Open Access
Add to BibSonomy
Abstract
We experimentally demonstrated high-speed diffuse line-of-sight optical wireless communication across a wavy water-air-interface. The testbed channel was evaluated, in terms of data rate, coverage and robustness to the dynamic wave movement, based on the performance of different modulation schemes, including non-return-to-zero on-off keying (NRZ-OOK) and quadrature amplitude modulation (QAM)-orthogonal frequency division multiplexing (OFDM). Under the emulated calm water condition, 8-QAM-OFDM offers a data rate of 111.4 Mbit/s at the aligned position, while only 55 Mbit/s is achieved using NRZ-OOK. On the other hand, effective communication can still be maintained at a high data rate of 11 Mbit/s when the photodetector is off aligned laterally by 5 cm based on NRZ-OOK modulation, leading to a coverage of ~79 cm2. By utilizing OFDM modulation scheme, a data rate of 30 Mbit/s can be achieved up to 2.5-cm misalignment, leading to a coverage of ~20 cm2. Furthermore, in the presence of strong waves (15-mm wave height, causing a scintillation index of 0.667), 4-QAM-OFDM modulation showed a better resilience to channel instability than NRZ-OOK modulation. Our studies pave the way for the eventual realization of communication across a challenging water-air interface without the need for an interface relay, which is much sought-after for implementing a robust and large-coverage underwater-to-terrestrial internet-of-things.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Owing to growing underwater activities [1], underwater wireless communication is of great interest to the military, industry, and scientific communities. Radio frequency (RF) [2], acoustic [3], and optical communications [4] are the available technologies deployed in subsea communications. However, in addition to focusing on communications in the sea, it is important to construct communication links between underwater platforms [e.g., autonomous underwater vehicles (AUVs) and sensors] and terrestrial platforms [e.g., unmanned aerial vehicles (UAVs) and base stations] for data transmission. This is especially important in underwater rescue missions, underwater monitoring, and communications with submarines, as shown in Fig. 1.
Fig. 1 The illustration of signaling and activities underwater for data transmissions from underwater platforms to airborne, terrestrial and space platforms. Communication techniques having high data rate, large coverage, and robustness to dynamic waves movement are essential for future internet-of-things (IoT) and internet-of-underwater-things (IoUT).
The difficulty in constructing such links across a water-air interface is mainly imposed by the different properties experienced by the signal carrier in different media. Although RF signals can travel a long distance up to tens of kilometers in the air, they evanesce exponentially in the water owing to the large attenuation coefficients (tens of dB/m) [5]. Therefore, underwater RF communications mainly leverage extremely low frequencies (ELF, 3–30 Hz) for their low water attenuation coefficients. The U.S. Navy conducted Project ELF in 1989 using an ELF as a carrier for communication with submerged submarines. This system used a huge antenna (23 km long) to send very high-power signals (2.6 MW) to submarines at depths of several hundred meters whether in open water or under an ice pack.
However, limited by the bandwidth of ELF, this system featured an extremely low data rate (on the order of minutes per character), and it was one-way communication because the submarines could not be mounted with such a huge antenna [6]. By contrast, the acoustic waves traveled over several km [5] in the water while experiencing severe reflections off the water surface [7]. Therefore, it is difficult to use a single type of acoustic or RF signal for communications across a water-air interface.
Conventional solutions for communications across a water-air interface rely either on AUVs or on partially submerged relays as data mules. The AUVs or relays are equipped with an acoustic transceiver to communicate with underwater users and interact with command users over radio airwaves [8,9]. However, both need to float on the surface of the water when performing data transmission, thus imposing security challenges as they may disclose their locations to an adversary.
Alternatively, by combining two techniques without using gateways, such as in acoustic RF [10] or acoustic-optical light [11], communications across water-air interface have also been demonstrated. By sending acoustic waves underwater (which travel as pressure waves), inducing the displacement of the water surface, and using RF or optical beams to detect the displacement, the MIT Media Lab [10] and U.S. Navy [11] demonstrated such across-water-air interface communication. However, because the displacements of the water surface induced by the acoustics are usually on the order of ~μm [10], airborne platforms need to be very close to the water surface to detect such tiny surface displacements, and such a system requires strict alignment between the transmitter and receiver along a vertical axis.
Optical wireless communication has been studied as an alternative solution for underwater wireless communication up to 100 m by the Woods Hole Oceanographic Institution [12], and for transmission in air at full Ethernet speed (10 Mbit/s) over a distance of 1–2 km by Reasonable Optical Near Joint Access [13]. Hence, it is possible to leverage optical light to transmit the signal across the water-air surface. Although a few high-speed (Gbit/s) and long-distance (26 m) across-water-air communications under calm water surface systems have been demonstrated [14,15], these line-of-sight (LOS) systems cannot support stable across-water-air interface communication owing to the strict optical alignment requirement. This is going to become more challenging, especially in the presence of terrestrial and/or underwater turbulence, which causes misalignment due to the variation of refractive index in the medium [16,17]. Furthermore, limited by the requirements on the alignment, such LOS communications cannot provide a large coverage area for mobile users.
Diffused optical light is promising because it boosts the coverage area such that the communication channel still remains effective even when a misalignment occurs between the transmitter and receiver [18]. However, optical light suffers from absorption, scattering [19], and turbulence [20–22] in the medium. As mentioned above, it can still travel long distances in both water and air [12,13]. High bandwidths of up to a few GHz and low latencies in the range of a few nanoseconds have also been demonstrated by optical communication technology [14,23,24]. Therefore, these distinctive features may enable diffuse optical communication in across water-air interface communication.
Note that although blue-green light is preferred in underwater wireless optical communication owing to the low attenuation in the water [25], ultraviolet (UV) light is adopted in this study as the signal carrier. This is because there is low background solar radiation and low device dark noise in the UV band owing to absorption from the ozone layer [26]. This feature is advantageous in complex channels, i.e., across water and air, for providing higher signal-to-noise ratios (SNRs). In addition, the natural scattering of UV radiation by abundant molecules and aerosols [27] offers an outstanding advantage when constructing a diffuse-line-of-sight (diffuse LOS) or even non-line-of-sight (NLOS) communication link [28,29].
In this work, we present the first experimental demonstration of high-speed, diffused line-of-sight optical wireless communications across a wavy water-air-interface. The performance of different modulation schemes are compared in this channel, including quadrature amplitude modulation (QAM)-orthogonal frequency division multiplexing (OFDM), which offers spectral efficiency and channel-instability resilience, as well as non-return-to-zero on-off-keying (NRZ-OOK), which is the simplest, less stringent SNR-requirement modulation-scheme. Under calm water condition, a data rate of 111.4 Mbit/s is achieved by using 8-QAM-OFDM at the perfectly aligned position, while NRZ-OOK can only offer a data rate of 55 Mbit/s at the same position. On the other hand, effective communication is can still be maintained when using NRZ-OOK at 11 Mbit/s when the photodetector is off-aligned up to 5 cm (i.e., a coverage of ~79 cm2), while that of OFDM is only 2.5 cm, leading to a coverage of ~20 cm2. In the presence of strong waves (15-mm wave height, causing a scintillation index of 0.6670), 4-QAM-OFDM is demonstrated to outperform 8-QAM-OFDM and NRZ-OOK in terms of the data rate achieved. Our proof-of-concept demonstration resolves the need for an intermediate signal relay in communication across a wavy water-air-interface, and thus offers a practical approach for future underwater-to-terrestrial internet-of-things implementation. Table 1 shows the comparison of different approaches for across water-air-interface communication.
Table 1. Approaches in across water-air-interface communication
2. Experimental setup
The left part of Fig. 2 shows a photo of the experimental setup used in this study. A UV LED (LG Innotek LEUA35W70RL00) is mounted at the bottom center of the water tank, and it transmits the modulated signal to the air through the water. A pair of plano-convex lenses is placed in front of the LED, making a transmission field of view (FOV) of 25°. The tank is made of acrylic, has dimensions of 35 × 35 × 35 cm, and is filled with emulated clearest sea water with an attenuation coefficient of 0.151 m−1 [28].
Fig. 2 Photo of experimental setup across water-air interface for diffuse-LOS communications, and the OFDM modulation processes.
The water turbidity level is changed by adding an accurate Maalox solution in an orderly fashion [19]. The water depth is 30 cm. A wave generator is installed near the water surface for emulating the dynamic wavy nature of the water-air interface in the real sea environment. An avalanche photodetector (APD) (Thorlabs APD430A2/M) with a bandwidth of 400 MHz is mounted at a height of 60 cm above the water surface.
The right part of Fig. 2 shows the schematics of the OFDM-modulated system setup. At the transmitter end, a series of pseudorandom binary sequences (PRBSs) are generated. After the serial-to-parallel (S/P) conversion, the signal moves through quadrature amplitude modulation (QAM). Each OFDM frame is composed of 180 OFDM data symbols including four training symbols for channel equalization and two extra training symbols for time synchronization. Different numbers of subcarriers are used for data transmission at different bit rates according to the channel conditions.
An inverse fast Fourier transform (IFFT) of length 1024 is performed to convert the signal into time-domain OFDM symbols. To efficiently eliminate the ISI, a cyclic prefix (CP) of 10 points is added in front of each OFDM symbol, and a parallel-to-serial (P/S) conversion is performed before sending the signal to the arbitrary wave generator (AWG, Tektronix AWG 70002A) to perform digital-to-analog (D/A) conversion. The sampling rate of the AWG is set to 250 MSamples/s. The amplitude of the AWG output is clipped within 0.5 Vpp. After that, the signal is amplified by a power amplifier (ZHL-6A-S + ) with a bandwidth of 2.5 kHz–500 MHz to modulate the 250-mW, 365-nm LED.
At the receiver end, the AC signal from APD was sent to a real-time oscilloscope (RTO, Tektronix DPO 72004C) through the same amplifier (ZHL-6A-S + ) and a 98-MHz low-pass filter (Mini Circuits BLP-100 + ) for removing the out-of-band radiation. The sampling rate of the DPO is set to 1.250 GSamples/s. Then, the signal is downloaded to a personal computer (PC) to perform offline analysis using MATLAB. Through resampling, window synchronization, serial-to-parallel (S/P) conversion, CP removal, a fast Fourier transform (FFT), channel estimation and QAM demodulation, and parallel-to-serial (P/S) conversion, the received bitstreams were compared bitwise to calculate the bit error rate (BER).
The NRZ-OOK modulation experiment utilizes a bit-error-rate tester set (ME522A) to generate a pseudorandom binary sequence (PRBS) data stream and measure the real-time bit error ratio (BER). The corresponding eye diagrams are recorded using a digital communications analyzer (Agilent 86100C). The bias and AC amplitude are the same as those used in OFDM.
3. Experimental results and discussion
Figure 3 shows the beam shape of the UV LED, obtained using a beam profiler (Ophir-Spiricon SP620U). It has a round shape with a diameter of 0.4 mm and a Gaussian distribution both on the X- and Y-axes. Therefore, after the light travels through a uniform channel, the Gaussian distribution is expected to be retained along every direction. Hence, in our study, for simplicity, the measurement is performed along one direction (named the X-direction) and then extended in all directions.
Fig. 3 Beam shape and power distribution of the LED source as recorded using Ophir-Spiricon beam profiler.
The power distribution in the air (60 cm above the water surface) formed the illumination coverage area. As shown in Fig. 4(a), the power decreased from 66 μW at the aligned point to almost half when the detection position was off-aligned by 5 cm (X-position = 5 cm). Based on this measurement, we determined the power distribution in the total area, i.e., the illumination area. As shown in Fig. 4(b), an illumination coverage area of 1963 cm2 is observed.
Fig. 4 (a) Received power in the air at 0.6 m above the water surface along the lateral direction from the aligned position (X-position = 0 cm) to off-aligned positions (X-position > 0 cm). (b) Power distribution extended to all directions in the air at 0.6 m above the water surface.
As a figure of merit, the normalized small-signal frequency response of the system is measured as shown in Fig. 5, which takes into account both the transmitter (i.e., the LED) and the receiver (i.e., the APD). Below 10 MHz, the curve is decently flat, which indicates no signal degradation in this frequency region. Furthermore, a −3-dB bandwidth of ~16 MHz and a −10-dB bandwidth of ~43 MHz are measured.
Fig. 5 Small-signal frequency response of the water-air channel. Dashed line indicates −3 dB/-10 dB bandwidth, which is approximately 16 MHz and 43 MHz, respectively.
To study the maximum achievable communication coverage and the corresponding data rate of this channel, we studied the NRZ-OOK performance in this channel. By fixing the APD at the aligned position (X-position = 0 cm), we collected the eye diagrams and measured BERs for different data rates. As shown in Fig. 6(a)–6(f), there are clear open eyes at 23 Mbit/s, and the eye openings decrease with an increase in the data rate to 53 Mbit/s. The corresponding BERs are also measured as shown in Fig. 6(g). It can be seen that a maximum data rate of 55 Mbit/s can be achieved with a BER of the same value at the forward-error-correction (FEC) limit, i.e. 3.8 × 10−3.
Fig. 6 (a)–(f) Eye diagrams at aligned position for data rates achieved by NRZ-OOK of 23 Mbit/s, 33 Mbit/s, 43 Mbit/s, 50 Mbit/s, and 53 Mbit/s, respectively. (g) BERs vs. NRZ-OOK data rate at the aligned position to misaligned positions, i.e. X-position = 0, 2.5 cm, 5 cm. (h) The effective NRZ-OOK communication coverage area and communication blind area.
To study the maximum communication coverage using NRZ-OOK, the APD is then translated along the X-direction from the aligned position to the off-aligned positions in steps of 2.5 cm. As shown in Fig. 6(g), the largest displacement of the APD is 5 cm, and the maximum data rates achieved at X-position = 2.5 cm and X-position = 5 cm are 43 Mbit/s and 11 Mbit/s, respectively. Similar to the power distribution, the maximum achievable data rates along all directions are plotted in Fig. 6(h). From this 2D plot, we observed an effective NRZ-OOK communication coverage of 78.54 cm2. This result suggests that a precise alignment between the transmitter and receiver is not necessary for this configuration. This demonstration elucidates a more practical and realistic approach in cross-medium communication.However, compared to the illumination coverage area, the effective NRZ-OOK communication coverage is much smaller. In other words, there is a communication blind area between the illumination coverage and NRZ-OOK communication coverage, as shown in Fig. 6(h). This is because the detected power on these off-centered area is too low to provide a sufficient SNR for effective communication.
To further improve the data rate, high-spectral-efficiency modulation techniques (OFDM schemes) are utilized. We tested 8-QAM- and 4-QAM-OFDM modulation for this system. Figure 7(a) shows the waveform of the captured 8-QAM-OFDM signal at the aligned position. The spacing (narrow, dipping lines) in the waveform corresponds to two training symbols for timing synchronization. Figure 7(b) is the corresponding spectrum showing a signal occupying a bandwidth of 41.5 MHz before subsiding to the background noise.
Fig. 7 (a) The waveform of captured 8-QAM-OFDM signal in time domain, and (b) the corresponding electrical spectrum of 8-QAM-OFDM with an occupied signal bandwidth of 41.5 MHz.
As the subcarrier spacing is defined as AWG sample rate/IFFT point (250 M / 1024 = 244 kHz), the total number of subcarriers is calculated to be 170. Therefore, for 8-QAM-OFDM over this channel, a gross bit rate of 124.5 Mbit/s is achieved. After removing the overheads of CP, FEC (7%), and training symbols for channel equalization, a net data rate of 111.4 Mbit/s is achieved. From the spectrum, owing to the limitations of the bandwidth, it can be observed that the SNR decreases in the higher-frequency region (from 10 MHz to 41.5 MHz). The corresponding constellation diagrams and BERs for different data rates are provided in Figs. 8(a)–8(c).
Fig. 8 (a)–(c) Constellations for 8-QAM-OFDM signals with net data rates of 91.8, 98.3, and 111.4 Mbit/s at aligned position. (d)–(f) Constellations for 4-QAM-OFDM signals with net data rates of 26.2, 28.4, and 30.2 Mbit/s at X-position = 2.5 cm.
As an alternative to NRZ-OOK modulation, an investigation of the communication coverage of OFDM over the same channel is appealing. A low-order modulation scheme, i.e., 4-QAM-OFDM, was also sent over this channel. Figures 8(d)–8(f) show the constellations for 4-QAM-OFDM at X-position = 2.5 cm at different data rates. It can be observed that more convergent constellations occur at decreased data rates.
The corresponding BERs vs. data rates at different positions and with different modulation schemes were also calculated and are shown in Fig. 9(a). It can be seen that a maximum net data rate of 111.4 Mbit/s is achieved by 8-QAM-OFDM at the aligned position, while only 94.1 Mbit/s is reached by 4-QAM-OFDM. In addition, a highest net data rate of 30.2 Mbit/s is achieved by 4-QAM-OFDM when the X-position = 2.5 cm. We plotted the maximum achievable data rates at different positions in Fig. 9(b). In the OFDM modulation technique, we observed an effective coverage area of 19.63 cm2.
Fig. 9 (a) BER vs. data rate at different positions using OFDM when APD scans from the aligned position to off-center position. (b) The effective OFDM communication area.
Compared to the effective coverage of NRZ-OOK mentioned above, the effective communication area of OFDM is smaller. In other words, the blind communication coverage area increased. However, a higher data rate at the aligned position is achieved by OFDM. Given these observations, we conclude that a higher-order modulation scheme can be utilized to achieve higher data rates in the aligned position, while lower-order modulation schemes should be applied in the off-center positions because they can be accommodated by the low SNR. Hence, adapting the modulation schemes based on the detection position and environment is necessary to achieve the highest possible data rate and largest effective communication coverage.
Lastly, we investigated the channel characteristics in the presence of surface waves. Such waves are expected to introduce the effect of multipath, thus increasing the intersymbol-interference (ISI) to the signals and frequency-selective signal fading. In this study, we created two artificial sinuous waves with wave heights of 5 mm and 15 mm at the water surface, as shown in Fig. 10. The effects of the waves on the signal transmitted are quantitatively characterized by the scintillation index , which is the normalized variance of the irradiance fluctuations. The scintillation index is expressed as [30]
where I is the intensity or irradiance of the optical wave, and the angle brackets 〈 〉 represent a long time average. Using a mixed domain oscilloscope (Tektronix MD03000), we collected 10 k irradiance fluctuation data samples at a sampling rate of 100 samples/s.As shown in Figs. 11(a) and 11(b), we collected the intensity of the received signals under the two wavy conditions within a period of 100 s. It is obvious that with stronger waves, the intensity of the received signal has a larger variance. The acquired intensity histograms for the two waves are provided in Figs. 11(c) and 11(d). Based on the histograms, the scintillation indices are found to be 0.0114 and 0.6670 for the weak and strong waves, respectively. It can also be seen that the Gaussian distribution is more suitable to describe the probability density function of the received intensity for weak waves, while a log-normal distribution better fit that of the stronger waves.
Fig. 11 Collected intensity of received signal during period of 100 s under: (a) weak wave and (b) strong wave. Histogram of normalized received intensity indicating scintillation index for: (c) weak wave and (d) strong wave.
For a communication system in the presence of waves, the amplitude of the received 8-QAM-OFDM signal will fluctuate, as shown in Figs. 12(a)–12(c) that represent the recorded signals at different instances under the same wave generating conditions, and given the notations of small signal-amplitude (SSA), middle signal-amplitude (MSA), and large signal-amplitude (LSA). We calculated the BER for all subcarriers and provided the mean BER for each amplitude-representation. As a larger signal amplitude will naturally give a higher SNR, an effective communication can only be evaluated when the corresponding SSA for a particular wave condition also provides a BER below the FEC limit. Under a weak wave condition ( = 0.0114), we send an 8-QAM-OFDM with a net data rate of 91.77 Mbit/s (140 subcarriers) over this channel. As shown in Figs. 12(d)–12(f), te recorded signals with SSA induces the highest mean BER (3.532 × 10−3), while the lowest BER (4.788 × 10−4) is enabled by the recorded signal with LSA. The signal with the MSA provides a intermediate mean-BER of 1.643 × 10−3. In addition, for each recorded signal, the BER for each subcarrier increases with an increase in the number of subcarriers. This is owing to the limited bandwidth and decreasing SNR offered by the channel.
Fig. 12 Recorded 8-QAM-OFDM signals in the presence of surface waves with varying degrees of amplitude fluctuation: (a) small signal-amplitude, (b) medium signal-amplitude, and (c) large signal-amplitude. Mean BER of each subcarriers for signals with different amplitudes for 8-QAM-OFDM signals: (d) 3.532 × 10−3 for small-signal amplitude, (e) 1.643 × 10−3 for medium signal-amplitude, and (f) 4.788 × 10−4 for large-signal amplitude.
To investigate the optimum modulation scheme for across-medium communication in the presence of waves, we also studied the performance of 4-QAM-OFDM. Figure 13(a) and 13(b) show the BER vs. data rates using 8-QAM-OFDM and 4-QAM-OFDM under weak waves, respectively. As can be observed in Figs. 13(a) and 13(b), all three BERs for signals with SSA, MSA, and LSA are calculated for each data rate. It is apparent that increasing data rates will lead to increased BERs.
Fig. 13 Measured BERs vs. data rates for: (a) 4-QAM-OFDM and (b) 8-QAM-OFDM. (c) NRZ-OOK in the presence of weak waves. Insets are eye diagrams for data rates of: (i) 53 Mbit/s, (ii) 50 Mbit/s, and (iii) 45 Mbit/s.
However, a maximum achievable net data rate of 91.77 Mbit/s for 8-QAM-OFDM is higher than that of 78.66 Mbit/s for 4-QAM-OFDM. This suggests that in the presence of weak waves ( = 0.0114), the higher-order modulation scheme (8-QAM-OFDM) still outperforms the lower-order modulation scheme (4-QAM-OFDM). This can be attributed to the fact that a high SNR is retained in the channel. For comparison, we also sent an NRZ-OOK data stream through the channel. By measuring the mean BERs of different data rates, we observed the highest achievable data rate of 52 Mbit/s with a BER below FEC, as shown in Fig. 13(c). Therefore, we conclude that small waves (wave height < 5 mm in our study) has limited effects on communication across a water-air interface.
However, stronger waves are supposed to degrade the communication performance severely owing to the decreased SNR and the ISI from the multipath effect. It is important to investigate the main contributing factors to the degradation in communication performance. By introducing larger waves ( = 0.6670), we studied the BER performance by using all three different communication schemes. By analyzing the obtained results, we also summarized the highest data rate achieved for calm water, weak waves, and strong waves for NRZ-OOK, 4-QAM-OFDM, and 8-QAM-OFDM. This is shown in Fig. 14.
Fig. 14 Summary of the highest data rates achieved for three modulation schemes under calm water, water with weak waves and water with strong waves.
It is noted that with an increase in the wave intensity, the highest achievable data rates for all three modulation schemes simultaneously decrease. In addition, in the presence of weak waves, 8-QAM-OFDM exhibits a higher achievable data rate than 4-QAM-OFDM and NRZ-OOK. However, in the presence of strong waves, the highest data rate achieved by 8-QAM-OFDM of 39.3 Mbit/s was smaller than the 43.7 Mbit/s by 4-QAM-OFDM. This can be explained by the fact that the waves degrade the SNR, and a lower SNR is required by 4-QAM-OFDM than that required by 8-QAM-OFDM. In addition, data transmissions based on NRZ-OOK scheme are less resilient in the presence of strong waves. This is because the NRZ-OOK scheme is most vulnerable to the severe ISI compared to the OFDM modulation schemes.
Therefore, owing to the lower SNR requirements and the ability to mitigate the ISI, lower-order OFDM modulation schemes are more practical in this situation. This benefits the across-water-air-interface communication link in terms of three key performance matrices: data rates, robustness against waves and communication coverage area. Experiments under the extreme condition, i.e. 2.5-cm misalignment + strong waves, showing an achievable data rate of 17.5 Mbit/s using 4-QAM-OFDM, further prove the practicality of the lower-order OFDM modulation schemes. In future work, adaptive bit loading and power allocation [31] according to the channel performance can further enhance the performance of this system.
Although these experimental results are remarkable in terms of the data rate, robustness to the waves, and communication coverage, they were achieved in a laboratory test bed with limited transmission distance. In real scenarios, a long-transmission-distance channel for the water-air interface is required. Future studies for implementing such a system can leverage adaptive modulation schemes with a high-power UV light source and highly sensitive UV photodetector. In addition, by deploying multiple light sources with a suitable configuration, these schemes are expected to achieve wider coverage. Besides, it is also appealing to investigate some other modulation schemes in such wavy channel. Pulse-amplitude-modulation 4 (PAM-4) [32], is supposed to increase the bandwidth utilization efficiency while requiring less SNR. While most of the studies on optical communications were concentrated on intensity modulation, polarization modulation is also feasible and attractive, despite its inherent complexity, to enhance the system performance in that: (1) it is more resilient to the intensity fading brought by various environmental factors, i.e. waves, oceanic and atmospheric turbulence; (2) the range over which information may be transmitted is substantially greater [33,34].
4. Conclusions
In this work, we established a laboratory test bed to emulate the oceanic environment and studied the optical wireless communication links across wavy water/air channels. A data rate of 111.4 Mbit/s was demonstrated for a diffuse-line-of-sight communication link through a calm water, as well as over tenths of Mbit/s across a wavy water-air-interface, all without the need for a signal relay. The data rate, enhancement of signal coverage, and mitigation of the wave-induced signal fading of the communication system with different modulation schemes were studied. It was found that NRZ-OOK modulation scheme exhibits a large coverage area of ~79 cm2. This will ease the alignment requirement between the transmitter and receiver. By using the OFDM modulation scheme, the cross water-air data transfer showed better resilience to degradation of the communication performance in the presence of dynamic waves movement. Our proof-of-concept demonstration offers a practical approach for potentially eliminating the need for an intermediate signal relay in communication across a wavy water-air-interface.
Funding
King Abdulaziz City for Science and Technology (KACST) Grant KACST TIC R2-FP-008; King Abdullah University of Science and Technology (KAUST) BAS/1/1614-01-01, KCR/1/2081-01-01 and GEN/1/6607-01-01.
References
1. Z. Zeng, S. Fu, H. Zhang, Y. Dong, and J. Cheng, “A Survey of Underwater Optical Wireless Communications,” IEEE Comm. Surv. and Tutor. 19(1), 204–238 (2017). [CrossRef]
2. K. P. Hunt, J. J. Niemeier, and A. Kruger, “RF communications in underwater wireless sensor networks,” in Proceedings of IEEE conference on Electro/Information Technology (IEEE, 2010), pp. 1–6.
3. M. Stojanovic, “Recent advances in high-speed underwater acoustic communications,” IEEE J. Oceanic Eng. 21(2), 125–136 (1996). [CrossRef]
4. F. Hanson and S. Radic, “High bandwidth underwater optical communication,” Appl. Opt. 47(2), 277–283 (2008). [CrossRef] [PubMed]
5. P. Lacovara, “High-Bandwidth Underwater Communications,” Mar. Technol. Soc. J. 42(1), 93–102 (2008). [CrossRef]
6. J. R. Wait, “Project Sanguine,” Science 178(4058), 272–275 (1972). [CrossRef] [PubMed]
7. W. Neubauer, Acoustic Reflection from Surfaces and Shapes (Naval Research Lab, 1986).
8. J. J. Puschell, R. J. Giannaris, and L. Stotts, “The Autonomous Data Optical Relay Experiment: first two way laser communication between an aircraft and submarine,” in Proceedings of IEEE conference on NTC-92: National Telesystems (IEEE, 1992), pp. 14/27–14/30.
9. M. Rhodes and D. Wolfe, “Underwater communications system comprising relay transceiver,” (Google Patents, 2011).
10. F. Tonolini and F. Adib, “Networking across boundaries,” in Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication - SIGCOMM ’18 (ACM, 2018), pp. 117–131. [CrossRef]
11. F. A. Blackmon, L. T. Antonelli, L. E. Estes, and G. Fain, “Laser-based acousto-optic uplink communications technique,” (Google Patents, 2005).
12. N. Farr, A. D. Chave, L. Freitag, J. Preisig, S. N. White, D. Yoerger, and F. Sonnichsen, “Optical Modem Technology for Seafloor Observatories,” in Proceedings of IEEE conference on OCEANS (IEEE, 2006), pp. 1–6. [CrossRef]
13. J. Söderberg, “Free Space Optics in the Czech Wireless Community: Shedding Some Light on the Role of Normativity for User-Initiated Innovations,” Sci. Technol. Human Values 36(4), 423–450 (2011). [CrossRef]
14. Y. Chen, M. Kong, T. Ali, J. Wang, R. Sarwar, J. Han, C. Guo, B. Sun, N. Deng, and J. Xu, “26 m/5.5 Gbps air-water optical wireless communication based on an OFDM-modulated 520-nm laser diode,” Opt. Express 25(13), 14760–14765 (2017). [CrossRef] [PubMed]
15. A. Wang, L. Zhu, Y. Zhao, S. Li, W. Lv, J. Xu, and J. Wang, “Adaptive water-air-water data information transfer using orbital angular momentum,” Opt. Express 26(7), 8669–8678 (2018). [CrossRef] [PubMed]
16. X. Zhu and J. M. Kahn, “Free-Space Optical Communication Through Atmospheric Turbulence Channels,” IEEE Trans. Commun. 50(8), 1293–1300 (2002). [CrossRef]
17. O. Korotkova, N. Farwell, and E. Shchepakina, “Light scintillation in oceanic turbulence,” Wave Random Complex 22(2), 260–266 (2012). [CrossRef]
18. M. S. Islam, M. Younis, and A. Ahmed, “Communication through Air Water Interface Using Multiple Light Sources,” in roceedings of IEEE conference on Communications (IEEE, 2018), pp. 1–6.
19. A. Laux, R. Billmers, L. Mullen, B. Concannon, J. Davis, J. Prentice, and V. Contarino, “The a, b, c s of oceanographic lidar predictions: a significant step toward closing the loop between theory and experiment,” J. Mod. Opt. 49(3–4), 439–451 (2002).
20. H. M. Oubei, X. Sun, T. K. Ng, O. Alkhazragi, M.-S. Alouini, and S. Boon Ooi, “Scintillations of RGB laser beams in weak temperature and salinity-induced oceanic turbulence,” in roceedings of IEEE conference on Underwater Communications and Networking Conference (IEEE, 2018), pp. 1–4. [CrossRef]
21. H. M. Oubei, R. T. ElAfandy, K.-H. Park, T. K. Ng, M.-S. Alouini, and B. S. Ooi, “Performance Evaluation of Underwater Wireless Optical Communications Links in the Presence of Different Air Bubble Populations,” IEEE Photonics J. 9(2), 1–9 (2017). [CrossRef]
22. H. M. Oubei, E. Zedini, R. T. ElAfandy, A. Kammoun, T. K. Ng, M.-S. Alouini, and B. S. Ooi, “Efficient Weibull channel model for salinity induced turbulent underwater wireless optical communications,” in Proceedings of IEEE conference on Opto-Electronics and Communications Conference (OECC) and Photonics Global Conference (PGC) (IEEE, 2017), pp. 1–2. [CrossRef]
23. R. M. Gagliardi and S. Karp, Optical Communications (Wiley, 1976), 445.
24. X. Liu, S. Yi, X. Zhou, Z. Fang, Z.-J. Qiu, L. Hu, C. Cong, L. Zheng, R. Liu, and P. Tian, “34.5 m underwater optical wireless communication with 2.70 Gbps data rate based on a green laser diode with NRZ-OOK modulation,” Opt. Express 25(22), 27937–27947 (2017). [CrossRef] [PubMed]
25. H. M. Oubei, C. Shen, A. Kammoun, E. Zedini, K.-H. Park, X. Sun, G. Liu, C. H. Kang, T. K. Ng, M.-S. Alouini, and B. S. Ooi, “Light based underwater wireless communications,” Jpn. J. Appl. Phys. 57(8S2), 08PA06 (2018). [CrossRef]
26. D. M. Reilly, D. T. Moriarty, and J. A. Maynard, “Unique properties of solar blind ultraviolet communication systems for unattended ground-sensor networks,” Proc. SPIE 5611, 244–255 (2004). [CrossRef]
27. Z. Xu and B. M. Sadler, “Ultraviolet Communications: Potential and State-Of-The-Art,” IEEE Commun. Mag. 46(5), 67–73 (2008). [CrossRef]
28. X. Sun, W. Cai, O. Alkhazragi, E.-N. Ooi, H. He, A. Chaaban, C. Shen, H. M. Oubei, M. Z. M. Khan, T. K. Ng, M.-S. Alouini, and B. S. Ooi, “375-nm ultraviolet-laser based non-line-of-sight underwater optical communication,” Opt. Express 26(10), 12870–12877 (2018). [CrossRef] [PubMed]
29. X. Sun, Z. Zhang, A. Chaaban, T. K. Ng, C. Shen, R. Chen, J. Yan, H. Sun, X. Li, J. Wang, J. Li, M.-S. Alouini, and B. S. Ooi, “71-Mbit/s ultraviolet-B LED communication link based on 8-QAM-OFDM modulation,” Opt. Express 25(19), 23267–23274 (2017). [CrossRef] [PubMed]
30. L. C. Andrews, R. L. Phillips, and Society of Photo-optical Instrumentation Engineers, Laser Beam Propagation through Random Media (SPIE, 2005), 152.
31. X. Huang, S. Chen, Z. Wang, J. Shi, Y. Wang, J. Xiao, and N. Chi, “2.0-Gb/s Visible Light Link Based on Adaptive Bit Allocation OFDM of a Single Phosphorescent White LED,” IEEE Photonics J. 7(5), 1–8 (2015). [CrossRef]
32. C.-Y. Li, H.-H. Lu, W.-S. Tsai, M.-T. Cheng, C.-M. Ho, Y.-C. Wang, Z.-Y. Yang, and D.-Y. Chen, “16 Gb/s PAM4 UWOC system based on 488-nm LD with light injection and optoelectronic feedback techniques,” Opt. Express 25(10), 11598–11605 (2017). [CrossRef] [PubMed]
33. W. Niblack and E. Wolf, “Polarization Modulation and Demodulation of Light,” Appl. Opt. 3(2), 277 (1964). [CrossRef]
34. J. Grosinger, “Investigation of Polarization Modulation in Optical Free Space Communications through the Atmosphere,” Technical University of Vienna (2008).
