-converting [Fig. 7(a)]. These 3 dB and 5 dB power penalties occur mainly because UTC-PD-2 and UTC-PD-3 have lower responsivities than UTC-PD-1.
Fig. 7. In system I, electrical spectra captured by down-converting the (a) 150-GHz 16 Gbaud 16-QAM sub-THz signal at 12.2 dBm UTC-PD input power, (b) 250-GHz 20 Gbaud 16-QAM sub-THz signal at 12.4 dBm UTC-PD input power, and (c) 325-GHz 28 Gbaud 16-QAM sub-THz signal at 12.7 dBm UTC-PD input power.
4. Conclusions
Transmission of sub-THz signals through fiber-FSO-5 G NR hybrid systems over a distance of 25-km SMF, 100-m FSO, and 30-m/25-m/20-m RF wireless are practically achieved. System simultaneously transports net bit rates of 59.813 Gb/s, 74.766 Gb/s, and 93.333 Gb/s in the 150-GHz, 250-GHz, and 325-GHz frequencies, respectively, through fiber-FSO-5 G NR hybrid integration. The system successfully transports different bit rates at different carrier frequencies, enabling a record-high aggregate net bit rate of 227.912 Gb/s. It is the longest transmission distance and the highest aggregate net bit rate for sub-THz signals over a fiber-FSO-5 G NR hybrid system. Good performance of satisfactorily low BERs, clear and distinct constellation diagrams, and acceptable power fluctuations in the electrical spectra are obtained, meeting the demands of 5 G NR and beyond in the sub-THz band. Such innovative demonstration of a fiber-FSO-5 G NR hybrid system signifies an important milestone towards the realization of 5 G sub-THz communications, paving the way for future advancements in 5 G NR communication systems.
Funding
Qualcomm Technologies, Inc. - NTUT Research Collaboration (NAT-514839); Hsinchu Science Park Emerging Technology Application Program (112AO02B); National Science and Technology Council (NSTC 110-2221-E-027-068-MY3, NSTC 111-2221-E-027-031-MY3).
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. V. M. Alevizaki, M. Anastasopoulos, A. I. Manolopoulos, and A. Tzanakaki, “Distributed service provisioning for disaggregated 6 G network infrastructures,” IEEE Trans. Netw. Serv. Manage. 20(1), 120–137 (2023). [CrossRef]
2. D. Chandramouli, P. Andres-Maldonado, and T. Kolding, “Evolution of timing services from 5G-A toward 6 G,” IEEE Access 11, 35150–35157 (2023). [CrossRef]
3. K. Tokugawa, J. Nakazato, H. Matsuo, K. Kubota, and K. Sakaguchi, “Implementation of the data conversion function for wireless environments for beyond 5 G applications,” IEEE 20th Consumer Commun. Netw. Conf. (CCNC)925–926 (2023).
4. T. Chauhan and P. N. Renjith, “Intelligent transformation of 6 G network by incorporating expert systems,” IEEE Intl. Students’ Conf. Electr. Electron. Comput. Sci. (SCEECS)1–5 (2023).
5. A. I. Salameh and M. E. Tarhuni, “From 5 G to 6G—challenges, technologies, and applications,” Future Internet 14(4), 117 (2022). [CrossRef]
6. G. Li, E. L. Claveau, S. K. Jawla, S. C. Schaub, M. A. Shapiro, and R. J. Temkin, “Measurement of time dependent reflection, transmission, and absorption in laser driven silicon and GaAs switches for 250 GHz radiation,” IEEE Trans. THz Sci. Technol. 13(4), 354–361 (2023). [CrossRef]
7. 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]
8. Y. Cao, K. Nallappan, G. Xu, and M. Skorobogatiy, “Add drop multiplexers for terahertz communications using two-wire waveguide-based plasmonic circuits,” Nat. Commun. 13(1), 4090 (2022). [CrossRef]
9. S. Jia, M. C. Lo, L. Zhang, O. Ozolins, A. Udalcovs, D. Kong, X. Pang, R. Guzman, X. Yu, S. Xiao, S. Popov, J. Chen, G. Carpintero, T. Morioka, H. Hu, and L. K. Oxenløwe, “Integrated dual-laser photonic chip for high-purity carrier generation enabling ultrafast terahertz wireless communications,” Nat. Commun. 13(1), 1388 (2022). [CrossRef]
10. C. Wang, K. Wang, Y. Tan, F. Wang, B. Sang, W. Li, W. Zhou, and J. Yu, “High-speed terahertz band radio-over-fiber system using hybrid time-frequency domain equalization,” IEEE Photon. Technol. Lett. 34(11), 559–562 (2022). [CrossRef]
11. 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. McClung, 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]
12. X. H. Huang, C. Y. Li, H. H. Lu, C. R. Chou, H. M. Hsia, and Y. H. Chen, “A bidirectional FSO communication employing phase modulation scheme and remotely injection-locked DFB LD,” J. Lightwave Technol. 38(21), 5883–5892 (2020). [CrossRef]
13. K. Mallick, P. Mandal, R. Mukherjee, G. C. Mandal, B. Das, and A. S. Patra, “Generation of 40 GHz/80 GHz OFDM based MMW source and the OFDM-FSO transport system based on special fine tracking technology,” Opt. Fiber Technol. 54, 102130 (2020). [CrossRef]
14. K. Mallick, P. Mandal, G. C. Mandal, R. Mukherjee, B. Das, and A. S. Patra, “Hybrid MMW-over fiber/OFDM-FSO transmission system based on doublet lens scheme and POLMUX technique,” Opt. Fiber Technol. 52, 101942 (2019). [CrossRef]
15. W. S. Tsai, H. H. Lu, C. Y. Li, T. C. Lu, C. H. Liao, C. A. Chu, and P. C. Peng, “A 20-m/40-Gb/s 1550-nm DFB LD-based FSO link,” IEEE Photonics J. 7(6), 1–7 (2015). [CrossRef]
16. C. Wang, J. Yu, J. Zhang, M. Zhu, Y. Tan, K. Wang, J. Ding, W. Zhou, T. Xie, J. Yu, and F. Zhao, “Beyond 300-Gbps/λ photonics-aided THz-over-fiber transmission employing MIMO single-carrier frequency-domain equalizer,” Opt. Lett. 48(6), 1363–1366 (2023). [CrossRef]
17. S. Jia, X. Pang, O. Ozolins, X. Yu, H. Hu, J. Yu, P. Guan, F. D. Ros, S. Popov, G. Jacobsen, M. Galili, T. Morioka, D. Zibar, and L. K. Oxenløwe, “0.4 THz photonic-wireless link with 106 Gb/s single channel bitrate,” J. Lightwave Technol. 36(2), 610–616 (2018). [CrossRef]
18. J. Ding, L. Zhang, J. Liu, W. Li, Y. Wang, K. Wang, L. Zhao, W. Zhou, J. Zhang, M. Zhu, and J. Yu, “THz-over-fiber transmission with a net rate of 5.12 Tbps in an 80 channel WDM system,” Opt. Lett. 47(12), 3103–3106 (2022). [CrossRef]
19. S. Wang, Z. Lu, W. Li, S. Jia, L. Zhang, M. Qiao, X. Pang, N. Idrees, M. Saqlain, X. Gao, X. Cao, C. Lin, Q. Wu, X. Zhang, and X. Yu, “26.8-m THz wireless transmission of probabilistic shaping 16-QAM-OFDM signals,” APL Photonics 5(5), 056105 (2020). [CrossRef]
20. X. Yu, S. Jia, H. Hu, M. Galili, T. Morioka, P. U. Jepsen, and L. K. Oxenløwe, “160 Gbit/s photonics wireless transmission in the 300-500 GHz band,” APL Photonics 1(8), 081301 (2016). [CrossRef]
21. H. H. Lu, C. Y. Li, W. S. Tsai, Y. E. Peng, W. C. Fan, C. J. Lin, Y. S. Tang, Y. X. Chen, and Y. S. Lin, “Two-way 5 G NR fiber-wireless systems using single-carrier optical modulation for downstream and phase modulation scheme for upstream,” J. Lightwave Technol. 41(6), 1749–1758 (2023). [CrossRef]
22. S. Zou, Y. Wang, Y. Shao, J. Zhang, J. Yu, and N. Chi, “Generation of coherent optical multi-carriers using concatenated, dual-drive Mach-Zehnder and phase modulators,” Chin. Opt. Lett. 10(7), 070605 (2012). [CrossRef]
23. H. H. Lu, C. Y. Li, W. S. Tsai, R. D. Lin, Y. S. Tang, Y. X. Chen, Y. S. Lin, and W. C. Fan, “An integrated fiber-FSO-5 G NR sub-THz link with 86.112 Gbps high aggregate data rates,” J. Lightwave Technol. 40(24), 7790–7798 (2022). [CrossRef]
24. T. Ji, C. Sun, H. Ji, Z. Xu, Y. Peng, and W. Shieh, “Theoretical and experimental investigations of interleaved carrier-assisted differential detection,” J. Lightwave Technol. 39(1), 122–128 (2021). [CrossRef]
25. A. S. Kucheryavenko, V. A. Zhelnov, D. G. Melikyants, N. V. Chernomyrdin, S. P. Lebedev, V. V. Bukin, S. V. Garnov, V. N. Kurlov, K. I. Zaytsev, and G. M. Katyba, “Super-resolution THz endoscope based on a hollow-core sapphire waveguide and a solid immersion lens,” Opt. Express 31(8), 13366–13373 (2023). [CrossRef]
26. H. H. Lu, C. Y. Li, X. H. Huang, C. J. Lin, R. D. Lin, Y. S. Lin, Y. S. Tang, and W. C. Fan, “A combined fibre/free-space-optical communication system for long-haul wireline/wireless transmission at millimetre-wave/sub-THz frequencies,” Commun. Eng. 2(1), 18 (2023). [CrossRef]
27. X. Li, J. Yu, Z. Dong, Z. Cao, N. Chi, J. Zhang, Y. Shao, and L. Tao, “Seamless integration of 57.2-Gb/s signal wireline transmission and 100-GHz wireless delivery,” Opt. Express 20(22), 24364–24369 (2012). [CrossRef]
28. D. C. Hogs, “Fun with the Friis Free-Space Transmission Formula,” IEEE Antennas Propag. Mag. 35(4), 33–35 (1993). [CrossRef]
29. T. Ishibashi and H. Ito, “Uni-traveling-carrier photodiodes,” J. Appl. Phys. 127(3), 031101 (2020). [CrossRef]
30. T. Ishibashi, Y. Muramoto, T. Yoshimatsu, and H. Ito, “Unitraveling-carrier photodiodes for terahertz applications,” IEEE J. Select. Topics Quantum Electron. 20(6), 79–88 (2014). [CrossRef]
31. C. Y. Li, X. H. Huang, H. H. Lu, Y. C. Huang, Q. P. Huang, and S. C. Tu, “A WDM PAM4 FSO–UWOC integrated system with a channel capacity of 100 Gb/s,” J. Lightwave Technol. 38(7), 1766–1776 (2020). [CrossRef]
32. M. M. Abadi and Z. Ghassemlooy, “A hybrid free-space optical (FSO)/radio frequency (RF) antenna for satellite applications,” Intl. Symposium on Commun. Systems, Netw. Digital Signal Process. (CSNDSP) 6, 185–190 (2022). [CrossRef]
33. O. A. Saraereh, “A novel broadband antenna design for 5 G applications,” Computers, Materials & Continua 67(1), 1121–1136 (2021). [CrossRef]
34. D. C. Shin, B. S. Kim, H. Jang, Y.-J. Kim, and S. W. Kim, “Photonic comb-rooted synthesis of ultra-stable terahertz frequencies,” Nat. Commun. 14(1), 790 (2023). [CrossRef]
