An FSO System Based on Mirrors for Early Warning of Frost: Primary Analysis

Abstract

This paper proposes a free-space optical (FSO) system for early warning and detection of natural disasters. The system consists of multiple sensor nodes equipped with mirrors, motors, and controllers. Under normal weather conditions, signals are transmitted between nodes. However, in the presence of frost, signal reflection is directed to a base station (BS) by adjusting the mirrors’ orientation based on the reflection angle determined using Snell’s law. By monitoring the round-trip time of the signal to the BS, the frost-affected node can be identified. The power received at various nodes is analyzed, considering reflections from mirrors and path attenuation. The results indicate that sufficient power levels can be achieved for six nodes, covering an area of approximately 314.16 km². The total time required to send an alarm signal to the BS is calculated and compared with the systems proposed in the literature. The proposed system demonstrates a time reduction of up to 69.7% compared to systems where the signal traverses all nodes before reaching the BS and a 7% reduction compared to systems employing dedicated transmitters and receivers. The proposed system is easy to deploy in a crop area with a fast response and relatively low power consumption, making it an efficient solution for early warning of frost. To the author’s best knowledge, the proposed system is the first one to exploit mirrors in free-space optics for early warning of frost. The analysis presented in this paper is very helpful for further investigations into using mirrors in FSO systems for early warning and notification of natural disasters such as frost.

1. Introduction

Early warning systems serve as vital tools in minimizing the negative impact of various natural and man-made disasters. These systems are designed to detect disasters such as frost damage and wildfire at an early stage and send warnings to receiving units, which helps in taking optimum actions at appropriate times. While disasters such as earthquakes and wildfires are considered devastating, others can also have a considerable effect. For example, frost damage poses significant risks to crops, particularly in temperate and subtropical regions such as Jordan, where frost damages large amounts of crops with yearly financial losses exceeding tens of millions [1]. Early warning systems utilize a set of technologies to detect and send alarm signals at an early stage. Free-space optics could be considered a new approach offering great prospects. Speed and quality are the two major fundamentals when it comes to disaster countermeasures [2]. Free-space optical (FSO) communication is designed for high-speed data transmission via light instead of air in the wired and wireless methods [3]. Due to the principles of using lasers or light-emitting diodes (LEDs) in areas with no obstacles, like space, a large amount of data can be transferred at high speeds with low latency. Therefore, it is quite valuable for modern communication systems [4,5]. In addition, FSO systems are immune to electromagnetic interference, providing a secure and reliable communication solution in environments where radiofrequency communication may be vulnerable or restricted [6,7]. Hence, FSO technology is very appealing for early identification and warning of natural disasters.

Several systems for early warning or management of natural disasters based on FSO were developed and proposed. In [8], a new emergency communication system using free-space optical (FSO) technology is proposed, suitable for disaster-stricken areas where traditional communication infrastructure may be compromised. That system employed full-duplex millimeter-wave (mm-wave) technology and optical relays via unmanned aerial vehicles (UAVs) to establish high-speed communication links. Key features include utilizing optical combs and Laguerre–Gaussian modes for data transmission, generating 60 GHz mm-wave signals, and implementing a single-hop scheme to overcome line-of-sight challenges. In [9], communication in natural disasters using a Fuzzy-Based FSO System, which was dependent on sensors and the Internet of Things to transfer data and management of disaster detection. In [10], FSO relays were mainly exploited for network recovery in scenarios like natural disasters or attacks, leveraging FSO’s wireless nature and high bandwidth.

The challenge of optimal relay placement, which is also crucial due to FSO’s weather-dependent links, was also tackled in that work. The problem was formulated as an integer linear program (ILP), considering link availability predictions and fairness to existing traffic flows. Using multiple-input multiple-output-free-space optical (MIMO-FSO) networks in [11] to improve transmission capacity in natural disaster cases by polarization division multiplexing (PDM). In the proposed MIMO-FSO, a single wavelength is supported by two independent data channels with orthogonal polarization states.

Several papers have studied FSO systems based on mirrors to enhance optical signal transmission in different applications. In [12], the described optical communication system was designed for data retrieval from remote sensors and identification over several hundred-meter ranges. The main component of this system was a modulated micro-electro-mechanical systems (MEMS) mirror, which switches between a flat reflective and a corrugated diffractive state, to increase the effectiveness of signal modulation. To overcome beam distortion induced by turbulent atmospheric conditions, the authors in [13] proposed a beam-relay approach. This system included a downward-directed laser from an orbiting relay mirror that served as a reference for atmospheric correction. To reduce the atmospheric distortions, the station on the ground used adaptive optics to control the uplink beam. Then the orbiting mirror directed the refined beam to a distant satellite, which increased the reliability of communication. In [14,15], a mirror diversity receiver (MDR) for the 2 × 2 multiple-input multiple-output (MIMO) visible light communication (VLC) systems was studied to reduce problems in the parallel data stream decoding. By using MDR to block light from one direction and increasing reception from another. The MDR with two photodetectors (PDs) was discussed in [14], while the MDR with multiple PDs was studied in [15], where the MDR was in the middle of four PDs and the MDR was in the middle of six PDs.

Free-space optical communication has evolved and developed as a high-capacity solution competing with radio frequency links, especially in situations where the spectrum is limited. Since 2020, the FSO systems have shown good maturity levels under challenging conditions, with many studies on their actual tolerance capacity. The study in [16] investigated the effect of optical aberrations and optimal segmented Gaussian beams in inter-satellite FSO communications. Models were developed to evaluate the FSO channel under different effects, such as joint jitter and misalignment pointing errors. The results were very helpful in designing an FSO-based satellite network by determining the required hardware performance levels as well as the number of satellites and orbital levels needed.

A mathematical model was proposed in [17] to assess and evaluate basic performance metrics such as average bit error and outage probabilities, and reliability on intra-satellite FSO links under transmitter pointing jitter and optical aberrations. The study specified optimal non-aberrated truncated Gaussian beams for different telescope architectures. It also showed that system performance is very sensitive to the optimal far-field irradiance. By introducing Seidel aberrations, the analysis revealed that coma aberration has the greatest impact due to its induced apparent angle-of-arrival shift. Monte Carlo simulations further demonstrated that the sensitivity of communication performance to aberration magnitude strongly depends on the telescope architecture.

Other models were also developed in [18], in which a newly developed combined stochastic gradient descent optimization algorithm aimed at compensating for optical distortions was explored. The algorithm exhibited linear time and space complexity. That model demonstrated low sensitivity to variations in input parameters, simplicity, speed, and power enhancement at the same time.

Similarly, a turbulence compensation algorithm for FSO was proposed in [19]. Fast-guiding mirror (FSM) and a deformable mirror (DM) were used to reduce turbulent front-wave errors as an effective adaptive and dynamic alignment method, which is highly needed in modern FSO systems.

Mirrors were also used in [20] for real-time wavefront correction. They were employed dynamically to reorient optical paths for latency-sensitive applications.

Free-space optical (FSO) communication based on mirrors provides a power-efficient solution for different applications. Unlike traditional satellite sensing systems that require sophisticated transceivers, mirrors can redirect optical beams passively. In contrast, space-borne remote Earth sensing systems require substantial continuous power to support high-gain radio frequency (RF) or optical transmitters, imaging payloads, onboard storage, and real-time communications with ground segments. Despite the modern solar panel improvements, power should be allocated not only for sensing functions but also for orbit maintenance and other tasks [21,22]. Moreover, the long distance of communication requires higher transmission energy per bit. This results in orders of magnitude lower operational power consumption compared with satellite-based remote sensing approaches [22,23].

Aircraft and unmanned aerial systems (UAV) equipped with specialized optical monitoring systems represent another well-established approach to remote inspection and sensing. However, they have limited endurance due to power constraints [24,25].

Therefore, the FSO system with passive mirrors provides a unique operational advantage. The system requires minimal operational energy after deployment, relies on ground-based transmitters rather than power-hungry mobile platforms. Such characteristics make reflective-relay FSO architectures especially suitable for long-term monitoring applications such as early notification of frost. Systems based on mirrors also provide a simpler solution, less or easier access for maintenance, and swifter response.

While such systems effectively utilize mirrors in controlling beam directions and improving the relevant communication process, none have specifically addressed the use of FSO-based mirrors for early warning or management of natural disasters. In comparison with intelligent reflecting surfaces (IRS) that were already proposed for early warning of frost [26], mirrors feature simplicity, low cost, and do not need external fields to control the angle of reflection.

Mirrors can offer a set of features for early warning systems as follows:

• Mirrors can redirect light signals without requiring additional energy for signal amplification or transmission [27,28]. This is essential for the continuous operation of the system in critical situations.
• In addition, mirrors retransmit optical signals faster than other devices that receive, process, and retransmit optical signals [29]. This provides a quick response to instantly redirect signals to base stations or other critical nodes when a disaster is detected.
• Wide coverage enabled by reflecting signals across long distances or around obstacles such as mountains or buildings [30].
• Mirrors can withstand harsh environments and conditions associated with natural disasters by utilizing protective coating and cleaning techniques [31].

Hence, a system of sensor nodes equipped with mirrors is proposed in this paper.

The system directs the signal from one node to another when no frost exists and to an external base station from the affected node. The mirror is directed at angles either to the next node or to the BS. The sensor nodes are arranged in a mesh topology covering an area resembling a crop field, which might be affected by frost. The time taken to detect the natural disaster at the affected node is calculated, indicating that the proposed system is faster than systems in which the signal needs to be transmitted through all nodes before it is sent back to the base station or others, based on the transmitter and receiver. The power received at each node is calculated, taking possible reflections from the mirror and attenuation through the path in between.

The contribution of this work can be summarized as follows:

• Mirrors have been rarely investigated for early warning systems in the literature. The proposed investigations in this paper indicate the effectiveness of mirrors in transmitting the alarm signal swiftly while reducing the overall power consumption.
• The proposed system achieves up to a 69.7% reduction in time compared to systems where the signal must traverse all nodes to reach the base station (BS) and a 7% reduction compared to systems using dedicated transmitters and receivers.
• The power received at each node is determined by accounting for attenuation for clear conditions and reflections introduced by the mirror. A link margin for a higher attenuation level is also calculated through link budget estimation. These calculations are crucial for gaining insights into real-world implementation scenarios of early warning systems.

This paper is arranged as follows: In Section 2, the system layout and principle of operation are presented and discussed. In Section 3, the analytical model, including mathematical equations and control methods involving algorithms for reflecting angle adjustment based on weather conditions, is presented and explained. In Section 4, the simulation results are discussed and analyzed. The paper is finally concluded in Section 5.

2. Layout and Principle of Operation

The proposed early warning system aims to identify and provide early warnings for natural disasters at an early stage. It comprises several sensor nodes, with each node equipped with a sensor, mirror, motor, and controller, as illustrated in Figure 1. In clear weather conditions and the absence of natural disasters, the signal is transmitted from one sensor node to another. However, in the presence of natural disasters, the signal is reflected to the base station. The direction of reflection is determined by the orientation of the mirror.

The weather conditions are evaluated at each node. Upon reaching the first node in the system, the sensors assess the weather conditions of its area. Using a control algorithm, the controller selects the reflection angle to adjust the direction of the mirror accordingly. This adjustment is made with the assistance of the motor, which influences the phase shift in the reflected signal either towards the next sensor node or back to the base station. It is worth indicating that several sensor types can be used in this case, including optical sensors in a single FSO system [32] and/or RF antenna sensors in which data are optimized for the controller in a hybrid FSO/RF system [33,34].

The layout distribution of nodes in the proposed system is shown in Figure 2. Mesh topology is selected for the following reasons:

• Reliability: Mesh topology is highly reliable by providing multiple communication paths between nodes, ensuring network continuity even if some nodes or links fail [35].
• Scalability: This methodology is suitable for increased coverage areas when additional nodes are added without impacting the network performance [36].
• Efficient communication: Mesh topology can reduce delays and response time by allowing direct or shortest-path communication between nodes [37].

All the above-mentioned reasons make this topology very appealing for early warning systems.

Planar mirrors are used to reflect the optical signal in nodes based on Snell’s law [38] as shown in Figure 3 and Equation (1).

$${\theta }_i={\theta }_r$$

(1)

where ${\theta }_i$ is the incidence angle about the incidence signal and ${\theta }_r$ is the reflecting angle about the reflecting signal.

The mirror can be oriented in a way that controls the signal reflected from it. The reflected signal makes an angle with the normal to the mirror plane, which is double that the mirror makes with the horizontal plane at which it was initially oriented. The change in the angle of reflection of the incident signal is performed with reference to (Equation (2)). The direction of the mirror is changed by an angle that is double the angle of reflection. This change does not affect the efficiency of the system in terms of increasing the time required for the signal to be reflected or increasing the losses. This process is indicated in Figure 4 [39,40].

$${\Phi }_{si}={2\Phi }_{sr}$$

(2)

where ${\Phi }_{si}$ is the mirror angle shift, ${\Phi }_{sr}$ is the reflecting angle shift about the reflecting signal.

3. The Analytical Model

In this section, the simulations and equations are discussed to evaluate an FSO system for disaster detection, focusing on time of arrival and received power using mathematical equations and MATLAB programming.

3.1. The Evaluating Parameters

In this section, the system model indicated in Figure 5 is simulated using MATLAB programming (MATLAB R2023b) [41]. Two important parameters are evaluated via simulations: the received power and time of arrival. Both parameters are important to ensure the continuity of the system in critical situations, and when the resources are limited, the time of arrival parameter indicates the effectiveness of the system in detecting the natural disaster at an early stage.

3.1.1. Time of Arrival and Coverage Area

The distribution of mirrors as nodes is shown in Figure 5. The distance between every two nodes and between each node and the BS is assumed to be 10 km, as indicated in

Table 1. Table distance in the proposed FSO network.

Elements (from–to)	Distance (km)
BS to node 1	$R$ = 10
Each node to the BS	$R$ = 10
Each node to the next node	$L$ = 10

. Using the sensed data, a decision is made to ascertain the presence or absence of natural disasters by using a control algorithm with a principle illustrated in Figure 6. If no natural disasters are detected, the signal proceeds to the next node; however, if natural disasters are detected, the signal is redirected back to the base station (BS). This process iterates at each node. Another node, designated as node T, has been incorporated into the system to introduce a time differential when frost is detected in the vicinity of node 5. Although the distance between node 5 and node T is minimal, it results in a time variance in the signal reaching the BS. When frost is present at node 5, the signal is transmitted directly to the BS. Conversely, in the absence of frost, the signal is routed through node T before reaching the BS.

It is worth noting that mirrors will be used only to direct the signal for notification purposes through the system, and sensor antennas will be used for the frost detection process, such as in [34].

The specific node affected by frost is determined based on the time required for the signal to return to the BS, as defined by the following speed equation (Equation (3)) [42].

$$D=V\times T$$

(3)

$$D=2R+\sum _{n^{\prime }=1}^{n^{\prime }}\left(n^{\prime }-1\right)L$$

(4)

where $D$ in meters is the distance traveled from the BS and back to it, $V$ in meters per second is the signal velocity, which is equal to free space velocity ($3\times {10}^8 \mathrm{m}/\mathrm{s}$), $n^{\prime }$ is the affected nth node, and $T$ in seconds is the signal’s round-trip time to the BS.

By analyzing the signal’s round-trip time to the BS and factoring in the estimated propagation velocity, we can determine the distance traveled. This measurement is then cross-referenced with the known distance between the node and the BS within the system. Consequently, we can pinpoint the location of the affected node.

Based on the layout distribution of nodes shown in Figure 5, the system’s coverage area is calculated considering a circular shape and using Equation (5) [43]. The distance from each node point to the BS is assumed to be equal, as well as the distance between every two successive points in the network, as outlined in Table 1.

$$A_c=\pi R^2$$

(5)

where $A_c$ in meters (m²) is the area of a circle which represents the coverage area of the proposed network as indicated above, R in meters (m) is the radius of the circle which represents the distance between the node and BS, and $\pi$ is approximately equal to 3.14. With reference to the above equation, a coverage area of 314.16 ${\mathrm{k}\mathrm{m}}^2$ was calculated. This area is large enough to cover a crop area that is potentially affected by frost damage.

It is worth noting that optical path alignment is always maintained between mirrors. In reality, this may be achieved by an acquisition, tracking, and pointing (ATP) system on one or both ends of the FSO link [44].

3.1.2. Power

Equation (6) delineates the calculation for received power, accounting for channel losses [45,46], which considers factors such as transmission aperture channel losses, beam divergence, atmospheric attenuation, and received aperture.

Table 2. The system sample values.

Sample	Value	Unit
d1 = d2	1	m
Pt	100	mW
B	10−9 near to ideal case	mrad
R	10	km
α	0.013	dB/km
λ	1550	nm
Pr	−80	dBm

provides all values of parameters for received signal power and link margin calculations.

$$P_r=P_t\ast {\displaystyle \frac{d_2^2}{{\left[d_1+\left(B\ast R\right)\right]}^2}}\ast {10}^{\left(-\alpha \ast R/10\right)}$$

(6)

where $P_r$ (W) is the received power, $P_t$ (W) is the transmitted power, $d_1$ is the transmission aperture diameter (m), $d_2$ is the receive aperture diameter (m), B is the beam divergence (mrad), and $\alpha$ is the atmospheric attenuation factor (dB/Km).

At the first node, the received power is the transmitted power by the BS after being subjected to losses due to the distance between the first node and the BS. At the second node, the received power is the transmitted power from the first node after being subjected to path losses and reflection losses introduced by the first mirror, and so on.

4. Results and Analysis

At first, the BS transmits a signal to node 1. If there is a natural disaster in node 1’s area, the signal returns to the BS; otherwise, it is directed to node 2. Upon the signal’s return to the BS, the distance traveled is determined using Equation (3) to identify the natural disaster-affected areas.

4.1. Power Results

Figure 7 illustrates the power distribution at each node, influenced by various losses, including reflection loss: contingent upon the material composition of the mirrors, typically ranging from 2% to 5% of the incident signal power [47]. For this system, a reflection loss of 5% for the mirrors is assumed, indicating that each reflection results in a 5% power loss when the total transmission power from the BS is 100mW, as depicted in Figure 7a. The amount of power lost due to reflection in the proposed system at node 6 is 27 mW, compared to 40 mW when calculating all the power lost due to reflection and propagation. The power lost due to reflection represents 32.5% of the power lost in the system, and to reduce this value, materials characterized by small losses, such as silver, can be used, as they exhibit losses of only 2% of the incident power. The effect of reflection losses of different mirror materials is depicted in Figure 7b [48]. Additionally, transmitted signals incur losses due to transmission through the channel, comprising geometric and atmospheric losses. Geometric loss is contingent upon beam divergence. The amount of geometric loss due to beam divergence in the proposed system at node 6 is 40.2 mW when beam divergence is equal to 10⁻⁹, compared to 96 mW when beam divergence is equal to 0.01.

The effect of beam divergence is indicated in Figure 8. Depending on the power values in Figure 8, this loss can be mitigated by reducing the beam divergence values as much as possible and making the signal narrower. Atmospheric losses, on the other hand, are influenced by atmospheric conditions such as humidity and visibility. This research assumes clear atmospheric conditions. As the number of nodes increases, the power loss problem becomes more complex, so optical amplification is added to some nodes in the system in order to boost the optical signal periodically.

To account for additional losses that may affect the performance, a link margin (Margin) is calculated using Equations (7) and (8). Calculations are conducted between every two nodes with a beam divergence of 1 × 10^−9.

$$P_r=P_t-nL-Margin$$

(7)

$$L=10{log}_{10}\left(\alpha R\right)$$

(8)

L is the attenuation loss in dB, n is the n-th sensor node, which is taken as six for this case. $P_r$ is the receiver sensitivity, which is assumed to be −80 dBm [49], $P_t$ is the transmitted power in dBm. $\alpha$ is assumed to be 4 dB/km in this case, which represents the attenuation constant of haze and fog that are common with frost [46]. R is kept as 10 km. In this case, a link margin of −3.876 dB is calculated.

4.2. Time of Arrival Results

In this section, a comparison between the proposed system and other systems regarding their response time to the presence of natural disasters is presented. In the proposed system, the signal travels directly from the natural disaster-affected node to the BS via mirrors.

Table 3. The total time of frost detection for the proposed FSO network.

If Natural Disasters in	Total Transmission Distance (km)	Time (µsec)
Node 1	20	66.7
Node 2	30	100
Node 3	40	133
Node 4	50	166.8
Node 5	60	200
Node T	60.1	200.3

displays the results of distance and time calculations for each point in the event of a natural disaster. This time is considered short enough for a quick detection process, as is explained in the following section.

5. Comparison with Other Systems Proposed in the Literature

The time taken for this system and others proposed in the literature is compared with the following two systems:

• System (1): A system in which the signal undergoes re-transmission from the affected node through a transmission system comprising a receiver and a transmitter [50]. This system is shown in Figure 9a.
• System (2): A system in which all nodes within the network are inspected, following which the signal is relayed back to the BS by the last node in the network [51]. This system is shown in Figure 9b. The signal needs to pass through all the nodes in succession to reach the last node, which is the only node in the system that can send the signal to the BS.

In both systems, the signal is received, processed, and retransmitted through 3 µsec [52], and the distance between nodes is the same for the proposed system. Also, mesh topology is considered for these two systems.

The comparison results are shown in Figure 10. The results in the figure indicate that the proposed system offers the swiftest response compared to the other systems. This is attributed to its elimination of the processing delay incurred in the second system, as well as its capability for direct transmission from any node within the system to the BS. As expected, the time increases while passing through more nodes for the proposed system and which uses Tx and Rx at nodes. For system (2), all nodes within the network are inspected before the signal is sent back to the BS. Hence, a fixed time of 215 s is noticed for 5 nodes. In such a system, the alarm signal must pass through the five nodes even if the disaster is detected at the first node. This is 69.7% longer than the time required to notify the BS about frost detected at the first node for the proposed system. The proposed system also outperforms system (1) by 7% when using five nodes. An additional time is always needed in both of these systems to receive, process, and transmit signals. Such time is not needed in the case of our proposed system, as stated above.

It is worth noting that achieving and maintaining perfect alignment between mirrors over 10 km distances is challenging. Minor vibrations, thermal expansion, or wind could easily misalign the beams, disrupting communication. To address the alignment issue, the system can utilize a low-power guide beam or beacon signal, as demonstrated in high-capacity FSO links [53], to help maintain precise alignment between nodes over long distances. Implementing an active beam-tracking subsystem would greatly enhance link reliability and ensure continuous communication performance, even under the atmospheric disturbances common during frost or fog. A sensor with a control algorithm can also be used to detect the spot offset, and alignment can be obtained through closed-loop control, as in [54].

6. Conclusions and Future Work

In this paper, a system based on mirrors for early warning and identification of natural disasters is proposed. By leveraging Snell’s law and a sophisticated reflection angle and mirror direction controller algorithm, the system achieves swift and accurate detection of frost-affected areas. Simulation results validate the system’s efficiency, demonstrating its capability to cover a significant area while accurately pinpointing frost-affected nodes. Compared to a system that relies on consecutive signal transmission between nodes or on transceivers between nodes, the proposed approach exhibits superior response time (over 69.7% and 7% faster), direct signal transmission capabilities, and reduced power consumption, making it a promising solution for early frost detection and warning systems.

While the system has a set of appealing and promising characteristics, the following points will be considered and elaborated in the future:

• The investigations in this paper considered a static refraction index, which is not the real case, as scattering and turbulence are expected in frost conditions. Models for adaptive refractive indices will be studied and used.
• The turbulence and low temperature effect will be studied and evaluated in a chamber mimicking the frost conditions. Results will be analyzed for different modulation types and locations of the turbulence effects between transmitter and receiver.
• Different pointing methods of the mirrors will be investigated. Dynamic solutions adapted by the mirrors will be explored and developed.
• Deployment and calibration of mirrors and other practical issues will be studied and investigated.