Evaluation of Optimal Visible Wavelengths for Free-Space Optical Communications

Abstract

Free-space optical (FSO) communications have emerged as a promising complement to conventional radio-frequency (RF) systems due to their high bandwidth, low interference, and license-free spectrum. Visible-light FSO communication, using laser diodes or LEDs, offers potential for short-range data links, but performance is highly wavelength-dependent under varying atmospheric conditions. This study presents an experimental evaluation of three visible laser diodes at 650 nm (red), 532 nm (green), and 405 nm (violet), focusing on their optical output power, quantum efficiency, and modulation behavior across a range of driving currents and frequencies. A custom laboratory testbed was developed using an Atmega328p microcontroller and a Visual Basic control interface, allowing precise control of current and modulation frequency. A silicon photovoltaic cell was employed as the optical receiver and energy harvester. The results demonstrate that the 650 nm red laser consistently delivers the highest quantum efficiency and optical output, with stable performance across electrical and modulation parameters. These findings support the selection of 650 nm as the most energy-efficient and versatile wavelength for short-range, cost-effective visible-light FSO communication. This work provides experimentally grounded insights to guide wavelength selection in the development of energy-efficient optical wireless systems.

1. Introduction

Free-space optical (FSO) communication transmits data through the atmosphere using laser beams, eliminating the need for physical transmission media such as fibers or cables. With global demand for higher data throughput and improved spectral efficiency rising, FSO systems have become increasingly attractive as both supplements to and replacements for conventional radio-frequency (RF) links, particularly in environments where the RF spectrum is congested or infrastructure deployment is challenging [1,2,3,4,5].

A closely related and rapidly evolving field is visible-light communication (VLC), which operates in the visible portion of the electromagnetic spectrum. VLC has demonstrated considerable promise for short-range applications, including indoor wireless networking (e.g., Li-Fi), building-to-building line-of-sight communication, vehicle-to-vehicle links, and rapidly deployable emergency systems. Its key advantages include the use of an unlicensed spectrum, strong immunity to RF interference, and compatibility with low-cost, readily available visible-light laser diodes capable of supporting high data rates [1,2,3,4,5].

Despite these benefits, the performance of FSO systems is highly susceptible to atmospheric conditions—most notably fog, rain, dust, and turbulence—which degrade signal quality through scattering and absorption [6,7]. Among these impairments, fog poses the most severe challenge, introducing significant optical attenuation that reduces the link margin and transmission distance. Since attenuation effects are wavelength-dependent, the choice of operating wavelength is a critical determinant of system reliability in limited-visibility scenarios.

While infrared (IR) wavelengths have traditionally dominated FSO research due to their favorable long-distance propagation characteristics, visible wavelengths, particularly 650 nm (red), 532 nm (green), and 405 nm (violet), present promising alternatives for short-range, low-cost systems [8,9]. These wavelengths offer high modulation bandwidth and potentially higher quantum efficiency, and they are compatible with widely available silicon-based photodetectors, which can also function as energy-harvesting devices in integrated VLC/FSO designs [10].

Numerous analytical and empirical models have been developed to estimate wavelength-dependent attenuation in foggy environments. However, many of these rely on detailed atmospheric parameters—such as particle size distribution and refractive indices—derived from Mie scattering theory, which are often difficult to obtain or model in practice [1,2,3]. Empirical formulations by Yu, Kruse, and Kim [5,11,12] attempt to correlate attenuation with visibility, while Al Naboulsi et al. [13] introduced refinements by distinguishing between advection fog and radiation fog, each with distinct scattering profiles. Recent studies [6,7,8,9] have also questioned the assumption that longer wavelengths are always superior under all weather conditions. Although these models provide valuable theoretical insight, their applicability to compact, real-world systems is often limited by their complexity and data availability.

In contrast, experimental research [14,15] suggests that visible wavelengths may offer favorable trade-offs in terms of energy efficiency and data throughput, particularly in short-range systems. Various modulation techniques—including amplitude, phase, and pulse schemes—have also been evaluated for FSO applications, with performance dependent on target range and system architecture [16,17]. Additional work on laser diode efficiency [18,19,20] further supports the potential of visible wavelengths for compact, power-conscious optical communication systems. However, direct experimental comparisons of visible wavelengths under identical conditions remain limited.

This study aims to address that gap by designing and implementing a modular laboratory platform capable of evaluating the optical and modulation performance of three visible-wavelength laser diodes—650 nm, 532 nm, and 405 nm—under consistent experimental conditions. The system includes an Atmega328p microcontroller (Microchip Technology Inc., Chandler, AZ, USA) for precision current and frequency control, a Visual Basic-based graphical interface for real-time adjustments, and a silicon photovoltaic cell that serves as both a detector and an energy-harvesting component in the FSO link.

The investigation pursues two core objectives: (1) to experimentally compare the optical power output and quantum efficiency of red, green, and violet laser diodes across varying current levels and modulation frequencies; and (2) to identify the wavelength that delivers the most practical and energy-efficient performance for short-range point-to-point FSO communication.

By providing controlled, side-by-side evaluation of visible wavelengths, this study highlights the significance of wavelength selection in the design of energy-efficient, low-cost optical wireless systems. The findings contribute to the development of compact FSO solutions tailored for indoor and urban environments, where short-range, high-reliability links are essential.

The remainder of this paper is organized as follows: Section 2 presents the theoretical background on atmospheric attenuation, laser diode efficiency, and modulation techniques. Section 3 describes the experimental methodology and testbed architecture. Section 4 details and analyzes the results, including trends in optical power, diode efficiency, and modulation performance, as well as comparing our findings with prior studies and linking them to theoretical trends predicted in atmospheric models. Section 5 concludes the study and outlines directions for future work.

2. Theoretical Background

Although this study is primarily focused on experimental evaluation within a controlled indoor environment, the inclusion of classical fog attenuation models provides a crucial theoretical framework for understanding wavelength-dependent performance in free-space optical (FSO) systems under realistic atmospheric conditions. These models—rooted in Mie scattering theory—offer insight into how fog, haze, and airborne particulates affect optical signal propagation, while also highlighting the limitations of the visibility-based approximations commonly used in FSO link design.

While such atmospheric conditions were not simulated in the current setup, a thorough understanding of these attenuation mechanisms is essential for the design of FSO systems that remain robust under environmental degradation. These models serve as a basis for predicting wavelength-dependent behavior and inform the planning of future experiments involving simulated fog, aerosol dispersion, and extended-range testing in both indoor and outdoor environments.

This theoretical discussion complements the technical focus of our work by situating wavelength selection within a broader atmospheric and photonic context. It also enables the interpretation of our laboratory measurements as baseline performance indicators under clear-air conditions, providing a foundation for future validation of model predictions in dynamic or adverse environments.

Moreover, the performance of any FSO system depends not only on atmospheric factors but also on component-level parameters such as beam attenuation, photon-to-electron conversion efficiency, laser diode characteristics, and modulation behavior. A comprehensive understanding of these interdependent elements is essential for optimizing visible-light FSO links, particularly in emerging applications such as indoor networking, IoT connectivity, and vehicular communication. The subsections that follow present the necessary theoretical constructs for interpreting the experimental results in Section 4.

2.1. Atmospheric Attenuation in FSO Systems

As an optical beam propagates through the atmosphere, it experiences attenuation due to scattering and absorption by air molecules and suspended particles. The total specific attenuation, denoted as ${\gamma }_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}$ (in dB/km), can be expressed as follows [1,2,3]:

$${\gamma }_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}\mathrm{ }={\gamma }_{\mathrm{a}\mathrm{b}\mathrm{s}}+\mathrm{ }{\gamma }_{\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{t}}$$

(1)

where

• ${\gamma }_{\mathrm{a}\mathrm{b}\mathrm{s}}$ accounts for molecular and aerosol absorption;
• ${\gamma }_{\mathrm{s}\mathrm{c}\mathrm{a}\mathrm{t}}$ represents scattering losses due to Rayleigh and Mie scattering mechanisms.

In the visible spectrum, Mie scattering dominates in environments with fog, haze, and smoke, where particle diameters are comparable to the optical wavelength. Mie scattering is wavelength-dependent and generally follows an inverse power-law relationship [1,2,3].

$${\gamma }_{\mathrm{M}\mathrm{i}\mathrm{e}}\propto {\lambda }^{-\mathrm{q}}$$

(2)

Here, $\lambda$ is the wavelength (in nm), and $\mathrm{q}$ is an empirically determined exponent related to particle size distribution, typically ranging from 0.7 to 2.

In contrast, Rayleigh scattering, caused by particles significantly smaller than the wavelength (e.g., air molecules), follows a steeper inverse relationship [1,2,3]:

$${\gamma }_{\mathrm{R}\mathrm{a}\mathrm{y}\mathrm{l}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}}\propto {\lambda }^{-4}$$

(3)

As a result, shorter wavelengths such as 405 nm (violet) are more susceptible to scattering losses compared to longer wavelengths such as 650 nm (red), especially in foggy or dusty atmospheric conditions.

2.2. Fog Attenuation Models in FSO

Fog is one of the most detrimental weather conditions for FSO links due to the high density of water droplets, which cause both scattering and absorption. While Mie theory provides a rigorous approach for modeling such effects, it requires detailed input data—such as the particle size distribution and refractive index—that are typically unavailable or impractical to obtain in real-time applications [6,7,17]. As a result, various empirical models based on visibility have been developed.

Kruse Model

The Kruse model estimates specific attenuation $\gamma$ as a function of wavelength $\lambda$ and visibility $\mathrm{V}$ (in km) [11]:

$$\gamma \left(\lambda \right)=\mathrm{ }{\displaystyle \frac{3.912}{\mathrm{V}}}\mathrm{ }{\left({\displaystyle \frac{\lambda }{550}}\right)}^{-\mathrm{q}}$$

(4)

where $\lambda$ is in nanometers and the exponent $\mathrm{q}$ depends on the visibility range

$$\mathrm{q}=\left\{\begin{array}{cc}\hfill 1.6 ,& \hspace{1em}V>50\hfill \\ \hfill 1.3 ,& \hspace{1em}6<V<50\hfill \\ 0.585{\mathrm{V}}^{{\displaystyle \frac{1}{3}}} ,\hfill & \hspace{1em}V<6\hfill \end{array}\right.$$

(5)

Kim Model

The Kim model refines the Kruse formulation by introducing improved empirical values for $\mathrm{q}$, particularly in dense fog conditions [12]:

$$\mathrm{q}=\left\{\begin{array}{cc}\hfill 1.6 ,& \hspace{1em}V>50\hfill \\ \hfill 1.3 ,& \hspace{1em}6<V<50\hfill \\ \hfill 1.6V+0.34 ,& \hspace{1em}1<V<6\hfill \\ \hfill V-0.5 ,& \hspace{1em}0.5<V<1\hfill \\ \hfill 0 ,& \hspace{1em}V<0.5\hfill \end{array}\right.$$

(6)

This model is considered more accurate in low-visibility regimes.

Al Naboulsi Model

Al Naboulsi et al. proposed models that differentiate between advection fog and radiation fog, based on empirical measurements in varying climatic conditions [13]:

$${\gamma }_{\mathrm{a}\mathrm{d}\mathrm{v}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{ }}=\mathrm{ }{\mathrm{a}}_{\mathrm{a}\mathrm{d}\mathrm{v}}{·.\lambda }^{-{\mathrm{b}}_{\mathrm{a}\mathrm{d}\mathrm{v}}}$$

(7)

$${\gamma }_{\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{ }}=\mathrm{ }{\mathrm{a}}_{\mathrm{r}\mathrm{a}\mathrm{d}}{·\lambda }^{-{\mathrm{b}}_{\mathrm{r}\mathrm{a}\mathrm{d}}}$$

(8)

where $\mathrm{a}$ and $\mathrm{b}$ are empirically derived coefficients reflecting fog type and regional characteristics. These models provide improved accuracy in practical deployment scenarios by accounting for specific fog morphologies.

2.3. Photon Energy and Quantum Efficiency

The energy $\mathrm{E}$ of a photon is inversely proportional to its wavelength [1,2,3]:

$$\mathrm{E}\mathrm{ }={\displaystyle \frac{\mathrm{h}\mathrm{c}}{\lambda }}$$

(9)

where

• $\mathrm{h}=6.626\times {10}^{-34}$ Joule·hertz⁻¹ is Planck’s constant;
• $\mathrm{c}=3.0\times {10}^8$ m/s is the speed of light;
• $\lambda$ is the wavelength in meters.

Shorter wavelengths (e.g., 405 nm) carry more photon energy than longer wavelengths (e.g., 650 nm). While this higher energy can enhance photoelectric conversion efficiency in some applications, it also increases atmospheric scattering and thermal load on optical components, potentially reducing system stability.

The quantum efficiency (QE) of a photodetector—defined as the ratio of generated charge carriers to incident photons—varies with wavelength. Silicon-based detectors, which are the most widely used in visible-light applications, exhibit peak QE in the red-to-near-infrared range (approximately 650–900 nm) [1,2,3,21]. This makes red light particularly well-suited for systems that leverage silicon photocells for both signal detection and energy harvesting.

While QE declines at shorter wavelengths, it remains sufficiently high across the entire visible spectrum to enable effective operation with green and violet light. The widespread adoption of silicon photodetectors is further supported by their low cost, low noise, high availability, and ease of integration with standard electronics. As a result, they are considered the default choice for VLC and short-range FSO applications, especially in experimental or cost-sensitive environments.

2.4. Laser Diode Efficiency

Laser diodes are favored in FSO systems for their compact size, high modulation capability, and optical efficiency. Their performance is commonly described by two parameters:

• Slope Efficiency ${\mathsf{\eta }}_{\mathrm{s}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}}$: Indicates how effectively the diode converts electrical current into optical power beyond the threshold current. It is given by the following [19]:

$${\mathrm{ }\mathsf{\eta }}_{\mathrm{s}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}}={\left({\displaystyle \frac{\mathsf{\Delta }{\mathrm{P}}_{\mathrm{o}\mathrm{p}\mathrm{t}}}{\mathsf{\Delta }\mathrm{I}}}\right)}_{\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{d}}\approx {\displaystyle \frac{{\mathrm{P}}_{\mathrm{o}\mathrm{p}\mathrm{t}}}{\mathrm{I}-{\mathrm{I}}_{\mathrm{t}\mathrm{h}}}}$$

(10)

where

• ${\mathrm{P}}_{\mathrm{o}\mathrm{p}\mathrm{t}}$ is the optical output power (in mW);
• $\mathrm{I}$ is the input current (in mA);
• ${\mathrm{I}}_{\mathrm{t}\mathrm{h}}$ is the threshold current (in mA).

• External Differential Quantum Efficiency (EDQE) ${\mathsf{\eta }}_{\mathrm{E}\mathrm{D}\mathrm{Q}\mathrm{E}}$: Relates the number of emitted photons to the number of injected electrons and is expressed as follows [20]:

$${\mathsf{\eta }}_{\mathrm{E}\mathrm{D}\mathrm{Q}\mathrm{E}}={\displaystyle \frac{\mathsf{\Delta }{\mathrm{P}}_{\mathrm{o}\mathrm{p}\mathrm{t}}/\mathrm{h}\mathrm{v}}{\mathsf{\Delta }\mathrm{I}/\mathrm{e}}}={\mathsf{\eta }}_{\mathrm{s}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}}{\displaystyle \frac{\mathrm{e}}{\mathrm{h}\mathrm{v}}}\approx \left({\displaystyle \frac{\mathrm{e}}{{\mathrm{E}}_{\mathrm{g}}}}\right){\displaystyle \frac{{\mathrm{P}}_{\mathrm{o}\mathrm{p}\mathrm{t}}}{\mathrm{I}-{\mathrm{I}}_{\mathrm{t}\mathrm{h}}}}$$

(11)

where

• $\mathrm{e}{=1.6\times 10}^{-19}\mathrm{C}$ is the elementary charge;
• $\mathrm{h}=6.626\times {10}^{-34}$ Joule·hertz⁻¹ is Planck’s constant;
• $\mathrm{v}$ is the frequency;
• $\mathrm{E}={\displaystyle \frac{\mathrm{h}\mathrm{c}}{\lambda }}$ is the photon energy.

These parameters allow for comparison of laser diodes based on their wavelength-specific energy conversion efficiency and are used in Section 4 to evaluate diode performance.

2.5. Modulation Techniques

Effective data transmission in FSO systems requires that information be encoded onto the optical carrier. Common modulation schemes include amplitude, phase, and frequency modulation, as well as pulse-based techniques such as pulse-width modulation (PWM) and on–off keying (OOK). Each method offers trade-offs in terms of implementation complexity, spectral efficiency, and resilience to noise [16,17,18].

In this study, frequency modulation (FM) was employed due to its robustness and ease of implementation using a microcontroller. The Atmega328p generated a square-wave control signal to vary the modulation frequency between 0.6 MHz and 6 MHz. FM was selected for its low susceptibility to flicker noise and ambient light interference, contributing to stable signal transmission even in the presence of environmental disturbances. These characteristics made FM a practical and reliable choice for our short-range visible-light FSO platform.

3. Methodology

This section details the experimental procedure used to evaluate the optical and electrical performance of visible-wavelength laser diodes in a short-range free-space optical (FSO) system. A custom laboratory testbed was developed using off-the-shelf components and a programmable microcontroller to regulate both current and modulation frequency for each laser diode under test.

3.1. Experimental Setup

The experimental platform was designed to emulate a short-range, point-to-point FSO link over a fixed transmission distance of 1 m. This specific distance was chosen to ensure experimental repeatability, minimize optical alignment errors, and eliminate environmental variables such as beam divergence and mechanical vibration. At this range, it becomes possible to isolate the intrinsic characteristics of each laser diode—such as threshold behavior, quantum efficiency, and modulation response—without interference from atmospheric or path-loss effects. While this configuration does not reflect full-scale FSO deployment, it serves as a baseline reference for diode-level performance under controlled conditions and provides a robust foundation for future, extended-range studies.

The main components of the system are summarized below:

• Laser Diodes: Three single-wavelength sources centered at 650 nm (red), 532 nm (green), and 405 nm (violet) were used as optical transmitters.
• Control Unit: An Atmega328p microcontroller, interfaced with a Visual Basic for Applications (VBA) GUI in Microsoft Excel, enabled real-time adjustment of current and modulation frequency with high precision.
• Photodetector: A silicon photovoltaic cell (VACTEC7–21G72S, Vactec Inc., St. Charles, MO, USA) acted as both the optical receiver and an energy harvester, converting incident light into electrical output.
• Signal Conditioning: The output of the photodetector was amplified using a low-power ICL7611 operational amplifier (Renesas Electronics Corporation, Tokyo, Japan) to match the 0–5 V input range of the microcontroller’s analog-to-digital converter.

This modular hardware architecture enabled precise electrical control and standardized optical evaluation across all diode configurations, ensuring a fair comparison under identical test conditions.

Figure 1 illustrates the block diagram of the complete system and the role of each subsystem in modulation, signal generation, detection, and data acquisition.

The components of the experimental setup shown in Figure 1 are described as follows:

• PC with VBA/Excel Interface: Provides the user interface for setting diode parameters and recording measurements via USB communication with the microcontroller.
• AVR Microcontroller Unit (MCU): Executes control logic, generates digital signals, and orchestrates modulation and current settings based on user input.
• Port IO 2–4: Digital output pins controlling laser diode activation.
• Port IO 8–11: Digital outputs regulating modulation frequency selection.
• Port IO AD0–AD3: Analog outputs for driving current control via DACs.
• Digital-to-Analog Converters (DACs): Convert digital control signals to analog voltages for regulating frequency (via the VCO) and diode current (via the voltage regulator).
• Voltage-Controlled Oscillator (VCO): Produces square-wave modulation signals; frequency is controlled by analog input from the DAC.
• Duty Cycle Control: Adjusts the on/off ratio of the modulation waveform to optimize spectral efficiency and reduce noise susceptibility.
• Voltage Regulator (LM317T, Texas Instruments, Dallas, TX, USA): Converts DAC output voltage into a stable current source for laser diode operation.
• Output Amplifier (ICL7611): Amplifies the regulated signal to ensure diode operation within its optimal power range.
• Laser Diode Module: A switchable module housing three visible laser diodes (650 nm, 532 nm, 405 nm); only one diode operates during each test.
• Photovoltaic Detector: The VACTEC7–21G72S silicon photocell detects modulated optical signals and converts them into measurable voltage output, while also supporting energy harvesting.
• Operational Amplifier (ICL7611, Renesas Electronics Corporation, Tokyo, Japan): Amplifies the photocell output to a readable level for the MCU’s analog input.

This configuration enables repeatable, high-precision measurements of laser diode behavior under dynamic modulation and electrical input conditions.

3.2. Current and Frequency Modulation Control

Two key system parameters—driving current and modulation frequency—were dynamically controlled during testing:

• Driving Current Control: The laser diode current was varied from 50 mA to 200 mA using digital signals from the microcontroller. These were routed through analog voltage regulators (LM317T, Texas Instruments, Dallas, TX, USA) and transistor driver arrays (ULN2003A, Texas Instruments, Dallas, TX, USA). Ports AD0–AD3 managed the digital-to-analog conversion for current regulation.
• Modulation Frequency Control: Square-wave signals were generated using a voltage-controlled oscillator (74S124, Texas Instruments, Dallas, TX, USA) in combination with dual D-type flip-flops (74S74N, Texas Instruments, Dallas, TX, USA) and MOSFET drivers (TC4428ACPA, Microchip Technology Inc., Chandler, AZ, USA). These components enabled frequency tuning from 0.6 MHz to 6 MHz via ports IO8–IO11.

This setup allowed for precise, real-time adjustment of optical signal parameters, enabling characterization of diode response across a broad modulation range.

3.3. Optical Signal Detection and Processing

The modulated optical signal emitted by the laser diode was received by the silicon photocell, which generated a voltage output proportional to the incident light intensity. This signal was then amplified using the ICL7611 operational amplifier before being digitized by the microcontroller’s analog-to-digital converter.

Two primary performance metrics were extracted from the digitized data:

• Optical Output Power (mW): Recorded as a function of both drive current and modulation frequency.
• Laser Diode Efficiency: Derived from the slope of the optical power versus current curve, normalized for electrical input.

To support reproducibility and clarity in signal processing,

Table 1. Specifications of the VACTEC7–21G72S silicon photocell.

Parameter	Value
Detection Area	$2\times 4 \mathrm{c}{\mathrm{m}}^2$
Open-Circuit Voltage	${\mathrm{U}}_{\mathrm{o}\mathrm{c}}=500 \mathrm{m}\mathrm{v}$
Short-Circuit Current	${\mathrm{I}}_{\mathrm{s}\mathrm{c}}=5 \mathrm{m}\mathrm{A}$
Standard Illumination	${\mathrm{E}}_{\mathrm{v}}=600 \mathrm{L}\mathrm{x}$
$\mathrm{Parallel} \mathrm{Cells} ({\mathrm{N}}_1$)	1
$\mathrm{Series} \mathrm{Cells} ({\mathrm{N}}_2$)	1
Load Resistance	${\mathrm{R}}_{\mathrm{h}}=0~1000 \mathsf{\Omega }$
Internal Capacitance	$0.2 \mathsf{\mu }\mathrm{F}$
Operating Temperature	$\mathrm{T}=295 \mathrm{K}$
Field of View	$\mathsf{\Psi }={90}^{\circ }$
Measurement Distance	$\mathrm{d}=1.0 \mathrm{m}$

provides the key physical and electrical parameters of the VACTEC7–21G72S photocell used in the receiver subsystem.

This methodology facilitated accurate and repeatable comparison of the three tested wavelengths, allowing optical power and conversion efficiency to be evaluated under standardized electrical and environmental conditions. The modular and reconfigurable design of the testbed further supports future extensions for longer-range or fog-simulated FSO testing.

4. Results and Discussion

This section presents a detailed evaluation of the experimental results obtained for three visible-wavelength laser diodes—650 nm (red), 532 nm (green), and 405 nm (violet)—tested under identical electrical and environmental conditions. Performance was assessed based on three primary metrics: optical power output, quantum efficiency (QE), and modulation frequency response.

4.1. Optical Power vs. Driving Current

Figure 2 illustrates the measured optical output power as a function of driving current, which was varied from 50 mA to 200 mA while maintaining a fixed modulation frequency of 3 MHz.

Key observations include:

• 650 nm (Red Laser): Demonstrated the highest optical output across the entire current range, with a quasi-linear and monotonic increase beyond the threshold current (~85 mA). This reflects efficient current-to-photon conversion and favorable stimulated emission characteristics.
• 405 nm (Violet Laser): Produced moderate optical output with a more variable slope. Threshold behavior was observed at approximately 100 mA, with greater sensitivity to current fluctuations than the red diode.
• 532 nm (Green Laser): Exhibited the lowest output power and a higher threshold current (~135 mA), indicating greater internal losses and reduced conversion efficiency.

These findings aligned with theoretical expectations related to wavelength-dependent emission efficiency and detector responsivity. The red diode’s spectral alignment with the responsivity peak of silicon detectors and its reduced susceptibility to scattering effects underscore its suitability for short-range FSO applications. Additionally, its smoother and more stable output trend across the 100–180 mA operating range simplifies system calibration and enhances modulation reliability—key attributes for low-power optical links.

4.2. Quantum Efficiency

Quantum efficiency (QE) was calculated using the slope of the optical power versus current curves, as defined in Equation (11) [19]. Figure 3a–c illustrate QE performance for each diode, highlighting significant differences in electrical-to-optical conversion efficiency.

Performance highlights:

• 650 nm (Red Laser): Achieved the highest peak QE at 55.1%, with optimal performance between 90 and 150 mA. Its efficiency curve shows a relatively smooth rise to this peak, followed by a gradual decline at higher currents due to thermal effects and carrier recombination losses.
• 405 nm (Violet Laser): Reached a peak QE of 34.0%, indicating moderate conversion efficiency. It exhibited greater sensitivity to thermal loading, resulting in steeper post-peak degradation.
• 532 nm (Green Laser): Displayed the lowest peak QE (19.9%), consistent with its higher operating threshold and reduced spectral compatibility with silicon photodetectors.

Across all wavelengths, the decline in QE at elevated current levels was attributed to non-radiative recombination, thermal loading, and carrier leakage. These factors reduce the proportion of injected current converted to useful optical output. The results emphasize the importance of current regulation and thermal management, particularly in systems employing continuous-wave operation or high current drive.

The clear advantage of the red laser in both energy conversion and photodetector compatibility supports its use in power-sensitive, visible light-based FSO systems.

4.3. Optical Power vs. Modulation Frequency

To assess frequency response, the modulation frequency was swept from 0.6 MHz to 6.0 MHz while maintaining a constant drive current of 200 mA. Figure 4 depicts optical output versus modulation frequency.

Performance trends:

• 650 nm (Red Laser): Maintained a stable and high optical output across the full frequency range, exhibiting minimal degradation and superior modulation stability. This confirms its suitability for data transmission over a wide range of bit rates with low distortion.
• 405 nm (Violet Laser): Demonstrated good high-frequency response but showed a moderate decline in output beyond 4 MHz. This can be attributed to the increased thermal load, higher threshold current, and reduced QE at elevated switching rates. Although bandwidth performance was acceptable, the diode’s high power consumption limits its viability for battery-powered or IoT applications.
• 532 nm (Green Laser): Exhibited the weakest frequency response, with output power decreasing markedly as frequency increased. A transient enhancement in the 1.8–4.2 MHz range was observed, forming a soft hump in the response curve. This anomaly may have resulted from internal resonances in the modulation circuitry or nonlinearities in the frequency-doubled laser architectures, such as relaxation oscillations or thermal-gain interactions. Further investigation is required to verify this hypothesis.

In summary, while the violet laser supports high-frequency modulation from a bandwidth perspective, its elevated energy demand makes it less practical for energy-constrained applications. The red laser diode offers the most balanced combination of frequency stability, energy efficiency, and photodetector compatibility, making it the optimal choice for visible-light FSO systems in short-range environments.

4.4. Summary of Findings

A comparative summary of key performance metrics is presented in

Table 2. Comparative performance of laser diodes for short-range FSO systems.

Parameter	650 nm (Red)	532 nm (Green)	405 nm (Violet)
Threshold Current (mA)	85	135	100
Max. Optical Power (mW)	High	Low	Moderate
Max. Efficiency (%)	55.1	19.9	34.0
Frequency Stability	High	Poor	Good
Modulation Bandwidth	Excellent	Limited	Suitable

.

Based on these results, the 650 nm red laser diode exhibited the most favorable overall performance across all evaluated criteria. It combined high energy efficiency, a broad modulation bandwidth, and excellent compatibility with silicon photodetectors, making it the most practical candidate for short-range FSO systems. In contrast, the 405 nm violet laser offered adequate modulation capabilities but was hindered by lower energy efficiency and increased thermal sensitivity. The 532 nm green laser underperformed across all categories, exhibiting higher threshold currents, lower output power, and poor modulation stability.

4.4.1. Comparison with Previous Experimental Studies

To contextualize our results, we compared them with the findings of Gaurav et al. (2018) and Abdullah et al. (2018), both of whom evaluated visible-wavelength laser performance in short-range FSO or VLC applications [9,14].

• Gaurav et al. analyzed four wavelengths (1550 nm, 850 nm, 650 nm, 532 nm) under simulated atmospheric conditions, evaluating the quality factor, bit error rate (BER), and received power [9]. The 650 nm wavelength demonstrated a higher Q-factor (4.87) and better BER than 532 nm (Q = 4.42), indicating improved transmission quality and stability.
• Abdullah et al. examined analog voice transmission using 650 nm, 532 nm, and 405 nm diodes in varying atmospheric conditions [14]. Their findings showed that 650 nm consistently delivered the best clarity and efficiency, due to lower scattering and stronger detector alignment. In contrast, 405 nm and 532 nm showed degraded signal clarity, especially under misty or humid conditions.

Our own results align with these conclusions, as shown in

Table 3. Comparative summary of wavelength performance across studies.

Wavelength (nm)	This Work —Max Efficiency (%)	Gaurav et al. (2018) —Q-Factor [9]	Abdullah et al. (2018) —Relative Efficiency [14]	Summary of Observed Behavior
650 nm (Red Laser)	55.1	4.87	Highest (stable day/night performance)	Best optical output and stability
532 nm (Green Laser)	19.9	4.42	Moderate (sensitive to atmospheric changes)	Susceptible to scattering and power degradation
405 nm (Violet Laser)	34.0	Not reported	Lowest (degraded under humidity and haze)	High drive requirement, lower atmospheric resilience

. Despite methodological differences, all three studies confirm that 650 nm offers superior performance in terms of energy efficiency, optical stability, and atmospheric resilience.

4.4.2. Theoretical Consistency with Fog Attenuation Models

Although the present experiments were conducted under controlled indoor conditions without atmospheric interference, the observed wavelength-dependent performance trends align with classical theoretical models of fog attenuation. These include the Kruse, Kim, and Al Naboulsi models, which predict how optical signal attenuation varies with wavelength under different visibility conditions [11,12,13].

The Kruse model, based on the Beer–Lambert law, expresses attenuation as $\gamma \propto {\lambda }^{-\mathrm{q}}$, where $\gamma$ is the attenuation coefficient, $\lambda$ is the wavelength, and $\mathrm{q}$ is a visibility-dependent parameter typically ranging from 0.5 to 1.6 [11]. For example, in moderate fog (visibility V = 2 km), $\mathrm{q}$ ≈ 1.3, and, in lighter fog (V = 6 km), $\mathrm{q}$ ≈ 0.7. Under such conditions, shorter wavelengths like 405 nm experience greater attenuation, while longer wavelengths, such as 650 nm, retain a clear transmission advantage. This trend is strongly reflected in our experimental results, where the 650 nm diode demonstrated the highest optical output power and quantum efficiency compared to the 532 nm and 405 nm.

The Kim model further extends Kruse’s work by refining the range of applicability and showing that longer wavelengths (typically in the 850–1550 nm range) experience reduced attenuation in light fog or haze [12]. Although the model is primarily used in the infrared regime, the attenuation behavior it describes also broadly applies to visible wavelengths, reinforcing the practical advantage of red light in short-range FSO links.

The Al Naboulsi model further distinguishes between advection and radiation fog. In cases of moderate fog, particularly radiation fog, attenuation varies with wavelength and is lowest in the longer visible and near-infrared bands [13]. This again corroborates our observation that the 650 nm diode outperforms the 405 nm and 532 nm sources, especially under conditions simulating moderate atmospheric disturbances.

To visually emphasize this consistency, Figure 5 presents a side-by-side comparison of the normalized experimental quantum efficiency and theoretical attenuation curves based on the Kruse model. The figure plots normalized attenuation curves for visibility values of 2 km (moderate fog) and 6 km (light fog), alongside the experimental data points at 405 nm, 532 nm, and 650 nm. The comparison illustrates a clear inverse relationship: wavelengths with higher theoretical attenuation correspond to lower measured efficiency, and vice versa. As shown, the 650 nm diode achieved the highest efficiency and corresponded to the lowest predicted attenuation, whereas the 405 nm diode experienced both reduced efficiency and higher theoretical losses. This supports the qualitative alignment between fog attenuation theory and practical system performance.

While future work will extend these findings through controlled fog-chamber testing and real-world FSO deployment, the present study establishes a solid, clear-air baseline. The consistency between our experimental findings and theoretical attenuation trends further validates the suitability of 650 nm wavelengths for energy-efficient, short-range FSO communication, particularly in environments with moderate visibility challenges.

5. Conclusions and Future Work

This study presented a comprehensive experimental evaluation of three visible-wavelength laser diodes—650 nm (red), 532 nm (green), and 405 nm (violet)—to determine their suitability for short-range free-space optical (FSO) communication. Utilizing a custom-developed laboratory testbed featuring an Atmega328p microcontroller and a silicon-based photovoltaic detector, each diode’s optical power output, quantum efficiency, and modulation response were systematically assessed under controlled electrical conditions.

The experimental findings demonstrate that the 650 nm red laser diode outperformed its green and violet counterparts across all key performance metrics. It exhibited the highest quantum efficiency, the most stable optical output across the frequency range, and strong alignment with the spectral responsivity of silicon photodetectors. These attributes, along with its lower sensitivity to scattering, establish the red diode as a compelling choice for energy-efficient and reliable short-range FSO systems.

In contrast, the 405 nm violet laser diode offered commendable modulation bandwidth and higher photon energy but was limited by reduced quantum efficiency and increased power consumption. This makes it less favorable for energy-constrained applications, such as battery-powered or IoT-integrated systems. The 532 nm green diode, while prevalent in visual signaling, demonstrated the weakest performance, with lower optical output and efficiency, and is therefore considered suboptimal for data-oriented FSO communication.

Although atmospheric attenuation models (the Kruse, Kim, and Al Naboulsi models) were not directly applied in this phase of the work, their integration provided essential theoretical context. These models will guide future studies aimed at evaluating system behavior under realistic environmental conditions. Planned expansions include testing under simulated fog, haze, and aerosol environments, using fog/smoke chambers and variable particle density modeling to assess attenuation effects more rigorously.

Building on the results of this study, future research will focus on the following directions:

-

Extending the transmission range to 5–10 m in controlled indoor environments (e.g., corridor setups) to examine signal integrity and beam propagation under more practical conditions.

-

Implementing advanced modulation schemes, such as pulse-position modulation (PPM) and orthogonal frequency-division multiplexing (OFDM) to enhance data rates, bandwidth efficiency, and noise resilience.

-

Investigating advanced photodetector architectures capable of simultaneous signal detection and energy harvesting, thereby improving receiver-side energy efficiency.

-

Embedding visible light-based FSO links into Internet of Things (IoT) platforms and Li-Fi networks, enabling scalable, low-power wireless communication for smart infrastructure and edge-computing applications.

-

Developing artificial intelligence (AI) models to predict environmental attenuation (e.g., fog, dust, ambient light), enabling dynamic system optimization and robust performance under varying outdoor conditions.

These future efforts will further advance the development of compact, adaptive, and energy-conscious visible-light FSO systems suitable for a wide range of applications, from smart indoor networking to environmental sensing and resilient outdoor communication infrastructures.