Open Access
Add to BibSonomy
Abstract
The link alignment is a challenge in underwater wireless optical communication (UWOC). This paper proposes a UWOC system adopting a fisheye lens with a field of view (FOV) of ±90° at the receiver to alleviate alignment requirement, and a mobile scanning device (MSD) is exploited to track the variation of the imaging position generated by the fisheye lens due to different incidence angles. In a 7-m tap water channel, a transmission with a data rate of 400 Mbps and an FOV of ±90° is realized with 16-quadrature amplitude modulating-orthogonal frequency division multiplexing (16-QAM-OFDM) modulation and orthogonal matching pursuit (OMP) channel estimation algorithm.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
In recent years, many researchers have committed to underwater wireless optical communication (UWOC) which is an important underwater communication method with high bandwidth and low latency. With the innovation in the design of optoelectronic devices and signal processing algorithms, breakthroughs have been made in transmission distance extension [1–3], data rate promotion [4–6], alignment requirement alleviation [7–9], and information encryption [10,11]. The UWOC can be divided into line-of-sight (LOS) and non-line-of-sight (NLOS) in terms of link structure. The NLOS communication systems mainly exploit the scattering or reflection of optical signals for communication, which can relax the alignment requirement [12,13]. However, the data rate and transmission distance of NLOS systems are usually limited. Instead, the LOS systems are prevalently adopted in UWOC studies to realize a higher data rate and a longer transmission range using a laser diode (LD) with a small divergence angle, while requiring strict alignment between the transmitter and receiver in the actual deployment.
To alleviate the alignment requirements of LOS system, researchers have made many attempts. It is natural to enlarge the radiation area of the transmitter. A prism array formed by three light-emitting-diodes (LEDs) was designed to implement a quasi-omnidirectional UWOC transmission with a data rate of 29.85 Mbps at a transmission distance of 35 m in a swimming pool [14]. In another study, light was generated by exciting the perovskite via an LD, and achieved a quasi-omnidirectional communication with a transmission distance of 35 m and a data rate of 40 Mbps in a swimming pool [15]. It is also noticed that the divergence angle of an LD is generally smaller than that of an LED, so an engineering diffuser and a ground glass diffuser were used to expand the divergence angle and homogenize the energy distribution of the light spot, achieving a field of view (FOV) of ±12° [16]. Moreover, the acquisition, pointing, and tracking (APT) method effectively solves the alignment problems. For instance, a transmitter used a camera to identify the target installed on the receiver for APT within 574 ms in a 0.28-m water channel [8].
Although expanding the coverage range of the transmitter can reduce the difficulty of alignment, it also introduces significant geometric loss of optical power that limits the communication distance. Therefore, relevant studies about alignment on the receiver have also been conducted. An APT method was realized by calculating the deviation angle of a laser beam spot captured by a camera at the receiver [7]. In addition, a 3 × 3 solar array with a size of 34mm × 34 mm was used as a detector to relieve the alignment difficulty by expanding the detection area of the receiver and realized a data rate of 150 Mbps over a 35-m underwater channel [17]. Moreover, fluorescent fibers were exploited in free-space optical communication (FSO) and exhibited advantages on FOV in UWOC systems [18,19]. Meanwhile, a visible light communication (VLC) system used five photoelectric detectors (PDs) distributed in a cube to increase the FOV of the receiver to 360° achieving a data rate of 20 Mbps with a range of 7.1 m [20].
In this letter, a fisheye lens is adopted in the UWOC system to expand the FOV of the receiver to ±90°. And stepper motors are also utilized to drive the PD scanning near the focal plane of the fisheye lens to track of the light spot formed by the signal. Then an envelope detector is used to detect the positive voltage amplitude of the received optical signals envelope, and a voltage threshold is set to determine whether it is the time to receive communication signals or start scanning for link alignment. And the proposed UWOC system realizes a data rate of 400 Mbps with the orthogonal matching pursuit (OMP) algorithm for channel estimation and zero forcing equalization in a 7-m tap water channel.
2. Principle
2.1 Fisheye lens
It is noticed that a fisheye lens has a large FOV. Therefore, when the fisheye lens is used as the optical antenna of a UWOC receiver, the FOV of the UWOC system can be enlarged. However, the aberration of the fisheye lens will significantly affect the system performance. Generally, to correct the aberration, the lens aperture will be adjusted to limit the off-axis light imaging during the design. Therefore, the received optical power (ROP) passing through the fisheye lens will decline when the incident angle is large. Besides, the incident light with various angles leads to a changing image positions (IPs) in the focal plane of the fisheye lens.
In this letter, a commercially available fisheye lens (SAMYANG) was adopted with a focal length (FL) of 12 mm, numerical aperture (NA) of 1/2.8, FOV of 180°, and diameter of 62 mm. Its pupil diameter (D) is 4.28 mm, which is calculated by
To show the ROP and IP variation of the fisheye lens, an experiment was conducted. In the measurement, an LD with an emitted optical power of 19.39 dBm is positioned on a circumference centered at the fisheye lens with a radius of 1 meter, as shown in Fig. 1(a). And the measured results are shown in Fig. 1(b). As shown in Fig. 1(b), there is an almost linear correlation between IP and incident angle. The ROP fluctuates slightly when the incident angle varies from 0° to 70°, and reduces by about 3 dB when the incident angle varies between 80° and 90°. The average attenuation caused by the fisheye lens is 15.7 dB. With the fisheye lens working as an optical antenna of the receiver, the UWOC system is able to achieve an FOV of ±90°.
2.2 Mobile scanning device
As mentioned earlier, the variation of the IP was unavoidable when the incident angle of the light changed. To solve this problem, we proposed a mobile scanning device (MSD) to assist the PD in tracking the spot. The combination of MSD and fisheye lens can realize a large angle reception. A power splitter was used to separate the received signal into two parts. One part was used for signal demodulation, and the other part for feedback controlling the movement of the PD.
Fig. 1. (a) The measurement setup,.(b) The ROPs and IPs of the light passing through the fisheye lens with different incident angles.
The structure of the MSD is shown in Fig. 2. A PD is connected to the input end of a power splitter, and the two output ends of the power splitter are connected to an oscilloscope and an envelope detector, respectively. The envelope detector with a cut-in voltage of 0.3 V and a drop of 0.2 V can obtain the positive voltage amplitude of the signal envelope. The envelope detector is followed by an analog-to-digital converter (ADC, PCF8591) connected to a Raspberry Pi. The PD is fixed at a horizontally moving rail driven by a stepper motor, and the rail itself is set at a longitudinally moving rail driven by the other stepper motor simultaneously. And the two motors are controlled by corresponding drivers using pulse width modulation (PWM) waves generated by the Raspberry Pi. Besides, to offer position information of the PD, a red LED connected to the Raspberry Pi is put 17-mm away longitudinally from the PD target surface. A wide-angle camera connected to the Raspberry Pi is utilized to capture the red LED and the light spot on the receiving screen. Finally, all of the required information is sent to the Raspberry Pi for further processing.
The decision tree of the proposed MSD is shown in Fig. 3. When the incident optical signal passes through the fisheye lens with a certain angle, the wide-angle camera obtains the position of the light spot relative to the red LED. Then, a rough alignment is implemented by moving the PD target surface to the location of the light spot with a relatively high speed. After getting the positive voltage of the signal envelope, a threshold is used to determine whether it is the right time to detect signal or start scanning. If the voltage of the signal envelope is greater than the threshold, the system starts to receive communication signals via the oscilloscope. Otherwise, when the positions of the LED and the light spot are not accurately obtained due to the reflected light on the reception plane and other errors, the fine alignment is conducted at a relatively low speed with the S-shaped scanning route until the voltage surpasses the threshold. The above process is repeated until the condition is met. The initial position of the PD is the position where it stops last time.
3. Experimental setup
The experimental setup of the proposed UWOC system is shown in Fig. 4. At the transmitter, a 16-quadrature amplitude modulating-orthogonal frequency division multiplexing (16-QAM-OFDM) modulated signal with the parameters listed in Table 1 was loaded into an arbitrary waveform generator (AWG) for generating a baseband electrical signal. Then, the signal was amplified by a power amplifier (AMP) with a 3-dB bandwidth of 150 MHz, and adjusted by a variable electrical attenuator (VEA) with the attenuation value of 8 dB. Then a Bias-Tee was used to combine the bipolar signal of 0.5 Vpp and a direct current (DC) of 0.38 A. Finally, the optical signal was generated by a 450-nm LD (NDB7875) with an output optical power of 24.47 dBm.
Fig. 4. Experimental setup of the proposed UWOC system. Insets: (i) transmitter, (ii) light in the water channel, (iii) receiver.
Table 1. OFDM Parameters
At the receiver, the size of the light spot was about 140 mm ×50 mm after the transmission of a 7-m tank filled with tap water, and the attenuation coefficient of water was 0.05 m−1 (0.23 dB/m). In this work, the fisheye lens and MSD device were placed on a rotary table for changing the incident angle. An avalanche photo diode (APD, APD430A2) with a target surface diameter of 0.2 mm in the MSD was placed 47-mm away from the fisheye lens, and such a distance was farther than the focal distance of the fisheye lens. The size of light spot was about 2 mm on the receiving screen, so there was a 6.6-dB optical gain caused by the fisheye lens, considering the 4.28-mm pupil diameter of the fisheye lens. The received signal was equalized by zero forcing with the channel impulse response which was obtained by the OMP algorithm. The normalized frequency response of the proposed UWOC system was measured by a network analyzer (Hewlett Packard, 8753D), and the 3-dB bandwidth is about 150 MHz as shown in Fig. 5. The scanning speed and area were set to 0.1 mm/s and 36 mm2, respectively. As the fisheye lens is symmetrical about the optical axis, it only needs to measure the incident angle from 0° to 90° horizontally which is sufficient to demonstrate that the proposed UWOC system can achieve a receiving FOV of ± 90°.
4. Experiment results and discussion
In order to optimize the voltage threshold, we measured its impact on bit error ratio (BER) without the rough alignment. The voltage threshold was set from 0.1 V to 0.8 V under the incident angle of 90° and a data rate of 50 Mbps. The BER as a function of voltage threshold with the least square (LS) algorithm for channel estimation and zero forcing equalization is shown in Fig. 6. When the threshold increases, the BER gradually decreases, and the variation of voltage threshold has limited influence on BER in the range of 0.5 V to 0.8 V.
Fig. 6. The BER with the threshold from 0.1 V to 0.8 V under the incident angle of 90° and a data rate of 50 Mbps.
The relationship between the voltage of the signal envelope and the data rate, as well as the relationship between the voltage of the signal envelope and the ROP under the data rate of 100 Mbps was measured at the incident angle of 90°, as shown in Fig. 7. As shown in Fig. 7(a), when the data rate increases, it is observed that the voltage of the signal envelope decreases. When the ROP increases, the voltage of the signal envelope becomes bigger as well shown in Fig. 7(b).
Fig. 7. (a) The relationship between voltage of signal envelope and the data rate. (b) The relationship between voltage of signal envelope and the ROP under a data rate of 100 Mbps.
The data rate was set to 100 Mbps, and the voltage threshold was set to 0.5 V. Once the signal envelope voltage reached the threshold, the stepper motors stopped moving, and the APD started to receive the signal. Then the data rate was reset to 200 Mbps, 400 Mbps, 625 Mbps and 800 Mbps, respectively. Attributed to the characteristics of the UWOC channel, the compression sensing theory can be used for channel estimation, which had been tested in simulation [21,22]. The LS algorithm was employed for channel estimation by ignoring the noise, which led to large errors when the signal-to-noise ratio (SNR) was low. The OMP algorithm decomposed errors several times until the given sparsity level was met. So the OMP algorithm can obtain higher channel estimation accuracy compared with the LS algorithm [23]. And the OMP algorithm was used for channel estimation in this work.
When the incident angle of the light varies from 0° to 90°, the BER performance under different data rates using the OMP algorithm with different sparsity levels is shown in Fig. 8. And the red dotted line refers to the forward error correction (FEC) threshold of 3.8 × 10−3. When the sparsity level is small, the BER rapidly decreases as the sparsity level increases. After the sparsity level reaching a certain value, the BER keeps steady. The BERs with different data rates and sparsity levels under the incident angle of 90° are summarized in Fig. 9. Besides, it is noticed that a higher data rate requires a higher sparsity level. The sparsity level of 9, 11, 13, and 15 are selected in the data rate of 200 Mbps, 400 Mbps, 625 Mbps, and 800 Mbps, respectively.
Fig. 8. The BERs with different sparsity levels under the data rate of (a) 200 Mbps, (b) 400 Mbps, (c) 625 Mbps, (d) 800 Mbps.
The BERs with the OMP algorithm and the LS algorithm are compared in Fig. 10(a). It can be observed that all the BER curves present an upward trend with the increase of the incident angle. And there is a BER variation among incident angles, which is attributed to different scanning routes and ROPs. Compared to the LS algorithm, applying the OMP algorithm can reduce the BER by 59.5%, 43.3%, 40.4%, and 34.9%, corresponding to the data rate of 200 Mbps, 400 Mbps, 625 Mbps, and 800 Mbps, respectively. Especially, with the incident angles of 90° and the data rate of 400 Mbps, the BER with the LS algorithm can be decreased from 5.75 × 10−3 to 3.37 × 10−3 by the OMP algorithm. Figure 10(b) shows the BERs of a UWOC system with a bare APD at the same experimental conditions. The ROP received by the APD is proportional to the cosine of the incidence angle. As the incident angle increases, the ROP decreases, leading to the decrease of SNR and the increase of BER. The maximum angle corresponding to BER lower than FEC threshold can be considered as the FOV. When the incident angle is larger than FOV, the advantage of the OMP algorithm over the LS algorithm is not obvious. As the data rate increases, the FOV becomes smaller, and its FOV is ±60° under the data rate of 100 Mbps. Besides, the BER under the data rate of 400 Mbps is at the level of 10−2 without the fisheye lens, and other angles without BER represent that the waveforms of signal are not captured by the oscilloscope in Fig. 10(b). Therefore, the fisheye lens not only enlarged the FOV of the proposed UWOC system to ±90°, but also extended the communication distance with an optical gain of 6.6 dB according to the optical power budget.
Fig. 10. The BERs of the LS and OMP algorithms under different data rates and incident angles in (a) the proposed UWOC system, insets: constellation diagram of 90° in the data rate of 400 Mbps with the OMP algorithm, (b) the UWOC system with a bare APD, insets: constellation diagram of 70° in the data rate of 100 Mbps with the OMP algorithm.
The scanning time and ROP corresponding to different incident angles are shown in Fig. 11. The overall trend of the ROP is similar to that in Fig. 1(b). The average alignment time is 127 seconds, and the alignment time for five angles is within 10 seconds which means the link was aligned only with the rough alignment. The alignments for large angles take relative long time, which may be due to the lower ROP.
5. Conclusion
In a 7-m underwater channel, the proposed UWOC system with a fisheye lens and MSD can achieve an FOV of ±90° at the receiver, as well as a data rate of 400 Mbps via modulation format of 16-QAM-OFDM and the OMP algorithm for channel estimation. Compared to the LS algorithm, the OMP algorithm has better performance on BER in proposed UWOC system. In the future, a duplex UWOC system will be tested with the MSDs, fisheye lens and multi-pixel photon counters (MPPCs) for dynamic reception in a swimming pool.
Funding
Science Foundation of Donghai Laboratory (DH-2022KF01015); National Key Research and Development Program of China (2022YFB2903403, 2022YFC2808200); National Natural Science Foundation of China (61971378); Strategic Priority Research Program of the Chinese Academy of Sciences (XDA22030208).
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. X. Chen, X. Yang, Z. Tong, Y. Dai, X. Li, M. Zhao, Z. Zhang, J. Zhao, and J. Xu, “150 m/500 Mbps Underwater Wireless Optical Communication Enabled by Sensitive Detection and the Combination of Receiver-Side Partial Response Shaping and TCM Technology,” J. Lightwave Technol. 39(14), 4614–4621 (2021). [CrossRef]
2. Y. Dai, X. Chen, X. Yang, Z. Tong, Z. Du, W. Lyu, C. Zhang, H. Zhang, H. Zou, Y. Cheng, D. Ma, J. Zhao, Z. Zhang, and J. Xu, “200-m/500-Mbps underwater wireless optical communication system utilizing a sparse nonlinear equalizer with a variable step size generalized orthogonal matching pursuit,” Opt. Express 29(20), 32228 (2021). [CrossRef]
3. J. Wang, C. Lu, S. Li, and Z. Xu, “100 m/500 Mbps underwater optical wireless communication using an NRZ-OOK modulated 520 nm laser diode,” Opt. Express 27(9), 12171–12181 (2019). [CrossRef]
4. C. Fei, X. Hong, G. Zhang, J. Du, Y. Gong, J. Evans, and S. He, “16.6 Gbps data rate for underwater wireless optical transmission with single laser diode achieved with discrete multi-tone and post nonlinear equalization,” Opt. Express 26(26), 34060–34069 (2018). [CrossRef]
5. Y. Zhou, X. Zhu, F. Hu, J. Shi, F. Wang, P. Zou, J. Liu, F. Jiang, and N. Chi, “Common-anode LED on a Si substrate for beyond 15 Gbit/s underwater visible light communication,” Photonics Res. 7(9), 1019–1029 (2019). [CrossRef]
6. 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]
7. J. Tang, R. Jiang, Z. Chen, and Z. Zhu, “Monocular vision aided optical tracking for underwater optical wireless communications,” Opt. Express 30(9), 14737–14747 (2022). [CrossRef]
8. J. Lin, Z. Du, C. Yu, W. Ge, W. Lü, H. Deng, C. Zhang, X. Chen, Z. Zhang, and J. Xu, “Machine-vision-based acquisition, pointing, and tracking system for underwater wireless optical communications,” Chin. Opt. Lett. 19(5), 050604 (2021). [CrossRef]
9. N. D. Hardy, H. G. Rao, S. D. Conrad, T. R. Howe, M. S. Scheinbart, R. D. Kaminsky, and S. A. Hamilton, “Demonstration of vehicle-to-vehicle optical pointing, acquisition, and tracking for undersea laser communications,” Proc. SPIE 10910, 35 (2019). [CrossRef]
10. J. Du, Y. Wang, C. Fei, R. Chen, G. Zhang, X. Hong, and S. He, “Experimental demonstration of 50-m/5-Gbps underwater optical wireless communication with low-complexity chaotic encryption,” Opt. Express 29(2), 783–796 (2021). [CrossRef]
11. H. Deng, Z. Du, J. Xiong, X. Yang, Y. Hua, and J. Xu, “Security enhancement for OFDM-UWOC system using three-layer chaotic encryption and chaotic DFT precoding,” Chin. Opt. Lett. 20(11), 110601 (2022). [CrossRef]
12. M. Sait, X. Sun, O. Alkhazragi, N. Alfaraj, M. Kong, T. Khee Ng, and B. S. Ooi, “The effect of turbulence on NLOS underwater wireless optical communication channels,” Chin. Opt. Lett. 17(10), 100013 (2019). [CrossRef]
13. X. Sun, M. Kong, O. Alkhazragi, C. Shen, E.-N. Ooi, X. Zhang, U. Buttner, T. K. Ng, and B. S. Ooi, “Non-line-of-sight methodology for high-speed wireless optical communication in highly turbid water,” Opt. Commun. 461, 125264 (2020). [CrossRef]
14. Z. Tong, X. Yang, X. Chen, H. Zhang, Y. Zhang, H. Zou, L. Zhao, and J. Xu, “Quasi-omnidirectional transmitter for underwater wireless optical communication systems using a prismatic array of three high-power blue LED modules,” Opt. Express 29(13), 20262 (2021). [CrossRef]
15. X. Li, Z. Tong, W. Lyu, X. Chen, X. Yang, Y. Zhang, S. Liu, Y. Dai, Z. Zhang, C. Guo, and J. Xu, “Underwater quasi-omnidirectional wireless optical communication based on perovskite quantum dots,” Opt. Express 30(2), 1709–1722 (2022). [CrossRef]
16. C. Yu, X. Chen, Z. Zhang, G. Song, J. Lin, and J. Xu, “Experimental verification of diffused laser beam-based optical wireless communication through air and water channels,” Opt. Commun. 495, 127079 (2021). [CrossRef]
17. Z. Tong, X. Yang, H. Zhang, Y. Dai, X. Chen, and J. Xu, “Series-connected solar array for high-speed underwater wireless optical links,” Opt. Lett. 47(5), 1013–1016 (2022). [CrossRef]
18. T. Peyronel, K. J. Quirk, S. C. Wang, and T. G. Tiecke, “Luminescent detector for free-space optical communication,” Optica 3(7), 787–792 (2016). [CrossRef]
19. C. H. Kang, A. Trichili, O. Alkhazragi, H. Zhang, R. C. Subedi, Y. Guo, S. Mitra, C. Shen, I. S. Roqan, T. K. Ng, M. S. Alouini, and B. S. Ooi, “Ultraviolet-to-blue color-converting scintillating-fibers photoreceiver for 375-nm laser-based underwater wireless optical communication,” Opt. Express 27(21), 30450–30461 (2019). [CrossRef]
20. P. Nabavi and M. Yuksel, “Comprehensive Design and Prototype of VLC Receivers with Large Detection Areas,” J. Lightwave Technol. 38(16), 4187–4204 (2020). [CrossRef]
21. W. Ding, F. Yang, W. Dai, and J. Song, “Time-Frequency Joint Sparse Channel Estimation for MIMO-OFDM Systems,” IEEE Commun. Lett. 19(1), 58–61 (2015). [CrossRef]
22. X. Liu, J. Hu, K. Zhang, X. Tang, and Y. Dong, “On Channel Estimation Based on Compressed Sensing for OFDM UWOC Systems,” Asia Communications and Photonics Conference, 1039–1042 (2022).
23. G. Z. Karabulut and A. Yongacoglu, “Sparse channel estimation using orthogonal matching pursuit algorithm,” 60th Vehicular Technology Conference3880 (IEEE, 2004).
