LEDs for Underwater Optical Wireless Communication

Abstract

LEDs are readily controllable and demonstrate rapid switching capabilities. These attributes facilitate their efficient integration across a broad spectrum of applications. Indeed, their inherent versatility renders them ideally suited for diverse sectors, including consumer electronics, traffic signage, automotive technology, and architectural illumination. Furthermore, LEDs serve as effective light sources for applications in spectroscopy, agriculture, pest control, and wireless optical transmission. The capability to choice high-efficiency LED devices with a specified dominant wavelength renders them particularly well-suited for integration into underwater optical communication systems. In this paper, we present the state-of-the-art of Light-Emitting Diodes (LEDs) for use in underwater wireless optical communications (UOWC). In particular, we focus on the challenges posed by water turbidity and evaluate the optimal wavelengths for communication in coastal environments, especially in the presence of chlorophyll or suspended particulate matter. Given the growing development and applications of underwater optical communication, it is crucial that the topic becomes not only a subject of research but also part of the curricula in technical school and universities. To this end, we introduce a simple and cost-effective UOWC system designed for educational purposes. Some tests have been conducted to evaluate the system’s performance, and the results have been reported.

1. Introduction

The facility of control and rapid switching of Light-Emitting Diodes (LEDs), in conjunction with flexible, cost-effective drivers, facilitates their effective integration across a broad range of applications. They are well-suited for diverse sectors—from architectural lighting and automotive systems to consumer electronics and road signage. Moreover, LEDs serve as efficient light sources for agriculture [1], spectroscopy [2], disinfection processes [3,4], and wireless optical communication [5,6,7,8]. The development of electronic devices advanced significantly in 1896 when Guglielmo Marconi eliminated the need for physical wiring in telegraph communications, effectively pioneering wireless communication. The origins of the LED can be traced to the various diode and triode variants that were invented and developed during the early stages of radio technology [9]. In 1907, H. J. Round, a radio pioneer and personal assistant to Marconi, observed the emission of yellow light upon applying a voltage to a piece of silicon carbide (SiC). However, in Ref. [10], Round reported only the experimental observation of this “curious phenomenon” without providing a physical explanation. This “curious phenomenon,” later termed electroluminescence, sparked a series of investigations that ultimately conducted to the development of modern LEDs. Indeed, these early light-emitting devices were subsequently identified as LEDs due to their rectifying current–voltage characteristics. The emitted light resulted from the formation of a rectifying Schottky contact between the SiC crystal and metal electrodes [11]. Round’s discovery did not immediately lead to the development of LED technology, as the underlying physical mechanisms were not yet fully understood. A significant turning point occurred in 1927, when Russian scientist O.V. Losev published a scientific paper detailing the emission of light from a junction diode [12]. He conducted extensive experiments on light emission from semiconductors and developed the first LED prototypes using silicon carbide crystals [13,14,15]. LED technology continued to evolve throughout the 20th century, driven by the contributions of numerous scientists and engineers [16,17,18]. Likewise, LEDs can be used not only as light emitters but also as photosensors [19,20,21,22]. In 2014, the Nobel Prize in Physics was awarded to Isamu Akasaki, Hiroshi Amano, and Shuji Nakamura, “for having invented a new energy-efficient and environment-friendly light source—the blue light-emitting diode (LED)” [23].

Scientific research, industrial and military operations, and recreational activities are driving a steadily increasing demand for underwater communication. Real-time operations and the rapid transmission of large volumes of data—as well as advanced navigation, high-resolution imaging, and modern physical sensors—require moderate-to-high bandwidth. While cables and optical fibers can meet these demands for fixed infrastructures, their deployment is often challenging and may involve costly and impractical installations. Furthermore, wireless communication is crucial for mobile underwater applications, such as scuba diving or the operation of underwater vehicles.

Wireless communication by means of acoustic waves is the principal method for data transmission in marine environments. However, this has several drawbacks due to the inherent physical properties of sound propagation in water [24,25,26,27]. The primary shortcomings include the following:

• Limited bandwidth, resulting in low data transmission rates;
• High latency, resulting in the relatively low speed of sound in water;
• Multipath propagation, where reflections from the sea surface, seabed, and underwater objects cause interference and/or signal distortion;
• Disclosure and jamming vulnerability, as acoustic signals can be detected and interrupted relatively easily;
• High power consumption, as acoustic systems typically require a significant amount of power to operate.

Underwater Optical Wireless Communication (UOWC) is an emerging technology that represents a major innovation for communication and data transfer in underwater environments [28,29,30]. Schematics application areas of Underwater Optical Wireless Communication (UOWC) are illustrated in Figure 1.

Underwater wireless optical communication systems employ both lasers and LEDs as optical sources. Each technology offers benefits and limitations in underwater communication applications.

In clear water conditions with low turbidity and minimal dispersion, lasers are certainly the best solution. The laser beam, naturally with low divergence, has a high-power density. This means that a laser can travel greater distances while still maintaining a high signal-to-noise ratio (SNR). However, the narrow beam requires very precise alignment; even small misalignments can cause significant problems in communication. LEDs, on the other hand, emit a light beam with much greater divergence than lasers. This results in a lower power density at the receiver. Therefore, systems using LEDs as the light source have a shorter range than those using lasers. However, the wider beam makes LED UOWC systems less dependent on perfect alignment, thus simplifying system design and reducing system costs.

When suspended particles are present, as occurs when water is turbid, they cause significant scattering and absorption. In these environmental conditions, the advantages of the narrow, high-intensity laser beam are diminished. The low divergence, which is an advantage in clear water, becomes a disadvantage in turbid conditions, as even small deviations can cause the beam to scatter rapidly. Furthermore, precision alignment becomes even more challenging in dynamic or naturally turbulent underwater environments.

Table 1. Advantages and disadvantages of LASER and LED light source of UOWC systems.

Light Source	Advantages	Drawbacks
LASER	High directionality. Lasers produce a low-divergence beam. This allows for long-distance transmission with low dispersion. High data rate. Due to the coherent nature of the light emitted by lasers, systems using this type of light source can support high-speed data transmission.	Alignment sensitivity. The narrow beam of light, typical of lasers, requires precise alignment between the transmitter and receiver. This makes lasers difficult to use in dynamic underwater environments. Dispersion issues. Particles suspended in the water can cause dispersion, reducing signal quality. High cost. Laser systems tend to be more expensive than LED-based solutions.
LED	Wide Beam. LEDs emit a relatively wide beam of light. This reduces the need for precise alignment between the transmitter and receiver, making it easier to implement systems that can be used in dynamic underwater environments. Cost-effective. LED-based systems are generally cheaper to manufacture and maintain. Resistance to water turbulence. The wider beam is less affected by small movements or disturbances in the water.	Lower data rates. LEDs cannot be modulated at high frequencies. In addition, LEDs are poorly suited to complex modulation schemes. Data rates are generally in the Mbps range. Low communication distance. Due to the lower optical power density (divergent beam), the effective communication range is in the order of 10 m.

LEDs are ideal for short-range applications, especially for applications in moving systems and turbid water. They also enable the development of low-cost systems that require little maintenance. On the other hand, systems that use lasers as the light source perform better in long-range communications and when high bandwidth is required.

summarizes the advantages and disadvantages of each technology [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47].

Unfortunately, the propagation of an optical signal through an underwater channel is significantly affected by absorption and scattering, as can be seen in Figure 2, which have a considerable impact on the communication range and data transmission rate.

Seawater is generally classified into four different types based on the way it scatters and absorbs light: pure seawater, clear ocean water, coastal ocean water, and turbid harbor water [37,48,49,50]. In contrast, clear ocean water contains a variety of dissolved and suspended matter, which further absorbs and scatters light [51,52,53,54,55]. Coastal ocean water is richer in Phytoplankton and suspended particulates, all of which have a strong impact on both the absorption and scattering of light. Turbid harbor water, with its extremely high levels of dissolved and suspended particles, significantly absorbs and scatters light, significantly reducing the range of underwater optical communications [55,56,57,58,59,60,61].

According to the literature [62,63], blue-green light is the dominant information carrier in Underwater Wireless Optical Communication (UWOC) systems. In pure water, red light is absorbed more strongly than blue-green light. However, red light is less affected by scattering caused by suspended particles. In addition are the following two factors:

(1)

red light sources (LEDs and laser diodes) are widely available, featuring a higher modulation bandwidth compared to blue-green light sources, and they are significantly less expensive;

(2)

conventional silicon photodetectors also exhibit higher sensitivity to red light.

These factors help, at least partially, to compensate for the higher absorption of red light in water.

The objective of this paper is to determine the most suitable color of LED light for wireless optical transmission in turbid waters. Although optical transmission in turbid waters can only occur over short distances (a few meters), it is useful in marine environments with poor visibility. For example, underwater operations can disturb muddy seabeds, generating sediment plumes that substantially reduce visibility. Marine exploration and underwater archaeological surveys are frequently carried out in environments characterized by high turbidity. Moreover, the development of underwater monitoring networks for environmental surveillance—particularly in the vicinity of aquaculture facilities or wastewater discharge zones—may benefit from the use of wirelessly transmitted optical signals. Finally, underwater wireless networks that remain functional in turbid conditions can enhance the performance of remotely operated vehicles (ROVs) [64] in tasks involving the detection and retrieval of objects, evidence, or individuals under low-visibility circumstances. Even in comparatively clear waters, robust communication strategies to mitigate attenuation and interference are essential. For example, when monitoring a potential subsea pipeline failure using an autonomous underwater vehicle (AUV), a leak may produce bubbles that hinder wireless communication even over short distances. Figure 3 illustrates a typical scenario where leak-induced bubbles can hinder communication between the sensor and the vehicle.

The main purpose of the paper is to investigate the possibility of transmitting in turbid waters by means of LED light. For this purpose, a simple teaching system has been realized, useful for understanding the feasibility of optical transmission in turbid waters and for training students in wireless optical transmission in water.

2. Absorption and Scattering

Absorption and scattering are the primary factors governing the attenuation and propagation of light in water [65,66,67]. As an optical beam penetrates the aqueous medium, photons interact with both water molecules, dissolved substances and suspended particles, resulting that the radiation intensity reaching the receiver is attenuated. A schematic description of light absorption and scattering in water is shown in Figure 4.

When an optical beam penetrates the aqueous medium, a portion of the photons that compose it are absorbed by the water molecules, suspended particles, and dissolved substances. The absorption coefficient $a\left(\lambda \right)$ quantifies the conversion of photon energy into thermal and/or chemical energy. Numerically, $a\left(\lambda \right)$ it is defined as the fraction of energy absorbed by a beam of light per unit distance traveled through the medium. During transit through the water, photons that survive absorption may undergo scattering. In this process, the direction of photon propagation is altered by particles suspended in the aqueous medium, while their wavelength remains unchanged. The scattering coefficient $b\left(\lambda \right)$ is defined as the fraction of energy scattered per unit distance traveled by a light beam in a scattering medium. The combined effects absorption and scattering give rise to signal attenuation. In other words, absorption and scattering decrease the number of photons collected by the photodetector.

The overall wavelength-dependent signal attenuation, generally indicated with attenuation coefficient $c(\lambda$), is the sum of absorption and scattering coefficients:

$$c\left(\lambda \right)=a\left(\lambda \right)+b\left(\lambda \right).$$

(1)

The attenuation coefficient $c(\lambda$), like the absorption and scattering coefficients, depends on the wavelength of light $(\lambda$) [28]. Figure 5 shows the absorption $a\left(\lambda \right)$ and scattering $b\left(\lambda \right)$ coefficients of pure sea water [68,69,70]. It is noteworthy that the energy loss due to scattering in a visible light beam (400–700 nm) propagating through pure water is practically negligible: $c\left(\lambda \right)\approx a\left(\lambda \right)$. In other words, when propagating through pure water, visible light undergoes only absorption.

On the contrary, in natural seawaters the attenuation coefficient c(λ) is also strongly influenced by the diffusion and absorption of chlorophyll. Figure 6 shows the absorption spectrum of chlorophylls [71,72,73,74,75,76].

Table 2. Absorption and scattering characteristics of different seawater components.

Component	Absorption Coefficient	Scattering Coefficient
Pure water	The absorption of UV, blue and green light is low.	In the range (400–700 nm) can be neglected.
Chlorophyll and humic and fulvic acids	Low value of absorption in the range 550–630 nm.	Can be neglected.
CDOM (Colored Dissolved Organic Matter)	Vary with the concentration of CDOM. The absorption of blue light is strong.	Can be neglected.
Plankton	Vary with the concentration of plankton.	Mie scattering. Vary with the concentration of plankton. Decrease monotonously with λ.
Detritus	Vary with the concentration of detritus. Decrease monotonously with λ.	Mie scattering. Vary with the concentration of detritus. Decrease monotonously with λ.

lists the absorption and scattering coefficients of various seawater components [77].

The attenuation coefficient $c\left(\lambda \right)$ quantifies the total reduction in light intensity during propagation in an aquatic medium, due to both absorption and diffusion. Beer–Lambert’s law can then be used to model signal loss over a given distance $d$ [78,79,80]. The loss along the propagation path $L_p\left(\lambda ,d\right)$, as a function of the attenuation coefficient and the distance $d$, is given by [43]:

$$L_p\left(\lambda ,d\right)=\exp \left[c\left(\lambda \right)·d\right].$$

(2)

In marine waters, the primary contributors to absorption are pure seawater, chlorophyll concentration, and two key components of colored dissolved organic matter: fulvic acid and humic acid. Therefore, for the study of light propagation in water, the absorption coefficient can be expressed as follows:

$$a\left(\lambda \right)=a_w\left(\lambda \right)+a_{cl}\left(\lambda \right)·C_{cl}^{0.602}+a_f\left(\lambda \right)·C_f\exp \left(-k_f\lambda \right)+a_h\left(\lambda \right)·C_f\exp \left(-k_f\lambda \right),$$

(3)

where $a_w\left(\lambda \right)$ denotes the absorption coefficient of pure seawater, $a_{cl}\left(\lambda \right)$ that of chlorophyll, and $a_h\left(\lambda \right)$ and $a_f\left(\lambda \right)$ the absorption coefficients of humic and fulvic acids, respectively [81,82,83].

The coefficients $C_{cl}$, $C_f$, and $C_h$ denote the concentrations of chlorophyll, fulvic acid, and humic acid, respectively. The factors $C_f$ and $C_h$ can be expressed in terms of the chlorophyll concentration as [84,85]:

$$\begin{gathered} C_f=1.74098·C_{cl}·\exp \left(0.12327·C_{cl}\right) \\ {C}_h=0.19334·C_{cl}·\exp \left(0.12343·C_{cl}\right). \end{gathered}$$

(4)

When light travels through natural seawater, in addition to absorption, there is loss of photons to the photodetector due to the phenomenon of scattering. The scattering coefficient $b\left(\lambda \right)$ is generally expressed as [74]:

$$b\left(\lambda \right)=b_w\left(\lambda \right)+b_{small}\left(\lambda \right)·C_{small}+b_{large}\left(\lambda \right)·C_{large},$$

(5)

where $b_w\left(\lambda \right)$ denotes the scattering due to seawater, $b_{small}\left(\lambda \right)$ represents the scattering coefficient of small particles, and $b_{large}\left(\lambda \right)$ the scattering coefficient of large particles. The parameter $C_{small}$ and $C_{large}$ are the concentration of small and large particles, respectively. The scattering coefficients are typically expressed as follows [86]:

$$\begin{gathered} b_w\left(\lambda \right)=0.005826·{\left({\displaystyle \frac{400}{\lambda }}\right)}^{4.322} \\ {b}_{small}\left(\lambda \right)=1.1513·{\left({\displaystyle \frac{400}{\lambda }}\right)}^{1.7} \\ {b}_{large}\left(\lambda \right)=0.3411·{\left({\displaystyle \frac{400}{\lambda }}\right)}^{0.3}. \end{gathered}$$

(6)

Although these Equations (1)–(6) are indispensable for modeling light transmission in water, seawater exhibits a highly heterogeneous mixture of dissolved salts, organic compounds, and suspended particulates. Therefore, to investigate light penetration in marine waters, seawater is categorized using the Jerlov classification scheme [87,88,89,90]. By considering the influence of dissolved and suspended particles on light absorption and scattering, this scheme defines ten water types, ranging from the crystal-clear open-ocean waters (Jerlov I) to the turbid coastal waters (Jerlov 9C).

Table 3. Jerlov water types vs. turbidity.

Open ocean	Jerlov I Jerlov IA Jerlov IB Jerlov II Jerlov III	Very clear Very turbid
Coastal	Jerlov IA Jerlov 3C Jerlov 5C Jerlov 7C Jerlov 9C

presents the various seawater classifications according to Jerlov’s scheme. Jerlov water types are defined by light transmission through a specified thickness and/or by the attenuation coefficient. Figure 7 plots, for each class, the percentage of light transmitted through a fixed path length as a function of wavelength, while Figure 8 presents the corresponding attenuation coefficients.

3. Underwater Optical Transmission in Turbid Water

While underwater wireless optical transmission remains of major scientific interest, its industrial and military applications have driven the development of commercially available systems.

Table 4. Commercially available UOWC systems.

Maker.	Model	Transmitter Type and Used Color	Data Rate (Max)	Operating Range
Hydromea Ref. [91]	LUMA™ X	LED 480 nm (blu)	10 Mbps	max 50 m depending on turbidity
Hydromea Ref. [92]	LUMA™ X-UV	LED 395 nm (UV)	10 Mbps	max 50 m depending on turbidity
Hydromea Ref. [93]	LUMA™ FLEX	LED 475 nm (blu)	500 kbps	max 100 m Values based on calculations. The actual achievable range depends on water conditions such as turbidity and ambient light levels.
Hydromea Ref. [94]	LUMA™ 500ER	LED 475 nm (LED)	500 kbps	greater than 50 m depending on turbidity
Sonardyne Ref. [95]	BlueComm 200	LED 450 nm (blue)	10 Mbps	max 150 m Optimal performance: Perfect for moderate-to-low turbidity dark water (>200 m depth or at night)
Sonardyne Ref. [96]	BlueComm 200 UV	LED 405 nm (UV)	10 Mbps	max 75 m maintains reliable communication even with artificial lighting present
Aquamodem Ref. [97]	AQUAmodem® Op2 (Subsea Optical Modem)	LED Cyan	115.200	Typically 1–2 m

shows the commercially available UOWC devices.

Using blue-green lasers, underwater wireless optical communication can reliably span several hundred meters in clear water [98].

Table 5. Papers related to advances in extending the link length of UWOC laser systems.

Year	Trasmitter	Optical Power	Detector	Distance	Reference
2018	Laser a stato solido (532 nm)	1 mJ (pulsed energy)	SPD	120 m	[99]
2019	Laser Diode (520 nm)	7.25 mW	APD	100 m	[100]
2020	3× Laser Diode (450 nm)	3 × 0.8 W	SiPM	100 m	[101]
2021	Laser Diode (250 nm)	288.4 mW	PMT	200 m	[102]
2021	Laser Diode (450 nm)	293.09 mW	PMT	150 m	[103]
2021	Laser Diode (450 nm) Laser Diode (520 nm)	~1.2 W ~1.0 W	2× PMT	139 m	[104]
2022	Fiber Laser (532 nm)	600 mW	PMT	99 m	[105]
2022	5× Laser Diode (520 nm)	5× 0.7 W (3.5 W)	4× PMT	100 m	[106]
2023	Laser Diode (450 nm)	~1.2 W	PMT	90 m	[107]
2023	Laser a stato solido (532 nm)	1 W	APD	80 m	[108]

SPD—Single-Photon; APD—Avalanche PhotoDiode; PMT—PhotoMultiplier tubes; SiPM—Silicon photomultiplier.

provides a summary of the literature related to advances in extending the link length of UWOC laser systems.

Lasers perform exceptionally well in fixed-infrastructure deployments but are poorly suited to mobile platforms. This limitation currently confines laser-based links almost exclusively to scientific research. For practical, real-world applications, active, continuous alignment systems between transmitter and receiver are required [109,110,111]. At present, this makes these configurations impractical and generally too expensive.

Suspended particles in water scatter light, altering photon trajectories and deforming the information-bearing pulses. Once the beam enters the underwater channel, it experiences not only attenuation but also scattering-driven spreading before reaching the receiver: each pulse exhibits greater rise and fall times and an overall longer duration. This temporal broadening constrains the achievable data rate [112,113,114]. Figure 9 schematically illustrates the effect.

As shown in Table 4, all commercially available UOWC systems utilize LEDs as light sources operating in the blue or UV wavelength range. Moreover, blue or UV radiation penetrates clear water more effectively. The selection of LEDs is driven by the need for systems that can operate under dynamic conditions while enabling easier transmitter–receiver alignment. Additionally, their less concentrated light beams are less susceptible to disturbances caused by suspended particles. Finally, LEDs facilitate the development of robust, long-lasting, and relatively easy-to-control systems.

Figure 7 and Figure 8 show that, in turbid waters (Jerlov 7C and 9C), wavelengths within the 560–630 nm optical band experience the best irradiation transmittance. Furthermore, it is well established that light scattering decreases as wavelength increases [115,116].

Studies on Underwater Wireless Optical Communication (UWOC) primarily focus on the light source, the receivers, signal modulation techniques, and error correction methods. Additionally, research aims to extend communication range and enable transmission in turbid water conditions. Depending on the specific application, certain system aspects may be prioritized over others. Moreover, factors such as system simplicity and cost-effectiveness can also become critical considerations. In this study, we focus on designing a system that enables audio communication between divers. Considering that divers primarily operate in coastal and, in some cases, turbid waters, the system must be ‘turbidity-resistant’ and capable of functioning even when the transmitter/receiver is in motion (albeit at low speed). The system does not require high transmission bandwidth; 40 kHz sampling rate is sufficient for sampling/transmitting audio signals. Additionally, the communication range is relatively short, approximately 5–10 m. Furthermore, since the goal is to design a system suitable for recreational and leisure excursions, it must be robust, simple, and cost-effective. Therefore, for this application, UOWC systems utilizing LED light sources and PIN diodes as photodetectors represent the optimal choice. In highly turbid water, red light attenuates less than blue-green light [62,63]. In turbid water, scattering dominates light attenuation; Figure 7 shows that for Jerlov type 9C water the optimal wavelength is roughly 575 nm.

Today, commercially available high-power red, amber, and yellow LEDs are made of AlInGaP (aluminum indium gallium phosphide). In contrast, the shorter-wavelength LEDs—green, blue, and UV—are based on InGaN (indium gallium nitride) [117,118]. These semiconductors possess a tunable band-gap: adjusting the alloy composition changes the photon energy. In InGaN, increasing the indium content narrows the band-gap, pushing emission from the near-UV (~360 nm) through the visible spectrum to yellow-green (~580 nm). AlInGaP shows the mirror image behavior—raising the aluminum fraction widens the band-gap, allowing emission to shift from deep red (~650 nm) up to yellow (~580 nm) [119]. Unfortunately, current LED manufacturing technologies still suffer from the well-known ‘green gap’ phenomenon [120,121,122,123] (see Figure 10); as a result, high-efficiency LEDs emitting around 575 nm are currently not commercially available.

In UOWC systems, five different types of photodetectors are mainly used: Single Photon Avalanche Diode (SPAD); Silicon PhotoMultiplier (SiPM); PIN photodiode; Avalanche Photodiode (APD); PhotoMultiplier Tubes (PMT).

Table 6. Low-light sensing detector categories.

Feature	SPAD [128,129]	SiPM [130,131]	PIN [132,133]	APD [134,135]	PMT [136,137]
Responsivity	Very High	Very High	Moderate	High	Very High
Bandwidth	Fast	Fast	Fast	Slow	Fast
Power Consumption	Medium	Medium	Very Low	Medium	High
Spectral Range (nm)			400–1100	400–1100	Variable *
Peak Sensitivity (nm)	400–600 nm	420	900	620	Variable *
Dark noise	High	High	Low	Medium	Low
Operative Voltage	Medium	Medium	Low	Medium	High
Cost	Moderate–High	High	Low	Medium	High

* It depends on the material the photocathode is made of.

summarizes the principal characteristics of the various photodetector types used for low-light detection [124,125,126,127,128].

After evaluating the available options, the implementation of a simple and low-cost underwater wireless optical link for audio communication between divers should be implemented using red-amber LEDs as the transmitter and a PIN photodiode as the receiver. Because this study focuses exclusively on audio transmission between divers, we do not address data communication protocols [138,139,140,141]—an interesting topic of research.

4. Test System for Transmission UOWC Study

The main purpose of this paper is to study the possibility of transmitting an audio signal through LED light in turbid water. For this purpose, a simple test system has been realized, useful for understanding the feasibility of optical transmission in turbid waters and for training students in wireless optical transmission in water.

Laboratory activities are a fundamental component of the learning process, particularly in disciplines such as physics and engineering, where theoretical concepts must be both integrated and refined. For this reason, it is essential to design and implement laboratory experiences that explore complex topics, especially when they are presented in a clear and accessible way [142,143,144].

The system is designed to transmit, via Frequency-Shift Keying (FSK) of the optical signal, an audio signal [145]. The carrier used was 40 kHz.

The light sources were made with LED clusters. This allows to obtain, near the source, a light beam with a “wide” width, while at a great distance (10–20 times the size of the source) it becomes equivalent to a point source (naturally wide beam) [146]. This arrangement means that the “scattered” particles near the source have little influence on the propagation of the beam. Obviously, at a great distance the beam is naturally wide so that the scattering particles intercept only a small part of the beam. The use of this type of source, in turbid water, has significant advantages over the use of laser sources.

A 30 W × 50 L × 30 H cm aquarium tank was used to simulate the communication channel. These dimensions were chosen to ensure ease of use in educational settings; the liters of water used allow for relatively easy emptying and refilling. Figure 11 shows the aquarium tank used. During the measurements, a stirrer (home-made) was inserted into the aquarium to keep the suspended particles moving and prevent them from settling on the bottom of the aquarium.

For the experiments, five different light sources were assembled, each incorporating a cluster of high-brightness LEDs. Each LED is capable of emitting an intensity greater than 25 cd within a 20-degree viewing angle. Figure 12 shows the realized light sources along with their corresponding emission spectra. One source consists of 9 blue LEDs, another of 12 green LEDs, a third of 16 amber LEDs, a fourth of 16 red LEDs, and finally one source combines 6 amber LEDs with 9 blue LEDs.

The light sources are designed for use both outside and submerged within aquariums. The aquarium glass does not affect the performance of the system. The ability to submerge the sources in water enables control over the thickness of the water column traversed by the light. Figure 13 shows two light sources immersed in the aquarium.

In this study, we focused on determining the optimal wavelength for transmission in turbid water. The geometry of the light sources and the number of LEDs used were not considered. The only precaution taken was to ensure that all sources emitted light with the same intensity (approximately 200 cd).

The light sources are powered by a driver capable of supplying either a constant or appropriately modulated current. The LED cluster is formed by connecting the devices in series, ensuring that the same current flows through all LEDs. A resistor, selected from a predefined set, is connected to the drain of a MOSFET via a selector. The selected resistor, together with the voltage drop across the diodes, determines the current flowing through the LEDs. By choosing different resistors, it is possible to equalize the light intensity emitted by the different sources. In our experiments, the light sources were configured to emit the same intensity, specifically 200 cd. Since this is primarily an educational system, a 48 V power supply was selected for safety reasons. In practical applications, however, higher voltages can be used, allowing for a greater number of LEDs in the cluster. The system includes two jacks that allow measurement of the current flowing in the LEDs.

Additionally, the driver features a switch that allows the light source to be activated even without modulation. The schematic diagram of the driver is shown in Figure 14.

To assess the feasibility of audio wireless optical communication link between divers, we investigated the possibility of transmitting an audio signal using digital frequency modulation. For this purpose, the circuit shown in Figure 15 was implemented. The output of the frequency-modulated audio signal is fed into the driver (i.e., the input of the circuit shown in Figure 14), which controls the LED cluster. In this configuration, the light emitted by the LEDs is modulated according to the frequency modulation of the input audio signal. This circuit converts an input audio signal (20 Hz–10 kHz) into a frequency-modulated digital output (30–50 kHz). The audio is first amplified by an op-amp; its output then feeds two paths: one into a second op-amp/FET stage that implements automatic gain control to stabilize varying input levels, and the other into a voltage-controlled oscillator (VCO) using a 40 kHz carrier. The VCO’s output is a digital signal whose frequency swings between 30 and 50 kHz in accordance with the input audio.

As a receiving sensor, in the context of this work, a PIN diode was chosen (to be precise, the OSI Optoelectronics PIN-5DI Photodiode). To increase the light-collection area, a 25 mm focal-length, 2.5 cm diameter lens was placed in front of the sensor.

Two different circuits can be connected to the photodiode to enable transduction of the incident light signal.

The first circuit is a classic transimpedance amplifier implemented using two low-noise operational amplifiers. The circuit diagram is shown in Figure 16. This configuration is used for all tests in which the light signal from the source is unmodulated—in other words, for experiments aimed at assessing the attenuation of the optical signal caused by water and dissolved or dispersed substances. The circuit also includes a digital output, which can be used to receive and decode modulated signals.

The second circuit is able to decode the frequency-modulated audio signal. In other words, it decodes the modulated signal via the system shown in Figure 15. The electrical diagram of this second light decoder is shown in Figure 17.

In the circuit shown in Figure 17, the first stage (OP-AMP 1) is a transimpedance amplifier that converts the photocurrent generated by the PIN photodiode into a voltage. The two subsequent operational amplifiers (OP-AMP 2 and OP-AMP 3) form an active band-pass filter, designed to pass the 40 kHz carrier frequency used by the transmitter. The fourth op amp (OP-AMP 4) is used as a squaring amplifier. The output of this amplifier is sent to a Phase-Locked Loop (PLL) circuit. In our system, the PLL is designed with a center frequency of 40 kHz. When the PLL locks onto the 40 kHz carrier frequency (used for modulation), the demodulated audio signal is produced at the output. This signal is then passed through a buffer stage and a low-pass filter before being sent to an audio amplifier. The entire circuit draws a current of 40 mA in the absence of modulation, and approximately 150 mA during modulation at half output volume.

5. Experimental Tests

Since our goal is to assess the suitability of LEDs for transmitting information even in turbid water, we first investigated how the light signal is attenuated in the presence of scattering particles.

Using the four custom-built monochromatic light sources, and adjusting the drivers to produce the same intensity at the detector in clean water, we introduced varying amounts of clay into the aquarium. As expected, the measurements show (seen Figure 18) that attenuation of red light is lower than that of blue light. In other words, the results indicate that the loss of light for scattering decreases as the wavelength of the light increases.

Another particularly noteworthy result is that, using the previously described system, audio transmission occurs reliably even at clay concentrations of 40 mg/l when the red-light source is used, whereas correct audio transmission is not achieved with the blue light source. When conducting experiments with students, it is particularly instructive to demonstrate how increasing clay concentration leads to the disappearance of the audio signal when using blue light, while the signal remains clearly audible with red light. Red light transmission remains effective even in highly turbid water. Figure 19 illustrates the propagation of red light under such conditions.

Figure 20a–c show the waveform applied to drive LED sources and the corresponding signal received at the test point of the circuit shown in Figure 17, for a light source consisting of 16 red LEDs. Figure 20a refers to transmission in pure water while Figure 20b refers to transmission in turbid water. Once the carrier is locked by the PLL module (the CD4046 IC in Figure 17), the receiver digitizes the signal—rendering it independent of the received-signal amplitude—so long as the signal-to-noise ratio remains above 20 dB (see Figure 20c). Both transmission and reception employ FM modulation, and the receiver extracts only the time-varying component that is phase-locked to the carrier. This architecture makes the system immune to ambient light, although it is still important to prevent the receiver from being saturated by excessive background illumination. Finally, Figure 20d shows a 2 kHz waveform (yellow trace) simulating a pure acoustic tone used as input to the system implemented with 16 red LEDs. This signal, transmitted through the optical channel formed by turbid water, is compared (blue trace) with the received and demodulated signal from the circuit in Figure 17. When the PLL is locked correctly, the “reconstructed” acoustic signal has a signal-to-noise ratio consistently above 35 dB and total harmonic distortion remains below 1%.

The previous tests were conducted with both the light source and the detector positioned outside the water-filled container. Subsequently, by immersing the light sources in water and varying both the optical path length and the turbidity, we determined the attenuation coefficient c and the maximum propagation distance that still enables reliable audio transmission for each turbidity condition and wavelength used.

Measurements were carried out in different water conditions: clean water, salt water (simulating clear seawater), turbid sea water simulated with salt plus Maalox, and turbid water simulated with salt plus Maalox [147,148] plus chlorophyll.

Table 7. Attenuation coefficients vs. light wavelength and turbidity.

	Pure Water	Salt Water (35 g/L)	Salt Water (35 g/L) + Maalox (3 × 10−3%)	Salt Water (35 g/L) + Maalox (3 × 10−3%) + Chlorophyll (12 mg/m3)
Light source made with 9 blue LEDs (~470 nm)
Attenuation coefficient [m−1]	0.011	0.016	1.8	4.8
Propagation length [m]	~25	~18	~1.1	~0.3
Light source made with 12 Green LEDs (~520 nm)
Attenuation coefficient [m−1]	0.033	0.048	2.2	3.9
Propagation length [m]	~16	~13	~1.3	~0.5
Light source made with 16 Amber LEDs (~610 nm)
Attenuation coefficient [m−1]	0.26	0.29	1.2	3.1
Propagation length [m]	~5	~4	~2.0	~1.0
Light source made with 16 Red LEDs (~635 nm)
Attenuation coefficient [m−1]	0.30	0.33	1.2	2.8
Propagation length [m]	~5	~5	~2.0	~1.2
Light source made with 9 blue LEDs (~470 nm) + Light source made with 8 Amber LEDs (~610 nm)
Attenuation coefficient [m−1]	0.011	0.016	1.1	3.1
Propagation length [m]	~25	~18	~2.0	~1.0

summarizes the results obtained.

As shown in Table 7, the attenuation coefficient in clean water is lower for blue light. Conversely, in turbid water, red light exhibits lower attenuation. The maximum transmission distance was estimated assuming all light sources emit an intensity of 200 cd, with a light collection area of approximately 5 cm², and taking into account the photodiode’s response curve and dark current. Finally, we tested audio signal transmission with FSK modulation. Under all operating conditions, the audio signal was successfully transmitted using red, amber, and amber + blue light sources.

To assess the system’s robustness to transmission errors, we measured the Bit Error Rate (BER) under various operating conditions. These measurements were performed without audio modulation. Using a dual-channel counter, we compared the carrier pulses sent to the LED driver (Figure 14) with those received by the transimpedance amplifier (Figure 16). Figure 21 presents the BER results for different turbidity levels using the light sources shown in Figure 12.

Furthermore, using the power transmitted in pure water as a reference, we determined the power loss of the transmitted optical signal for each light source as a function of the different types of water. Figure 22 illustrates the power loss in dB for each source and water type listed in Table 7.

6. Future Directions

To implement a diver-usable communication system in real-world conditions, the research will progress toward the development of a full-duplex optical communication system. Two distinct wavelengths in the amber/red band (610 nm and 635 nm) will be employed. One wavelength will be used for transmission in one direction, the other in the opposite direction. Figure 23 illustrates an example of full-duplex wireless optical communication between two devices using these wavelengths.

The system proposed in Figure 23 is composed of two different types of transmitting–receiving stations. One station (in the figure called Tx_{(635 nm)}–Rx_{(610 nm)}) transmits with light radiation at 635 nm and is able to receive radiation at 610 nm. The other station (in the figure called Tx_{(610 nm)}–Rx_{(635 nm)}) transmits with light radiation at 610 nm and is able to receive radiation at 635 nm. Each Tx-Rx station is made up, for the transmitting part, of a cluster of LEDs.

The receiving part is instead made up of the following:

(1)

A narrow band optical filter centered on the wavelength of the light radiation to be “decoded”. This filter must have a pass band of, at most, 20 nm.

(2)

A lens to collect the light and collimate towards the sensor surface. This lens is used to intercept a light surface larger than the sensitive area of the detector.

(3)

Quadrant photodiode. The position of the divers will “hardly” remain fixed during transmission. Therefore, an “automatic” system will be needed to keep the line of sight between the two communicating divers “relatively” fixed. It is therefore planned to use a quadrant detector that allows to understand how to move the Tx–Rx station to maximize the received signal. The transmitter–receiver head can be adjusted using piezoelectric actuators to optimize alignment. This allows for proper alignment with minimal energy consumption. Since high-precision alignment is not required, the same photodiode can be used for both signal reception and alignment purposes [149,150,151].

(4)

If the alignment system is not sufficient to establish a reliable connection between the receiver and the transmitter, the diver may use a handheld flashlight-like transmitter/receiver. To initiate communication when no connection is present, the diver points the device directly at the intended recipient.

Obviously, the system can only communicate correctly between two divers. In case of communication between a group of divers, it is necessary to implement a “communication network” (see Figure 24).

To ensure effective communication in all directions, a set of transceiver units can be mounted on a spherical cap structure positioned on the diver’s head. This configuration enables semi-omnidirectional communication (see Figure 25).

To achieve omnidirectional communication, a system similar to the one proposed in Refs. [152,153] can be implemented. The system could consist of six/eight sets of optical communication modules, each of which is composed of an LED source and a photodetector. The modules can be positioned on the diver’s suit (for example, one on the back, one on the front, two/four on the legs and two on the head).

7. Discussion

In light of the growing research interest in underwater optical wireless communications, we identified the need for an accessible, low-cost educational platform to support hands-on learning and experimentation in this field. Such a tool can enhance student engagement, foster a deeper understanding of system-level challenges, and potentially inspire innovation among future researchers. Tools of this kind are not commercially available and have not been documented in the literature. The existing literature predominantly focuses on assessing maximum transmission ranges and achieving high data rates in underwater optical communication. Conversely, there is a notable lack of studies addressing transmission performance in turbid water conditions.

In this paper, we have focused the study on audio transmission because it is highly impressive for educational purposes. This also allowed us to investigate how emerging technologies can address real-world challenges. The communication between divers, for example, is essential for both recreational diving and specialized professional activities. Divers currently communicate by hand signals or acoustic intercoms. Acoustic links are easily detected, can disturb marine life, and draw significant power. Optical communications, by contrast, are energy-efficient, have minimal impact on marine fauna, and are far less detectable. However, in coastal environments they must still perform reliably in turbid waters.

8. Conclusions

The test system realized and the related preliminary tests carried out in the laboratory have demonstrated that there is a concrete possibility of creating a wireless optical transmission system that can be used for “audio” communication between divers. In particular, it is possible to create a system capable of communicating, even in very turbid murky waters, using red light.

Furthermore, the system serves as an excellent educational tool, enabling experimentation with various types of light sources and receivers. The configurations presented in this work represent only a few of the possible setups. For example, different types of receivers can be employed, and a greater number of LEDs can be used. Additionally, alternative modulation techniques for audio transmission can be explored. Thanks to the system’s compact size, the water and dissolved substances can be easily replaced, making it especially suitable for educational applications.