1. Introduction
The use of unmanned aerial vehicle (UAV) communications could make fifth-generation (5G) and sixth-generation (6G) wireless mobile communication more powerful by combining networks in space, the air, the ground, and the sea [1,2]. Achieving ubiquitous and uninterrupted wireless communication necessitates the adoption of novel methods. Among these methods is the integration of UAVs into the networks to provide coverage from aerial positioning [3]. Because of their excellent mobility, UAVs are well suited for building emergency communication services that prioritize ground sensor nodes and users, and they effectively help catastrophe victims in rugged mountainous terrains [4,5]. As a result, UAV-based emergency communication systems are gaining popularity as revolutionary ways of improving disaster response and ensuring safety in hostile circumstances [6,7,8]. Coordination and management, including wireless communication engineers, police, firefighters, and medical personnel, are critical for effective emergency response.
Identifying emergency responses in disaster zones is vital, as is maintaining good signal quality with minimal latency [9]. For situational awareness, high-quality video feeds must be sent to central control continuously in real time. Bidirectional, high-bandwidth wireless communication that sends real-time 3D data to respond is what makes this possible and offers adaptability across a wide range of cellular network applications, including access to distant locations as flying base stations employing lightweight gear [10,11]. The rapid development of mobile and wireless communication technologies has overtaken spectrum availability for cellular networks. The sub-6 GHz frequency bands, widely utilized by most mobile technologies, are becoming insufficient to satisfy the capacity needs of future communication networks [12]. As a result, a shift towards mmWave frequency bands with their capacity-enhancing larger bandwidths has been seen.
The mmWave spectrum, within the frequency range of 30–300 GHz, is rapidly being considered feasible for future wireless communication systems, with a particular interest in 5G and beyond [13,14,15]. This interest arises from its ability to support more incredible data speeds and increased capacity. However, the frequency is challenged by significant propagation loss, lower coherence time due to rapid channel fluctuation, superior power consumption in analog-to-digital conversion, increased sensitivity to radio-wave blockage, and low-power amplifier efficiency [16]. The propagation constraints inherent in mmWave communications shadowing, larger Doppler spreads, increased penetration loss, and increased air absorption pose significant challenges, particularly in NLOS channels, limiting their terrestrial mobile application [17]. In response, it is recommended that mmWave and microwave technologies be integrated to improve user experiences. Developing wireless networks that accommodate the projected surge in broadband wireless connectivity poses noteworthy difficulties. The dimensions of large-scale broadband wireless networks requires system designers to fulfill multiple objectives [3]. The exploration of the mmWave spectrum is considered in 5G and beyond wireless systems. This promise stems from its ability to support significant data rates and enhanced network capabilities.
In the midst of all this, the dynamic landscape of wireless communication, as well as the challenges of network planning for mountainous and suburban locations, present distinct issues. This gap highlights the necessity for empirical research on signal intensity, interference difficulties, and the complex signal environments seen in these difficult terrains. Consequently, this paper aims to close the gaps, and the contributions to this work can be expressed as follows:
1.. Investigation of channel propagation characteristics in mountainous situations across four frequency bands (3.5, 6, 28 and 60 GHz) using Ray-Tracing method (RT). It focuses on path loss, angle and delay characteristics, phase shift, and losses caused by concrete, weather, and foliage. UAVs of varying heights were utilized to investigate the effects of fading and reflection, offering critical insights into signal behavior in tough terrains. The Received Signal Strength (RSS) models were derived based on a custom channel for each frequency.
2.. Investigation of regional variability in field signal strength and identification of the best UAV height for the ideal user coverage. It explores the features of mmWave channels in complicated mountainous terrain using the 3D RT technique.
In order to maximize services in difficult terrains, using multiple UAVs as flexible base stations can be advantageous. This work, however, considers the deployment of a single UAV to determine the optimal operational conditions under different constraints. The proposed models are general and can be extended to multiple UAV deployments.
The rest of the paper’s structure is as follows: Section 2 provides a thorough analysis of the literature, focusing on propagation modeling in mountainous areas and the prospects of Ray-Tracing in 5G and beyond. Section 3 describes the system model and elaborates on the analysis methodology to derive the targeted models. Section 4 presents and examines the performance outcomes and considers future works. Finally, Section 5 concludes the study.
2. Literature Review
2.1. Wireless Channel Behaviours in Mountainous Areas
Mountainous terrains frequently contain dense forests or thickets that can function as natural obstacles for wireless signals, resulting in substantial signal attenuation. Although elevated mountain positions have the potential to provide extended signal coverage, the presence of dense vegetation in these areas may diminish the advantages. Additionally, it is worth noting that the vegetation in mountainous regions experiences seasonal variations, particularly in areas with deciduous trees [18]. In addition, the average signal level also changed from month to month, with a seasonal variation of the most significant size in the winter months. The monthly average values of seasonal changes in signal levels were up to 4–5 dB [19]. Thus, these variations can result in changes in signal propagation characteristics at various times of the year in Pakistan. According to [20], an extensive measurement campaign in West Harlem, New York City, involved over 2000 link measurements across seven diverse buildings. The constructed path loss model revealed an average of 30 dB excess loss over free space beyond 50 m. The glass type emerged as the dominant factor in outdoor to indoor loss, with a noticeable 20 dB difference between scenarios featuring low- and high-loss glass. Additionally, variations in floor height were found to impact signal strength by 5 to 10 dB.
In contrast, suburban environments exhibit a combination of residential properties accompanied by gardens, community parks, and streets adorned with trees, resulting in a scattered distribution of vegetation. When higher frequencies, which are more susceptible to foliage interference, are involved, these vegetation regions might cause signal attenuation. Design is strongly emphasized in suburban landscapes, which often leads to implementing planned greenery and ornamental vegetation [21]. However, this can pose challenges for maintaining consistent signal propagation. The proximity of some suburban regions to woodlands or natural reserves introduces an additional level of intricacy. The diverse range of infrastructure, including single-family homes and multi-story buildings and the integration of natural foliage and constructions, creates a highly intricate signal environment. Consequently, it necessitates a meticulous approach to wireless network planning. The limited body of research specifically addresses the unique challenges and opportunities presented by mountainous scenarios, which hold practical significance. In complex mountainous environments, UAV-assisted sensor networks can facilitate critical operations such as fire warnings, rescue missions, surveillance, detection, and communication relays [22].
2.2. Navigating Multipath Challenges in Mountainous UAV Communication
In complicated mountainous scenarios, a UAV-assisted emergency communications network architecture is designed to address the challenges of diverse terrains and complex reflections. The network consists of a transmitter (Tx) represented by the UAV and a receiver (Rx) on the ground. The rugged and uneven terrain in mountainous regions gives rise to multiple reflections, causing multipath fading and adding intricacies to the communication link. The author of [4] provides directional propagation data at 60 GHz from outside to inside at the American College of Greece, with over 2.4 million power measurements collected from four runs covering up to 220 m. The results show that the Root Mean Square (RMS) errors are less than 2.5 dB, the excess loss is more than 30 dB beyond 20 m, and the azimuthal gain degradation is 7.5 dB for 90th percentile links.
It is essential to take into account all potential multipath types in order to effectively describe the UAV air-to-ground (A2G) channel in these difficult conditions. There are three primary parts to the communication link. The first part is the LOS path, which is the direct line of communication between the UAV and the ground. This path typically has the strongest signal because there are not any large obstacles in the way. The second component is the Single Bounce (SB) path, where the received signal reaches the ground end after a single reflection. The SB component can be further classified into two categories in mountainous environments. One arises from scattering due to ground reflections near the Rx, while the other originates from reflections off the mountains. These reflections can introduce additional delays and affect the overall signal quality. The third component is the Multiple Bounce (MB) path, where the received signal experiences the effects of multiple environmental characteristics during propagation. For instance, the signal might reach the Rx after reflecting off a mountainous area and undergoing another reflection from the ground. The MB power is often weaker and less easily detected by the Rx. The complex and dynamic mountainous terrain is critical in shaping the UAV A2G channel characteristics. Understanding and accurately modeling these multipath effects are essential for designing reliable and efficient communication systems in such challenging scenarios. Considering all possible reflections and path types, the UAV-assisted emergency communications network architecture can be optimized for effective and robust communication in mountainous regions during devastating disasters or emergencies.
2.3. Propagation Modeling for Mountainous Area
UAVs in complex mountainous landscapes exhibit distinct channel characteristics, such as obstructed paths due to mountains, unpredictable 3D flight paths, and dynamic UAV nodes [23]. Hence, it is imperative to comprehend the channel characteristics to utilize UAV-assisted emergency communication systems effectively in such scenarios. The research on UAV channel modeling can be broadly classified into two main categories: measurement-based empirical channel models and simulation-based deterministic or stochastic channel models [24,25]. Measurement systems based on time domain analysis have been employed to track flight paths in mountainous landscapes. However, these studies have primarily concentrated on frequency bands below 6 GHz in mountainous terrains rather than mountainous regions [26]. It was also necessary to utilize aircraft of considerable size to accommodate the measurement equipment, as it was unsuitable for low-altitude UAV communications. Furthermore, the plane did not incorporate mmWave technology [27].
The RT method has demonstrated exceptional performance in deterministic channel modeling for UAV communications, taking into thorough consideration the environmental conditions [6,28]. Nevertheless, applying this method in complex mountainous scenarios is subject to certain limitations. These limitations arise from the intricate environments, challenging experimental conditions, and high equipment costs. In contrast, stochastic channel modeling utilizes geometry-based stochastic models (GBSMs) [29]. Several studies have presented and examined different models based on GBSM. These models include a non-stationary maritime UAV-to-ship GBSM [30], MIMO channel models [31], and a 3D non-stationary GBSM for a UAV-to-vehicle mmWave beamforming channel [32]. The models consider factors such as 3D UAV trajectories and time-varying parameters to simulate movement. However, it should be noted that these models are primarily designed for urban scenarios and may not adequately account for the distinct features and challenges present in mountainous areas. Several UAV communication channel models, such as the third-generation partnership project standard, have been proposed. However, similarly, the models have not specifically addressed complex mountainous and suburban scenarios [33].
2.4. Foliage Losses Effects
Factors influencing the penetration loss include the mmWave signal frequency, antenna polarization, angle of incidence, surface roughness, material permittivity, and material thickness [34]. Foliage attenuation is a critical factor in mmWave communications. The presence of vegetation between the Tx and Rx introduces additional signal loss, potentially adversely affecting the Signal Strength of a wireless communication system. Foliage attenuation intensity depends on the vegetation density, with single trees having less impact than multiple trees or forests [35]. Compared to 3.5 and 6 GHz frequencies, mmWave frequencies exhibit higher penetration loss and struggle to penetrate most solid materials, including walls, doors, and furniture. Consequently, mmWave communications can be easily obstructed, particularly in densely built suburban environments with numerous buildings and people [36].
When a wireless signal propagates across a vegetation-covered region, the total propagation loss, PlTl (dB), may be calculated as [37,38]. Natural landforms substantially influence radio wave propagation [39,40]; for example, densely wooded locations provide difficulties such as multipath fading and NLOS channels, separating them from metropolitan situations [41]. Geographical characteristics peculiar to various places, such as terrain, vegetation density, and climate, exhibit variable degrees of impact on transmission [42]. Propagation models must consider these aspects to ensure accuracy, as inaccurate models might decrease network coverage. Researchers have developed channel models for forest settings. The authors of [43] investigated roadside woodland foliage and attenuation loss and discovered that it might be enhanced by up to 20 dB. They also investigated how the 28 GHz mmWave signal propagated through coniferous forests, resulting in more accurate automated site-specific models. An experimental method is proposed to calculate the average loss per meter of vegetation depth for various vegetation types. The average losses at 140 GHz range from 0.2 dB/m in open forests to 9.8 dB/m in dense forests [44]. Radio wave propagation in forests is classified as direct, reflected, and lateral waves [9]. Direct and reflected waves via vegetation suffer greater loss owing to dispersion and absorption. Lateral waves propagate over tree crowns, and it has been found that at 900 MHz, vegetation is the major propagation mechanism within 1000 m [45]. A modified exponential decay (MED) model, derived from the 1960s decay model, is the most commonly used to model the phenomena. The model can be represented as follows:
(1)
where A, B, and C are coefficients of amplitude, frequency, and distance dependency, respectively. f is the frequency, whereas D is the distance of propagation through the vegetation (depth). The models are adapted from empirical measurements and modified based on a proposal by Weissberger in the frequency range of 200 MHz to 95 GHz [34]. Additional models like FITU-R and COST 235 are tuned to specific frequency bands. The exponential decay models are empirical and not constrained by fundamental propagation mechanisms, as noted by Seville and Craig [38]. Table 1 summarizes the difference between A, B, C, D, and f values taken into consideration by the different models.2.5. Non-Line-of-Sight (NLOS) Propagation
NLOS propagation between Tx and Rx primarily results from refraction and scattering. In this context, transmitted signals can still reach the receiver due to reflections from nearby objects, bending, or diffraction. Reflection and scattering occur when obstacles between TX and RX are more significant than the propagating signal wavelength. Consequently, short wavelength wave signals experience increased shadowing and reflection and reduced diffraction [52]. The short wavelength of mmWave signals causes reflecting surfaces to appear rougher, resulting in increased signal dispersion and reduced specular or direct reflection. Consequently, diffuse reflection leads to lower received transmission power than specular reflection.
This article comprehensively investigates signal intensity and propagation characteristics over a broad spectrum of frequencies. The investigation can show the spectrum of signal characteristics, covering frequencies from 3.5 GHz to 60 GHz. One notable characteristic is the regulated number of reflections used in the imaging technique, whereby a maximum of two reflections is often allowed, apart from the shooting and reflections [53]. This latter approach was used for RT throughout the research. The modeling methodology incorporates significant variables such as buildings, differences in weather conditions, and a sizable vegetation cover measuring 5 m in depth. Enabling a comprehensive evaluation of the power received under different circumstances, this study enhances the understanding of the interaction between signals and their surrounding environment, along with the ramifications across various frequencies. These findings provide valuable insights that may be used to optimize wireless communication systems.
Table 2 presents a synopsis of selected prior efforts on mmWave measurement studies that typically necessitate specialized channel sounders and are categorized based on type: outdoor [9,16,23,25,28], frequency range, mountainous, and suburban environments. The measurements have been used to explore various environments, including suburban [3,9], and rural [33] mmWave deployment scenarios.
This research investigates how wireless communication works in rural mountain areas and suburbs that are exposed to different types of environmental challenges, which have not been studied much compared to the cities. The interest is to study how signals move at different frequencies and how the weather affects them, enabling specific model characterization for each frequency.
3. Methodology
In this study, MATLAB’s 2023a (Bangi, Malaysia) advanced 3D RT toolbox is used. The primary objective of this investigation was to examine and assess the practical application of UAVs in emergency scenarios in mountainous areas. RT technique is based on accurate point-to-point predictions between each UAV and the associated User Equipment (UE) locations. It utilizes a propagation channel model with multipath components in both space and time that give important details about the amplitude, phase shift, angle of arrival (AoA), and precise 3D angle of departure (AoD). This paper talks about a wireless network that works in the 3.5, 6, 28, and 60 GHz [56] bands. It carries this out by focusing on two very important issues: how path loss works and how to describe the channels. It also aims to examine and analyze the received power, coverage area, and path loss characteristics in the frequency ranges within two specific scenarios. The second focus is to compare the analysis with suburban settings with more buildings but fewer flats and no sturdy mountains.
3.1. System Model
The two scenarios considered in the study are Scenario 1 (S1), Remote mountainous Skardu, Pakistan, and Scenario 2 (S2), Suburban Bangi, Malaysia, through the wide campus area of the National University of Malaysia (UKM). The northern mountainous territory of Skardu, Gilgit Baltistan, Pakistan, was rebuilt from the OSM file from 75.4501° E to 35.4349° N, and the suburban range of UKM is between 2.9290° N and 101.7801° E. These captivating landscapes and their interplay are captured in Figure 1 and Figure 2. Skardu is a captivating and physically varied location known for its serene lakes, glacier landscapes, and majestic mountains. UKM, in comparison, has a relatively flat topography with minor undulations and no prominent mountainous or hilly characteristics. The presence of these unique environments provided the opportunity to investigate the obstacles and complexities associated with the implementation of UAVs for connectivity extension in a variety of contexts.
To enhance the precision and authenticity of the simulations, authentic building data obtained from the two locations were incorporated. The most significant human-made structures rise to 10 m in Skardu and 20 m in UKM. The buildings were modeled to represent their uneven forms and variable height effects on multiple propagations. In UAV technology’s signal characterization and propagation framework, exploring unique surroundings in various nations is justifiable, even though standard practice favors same-country comparisons. This work provides insight into signal variability related to geography, building density, and meteorological conditions. Although this study is primarily based on simulations, i.e., the real-world information incorporated through the use of OpenStreetMap (OSM) files, which accurately depict the landscape, terrain, and obstacles such as buildings and vegetation in the chosen environment is crucial for the propagation modeling. These parameters and variables impact the signal behavior and path loss characteristics. The frequency operations adhere to established standards, ensuring the practical relevance of the simulations. This foundational understanding paves the way for future real-world measurements, which can be conducted more efficiently and resourcefully, guided by the insights gained from these simulations.
Figure 1 and Figure 2 depict the distribution of 100 UEs on the ground in the S1 and S2 environments, respectively. The positions of all these UEs were determined randomly using MATLAB following a normal distribution. The UEs are shown as blue markers to emphasize their ground-based attributes, while a red marker represents the projected position of the UAV. The UAV position was determined based on the existing position of the terrestrial base station in the area.
The simulation was based on omnidirectional antenna deployment in both the UAV and at UE points. UAV height is varied, while UE points were set to be at 1.5 m above ground level (AGL). The required receiver sensitivity at the user end may vary based on the 5G New Radio (NR) FR1 and FR2 frequency bands [57]. The simulations ignored antenna gain at both the Tx and Rx ends for simplicity, and the receiver sensitivity was kept at −100 dBm irrespective of the frequency to analyze the results independent of this parameter of receiver design. The integration of omnidirectional antennas adheres to basic signal propagation and antenna design principles of the area and UAVs. Omnidirectional antennas have the unique property of producing signals evenly in all directions, making them especially well suited for dealing with the complex situations of signal reflection and multipath propagation [58]. The analysis concentrates on downlink performance with a maximum transmission power of 5 watts and assumes that the Tx operates at the highest allowed power without power control. This widely recognized technique allows for the accurate modeling of radio wave propagation paths, considering reflections and diffractions in complex environments. The method ensures that the calculated paths maintain precise geometric accuracy, further enhancing the reliability of the study’s findings.
The shooting and bouncing ray (SBR) method was applied to a maximum of ten reflections, signifying that in NLOS conditions, the transmitted signal can reach its intended destination by undergoing multiple reflections from various surfaces. This unique propagation behavior enables the signal to cover significant distances during transmission and develops dedicated positioning methodologies for UAV deployment in rural mountainous and suburban areas. Taking into consideration unique aspects of both mountainous rural and suburban environments, the RT method was applied to identify all critical ray pathways linking the UAV and the UE. Analysis was performed and compared across a 2 × 3 km² region in each scenario.
Figure 3 shows an example in which rays interact with buildings and mountains in S1, undergo reflection, and go through various atmospheric conditions such as rain, gases, and foliage. To effectively plan and construct a communication network operating at mmWaves and THz frequency bands, it is essential to use precise radio channel characterization and coverage-prediction methodologies in both locations. The deterministic RT technique is of great significance due to its ability to effectively analyze propagation in diverse contexts. UAV communication inherently encounters distinct propagation conditions in contrast to terrestrial communication. Consequently, the traditional route loss, LOS probability, and fading models used in terrestrial communication cannot be extrapolated to UAV communication.
By adopting this approach, the research aims to shed light on the relationship between UAV height, path loss, and received power in different scenarios. Moreover, the study explores the distinct scattering scenarios to capture real-world conditions, perfect reflectors, pure concrete walls, and complex rain and foliage settings. The foliage scenario specifically incorporates a foliage depth of 5 m to study the influence of different frequencies. By considering these diverse scenarios, the research provides valuable insights into the behavior of 5G networks under other environmental conditions.
Simulations were conducted across frequency centers at 3.5, 6, 28, and 60 GHz, enabling a comprehensive analysis of different frequency spectrum performance characteristics. The study also evaluates critical parameters such as path loss and received power, particularly emphasizing the impact of varying transmitter antenna heights. The transmitter power remains fixed at 5 watts throughout the analysis, while the UAV altitude ranges from 30 to 120 m. The simulation parameters are provided in Table 3. The analysis will be used to determine the best position of the UAV in each case.
UAVs are often fueled by high-energy lithium batteries, with flight periods ranging from 20 to 40 min. The battery charge duration is influenced by multiple factors, including the consumption of other onboard electronics and the power level of transmission, particularly when the UAV functions as a base station [59]. Factors such as signal strength, interference, and the need for signal re-transmissions in certain channel models can all contribute to increased battery usage. UAVs may need to adjust their communication protocols, flight paths, or transmission power to optimize battery consumption. The resulting model can contribute to the automatic adjustment and power-consumption-optimization strategies but is not within the scope of this study.
3.2. Modeling the Received Signal Strength
To model the Received Signal Strength (RSS) of the scenarios, the following considerations were integrated into the analysis.
3.2.1. Path Loss Considerations
The higher the altitude of the UAV, the greater the distance between the UAV and ground users (UEs), resulting in an increase in free space path loss. Furthermore, in a realistic environment, the building material and the presence of weather and foliage affect the Received Signal Strength. Therefore, this study includes the effect of concrete, weather, and foliage in calculating the Received Signal Strength at the user end. The RT propagation model used in Matlab determines the path loss of each ray using electromagnetic analysis. This calculation includes free-space, reflection, and diffraction losses according to the characteristics of building materials such as concrete, provided in the ITU standards [60,61]. For the consideration of losses due to weather, this study calculates the attenuation due to the presence of rain and atmospheric gases as per ITU standards [62,63]. The signal strength is also affected by the presence of foliage. Since we have chosen 5 m depth, the model [46] is used to include the losses due to foliage in this study.
3.2.2. Received Power Levels
The UAV reaches an optimal altitude, which strikes a compromise between reducing route loss and increasing the probability of LOS connection with ground users. NLOS scenarios might occur more often at lower altitudes, such as 45 m, resulting in decreased received power levels measured by the signal strength. The number of obstacles is reduced at a distance, leading to a higher likelihood of LOS communication and improved signal strength.
3.2.3. Frequency-Specific Observations
Frequencies in the lower range of 3.5 GHz and 6 GHz exhibit superior performance at lower altitudes and possess a greater ability to overcome obstacles, making them well suited for suburban or blocked settings. Higher frequencies, 28 GHz and 60 GHz, are appropriate for high-data-rate applications, but they need closer proximity to maintain a stable connection owing to increased route loss and air absorption. Lower altitudes may result in poorer signals owing to greater reflections and obstacles.
3.2.4. Propagation Delay Analysis
Propagation delay, which is the duration it takes for a signal to travel from the Tx to the Rx, is important for network planning. In our scenarios, it is a critical factor in accurately estimating distances for localization and positioning of UAVs. In this study, the propagation delay behavior is analyzed based on maximum, minimum, and average delay measurements at various altitudes between scenarios S1 and S2. Such analysis aims to provide a better understanding of the effect of height on signal transmission without diving into specific causes or variations.
3.2.5. Ranges for Height Optimization
The identification of the optimized height of UAVs for communication purposes is a complex task that involves considering several aspects to produce the best possible signal strength and coverage. The determination is restricted to regulatory limitations [64,65]. Some countries allow UAVs to fly up to 200 m, but in general, most countries agree to ensure that UAVs stay below 120 m to help maintain a restricted altitude range, which minimizes the chances of accidents with commercial and private manned aircraft that usually fly at higher altitudes. This measure enhances air traffic safety. UAV operators should follow national aviation rules. Exceeding the designated altitude of 120 m without explicit authorization might result in legal ramifications and provide potential hazards to safety. Special licenses for higher altitude operations may be given but must undergo stringent safety evaluations and collaboration with aviation authorities.
The flowchart in Figure 4 summarizes the methodology and the elements of consideration in the analysis.
4. Results
4.1. Received Signal Strength Modeling
The first results are given in Figure 5 and Figure 6 for S1 and S2, respectively, and at a fixed height of 75 m. For each of the scenarios, the graphs distinguish the simulation measurement following the four working simulated frequencies. The x-axis presents the distance (in m) of each UE to the Tx on the UAV but on a logarithmic scale. The results show the RSS or the actual obtained measurement (in dBM) on the primary y-axis and residual error on the secondary y-axis. The mean value for each frequency’s case is also displayed to quantify the measurements. The best results obtained when data are fitted follow quadratic polynomial regression. This technique is suitable and inherently consistent with the anticipated parabolic trend of signal attenuation in these terrains. Given the observed link between distance and signal strength, it is evident that a curved trajectory is followed. Consequently, the use of quadratic equations emerges as a logical approach for precise modeling.
In Figure 5, the graphs show a considerable fall in received signal power levels while migrating from the 3.5 GHz to higher frequencies in the 6 GHz, 28 GHz, and 60 GHz bands. Throughout this shift, the mean received power drops by around 5 dB, 24.2 dB, and 45.5 dB, respectively. Similarly, in Figure 6 when shifting from the 3.5 GHz band to the higher frequencies, the received power levels drop by around 5 dB, 24.2 dB, and 49.5 dB, respectively. The comparison between the 28 and 60 GHz bands is especially noteworthy. The same pattern can be recorded in other measurements using different UAV heights.
The residual errors illustrated under the secondary y-axis to quantify the quadratic fit were derived using the least squares method, which minimizes the sum of the squared differences between the observed data and the predicted values from the quadratic equation. Their simplicity and computational efficiency further justify the use of quadratic polynomials. These equations provide the ideal compromise between complexity and accuracy, capturing the essential traits without encouraging overfitting. The quadratic fit equation allows us to predict outcomes, and the coefficients may be interpreted easily to provide compelling support for adopting this strategy. Identifying anomalies in signal strength is crucial, considering the distinctive geographical features of these regions, as deviations from the mean may signify such anomalies. Outliers were also computed and are highlighted in the results. The resulting fit quadratic polynomial models, the RSS, is described by Equation (2),
(2)
where the coefficients a, b, and c represent the best fit quadratic values for each frequency d represents the distance between the Tx and Rx. The coefficients under consideration have significant relevance in understanding the correlation between distance and signal intensity, enabling precise predictions about the magnitude of the received signal. In many cases, the relationship between distance and signal intensity follows an inverse square law, although interference, barriers, and environmental constraints may necessitate adjustments to the mathematical model. These coefficients also make it easier to extrapolate signal intensity to untested distances, which is particularly useful in geographically difficult places where extensive measurement is impossible. Such information enables signal-level prediction and network planning.4.2. Analysis of the Optimal UAV Position
Several critical insights emerged when analyzing path losses versus altitude at different frequencies. First, path loss increases with UAV altitude at all frequencies: 3.5 GHz, 6 GHz, 28 GHz, and 60 GHz. The free-space path loss model predicts that signal attenuation increases as altitude and the distance between the UAV and the receiver increase. Additionally, frequency dramatically affects route losses at a given height. The maximum path losses at 30 m are 60 GHz, while the lowest are 3.5 GHz. At 75 m and 120 m, higher frequencies always cause more significant path losses.
Table 4 and Table 5 illustrate how average RSS in (dBm) varies across different frequencies and heights 30 m, 75 m, and 120 m, but this time with the sequential introduction of concrete material for building and the cold mountains, weather, and foliage depth for the green parts of the mountains. Higher frequencies generally experience more significant attenuation, resulting in lower signal strengths compared to lower frequencies. Specifically, at 60 GHz, the dBm values are more negative than at 3.5 GHz and 6 GHz across all elevations. Increased elevation typically improves LOS conditions, leading to better coverage and service capacity. However, this advantage diminishes at higher frequencies due to heightened susceptibility to environmental factors such as atmospheric absorption, reflection, and diffraction. At higher altitudes, transmissions with higher frequencies deteriorate more quickly. Signal strength exhibits greater variability at higher frequencies, primarily due to factors like weather conditions, vegetation, and physical obstacles.
The results also show the final average path loss values with the introduction of all elements. It can be seen that at 60 GHz, the path loss reached −145.9 dbm in S1 and 140.4 dbm in S2. These findings highlight the complexities involved in deploying high-frequency communication systems, particularly those used in 5G and beyond, necessitating advanced mitigation strategies to effectively manage signal loss. mmWave frequencies (28 GHz and 60 GHz) incur more path losses in comparison to lower sub-6 GHz frequencies. The impact of weather on UAV signal strength varies significantly with frequency and altitude. Lower frequencies of 3.5 GHz and 6 GHz are less affected by weather conditions such as rain, fog, and humidity compared to higher frequencies 28 GHz and 60 GHz, which suffer significant attenuation due to these factors. At lower altitudes of 30 m, UAVs are less exposed to severe weather but are more likely to encounter ground-level obstacles.
The data presented in Table 4 and Table 5 also demonstrate the disparities in signal propagation for different frequencies throughout S1 and S2. Both terrains suffer a rise in values at lower frequencies, such as 3.5 GHz and 6 GHz, suggesting probable signal attenuation. Higher frequencies, on the other hand, react differently. In S2, values decrease from 6 GHz to 28 GHz and 60 GHz, implying effective solutions, such as beamforming or massive Multiple-Input–Multiple-Output (MIMO), to compensate for the performance and improved signal quality for those closer to the transmitter. On the other hand, S1 consistently displays lower relative performance across all frequencies, emphasizing the inherent problems of the rough topography.
Figure 7 (for S1) and Figure 8 (for S2) show the RSS results with all effects in box plot representation, showing that signal intensity decreases with frequency in both locations. In all frequency bands, S2 has greater maximum and lowest signal intensities than S1, suggesting better signal coverage. Further median and quartile research indicates suburban dominance over mountainous places. S2 had larger median and quartile values, suggesting stronger middle-range signals. This shows better signal coverage and more durable suburban wireless communication infrastructure. Outliers are seen in both zones, especially at 28 and 60 GHz. These anomalies may reveal distinct environmental or interference issues impacting signal transmission. Investigating these outliers may provide helpful information.
The findings revealed that the mean received power at 75 m is higher than at 45 m and 30 m. When the UAV is at a low altitude of 45 m, the distance between terrestrial UEs and the UAV is small, resulting in low received power for some UEs. However, rays follow multiple paths to reach the receiver due to numerous reflections and obstructions in the environment. Furthermore, there are a considerable number of users along the cell edge. Increasing the UAV height from 45 m to 75 m increases the distance between terrestrial UEs and the UAV. However, the greater height enhances the possibility of LOS between the UAV and the user, increasing coverage inside the cell. This is clear in the figure, where the mean received power level with a 75 m UAV altitude is more significant than when the UAV is at higher heights. In other words, for the given situation and user grid, a UAV altitude of 75 m is determined to be an ideal height for received power.
The determination of the UAV altitude at 75 m in this context was made through a meticulous evaluation process that considered various factors and weighed potential compromises. Ensuring NLOS coverage for all 100 randomly generated UEs constituted a pivotal element. In both scenarios, between 30 m and 75 m, there are considerable differences in the number of users with total outages, as compared to 75 m and 120 m. Each is explained by the quantity of allowed reflections and diffraction. Selecting a height of 75 m confirmed that with one or two exceptions, all UEs would consistently receive signals, even when obstacles impede communication. If a shorter height, such as 60 m, had been chosen, it is plausible that there would have been situations in which specific UEs may have experienced a lack of signal reception because of signal obstruction. The assessment of signal strength was a pivotal factor in ascertaining the most favorable elevation. Although increasing the height to 90 m could have potentially improved signal strength, it would have resulted in a lack of signal reception for users beyond 2 × 3 km². Conversely, a reduced vertical dimension, such as 60 m, may have potentially yielded improved signal strength across the entirety of the coverage area. However, it is essential to note that even with this lower height, there remains the possibility of certain UEs situated at 2 × 3 km² experiencing a complete lack of signal reception. Therefore, selecting a 75 m distance achieved a compromise between signal strength and coverage, resulting in a more equitable dispersion of signal reception throughout the designated area.
Additionally, the choice to operate the UAV at an altitude of 75 m was also influenced by factors related to battery usage. In contrast to the elevated altitude of 90 m, the utilization of a lower altitude at 75 m resulted in reduced battery power consumption, thereby enhancing flight duration and overall operational effectiveness [52]. The decision to set the UAV height at 75 m was a carefully considered choice considering factors such as NLOS coverage, signal strength, and battery efficiency. Achieving an optimal equilibrium among these variables ensured the attainment of efficient communication and enhanced performance within the specified context. Figure 7 and Figure 8 illustrate the mean received power levels at various frequencies, supporting the chosen UAV height as optimal for the given conditions. Nonetheless, the comprehensive findings obtained from the study yielded significant insights into wireless communication efficacy within rural and suburban regions.
The signal-dispersion differences between mountainous and suburban areas affect wireless communication tactics. Mountainous terrain’s signal strength drops highlight topographical impediments like mountains and valleys. Signal repeaters or other technologies may help overcome these obstacles. Conversely, suburban regions’ stronger signal strengths may be due to less congested and larger surroundings. These results suggest that suburban areas may build dependable wireless communication infrastructure. Enhancing wireless communication systems in different situations requires understanding these signal strength changes across geographies and frequency bands. The findings may improve wireless communication infrastructure in rural mountainous and suburban areas by informing network design, signal coverage optimizations, and infrastructure upgrading.
4.3. Propagation Delay Analysis
In Figure 9 and Figure 10, propagation delay characteristics are thoroughly examined in two diverse environmental settings: mountainous regions (S1) and suburban (S2). A significant increase can be detected for UEs in S1 with more than 1000 m, where the order of delay is the same as in S2 but at 1200 m. A spike is also introduced around this distance for UEs in S1, and it can be attributed to users who are on the other parts of the mountain than the base station position. Despite the delay, these users still have signals with sufficient strength. A visible pattern appears as UAV altitude climbs from 45 to 120 m in the S2 environment. The maximum delay increases noticeably, rising from 5.81 s at 45 m to 5.84 s at 120 m. In contrast, the minimum delay has improved somewhat, from 0.259 s to 0.455 s. These findings are explained by the complicated suburban environment, where signal reflections and multipath interference become more prominent as altitude increases. Surprisingly, the average delay at a UAV height of 120 m reports a significant rise to 5.83 s, suggesting occasional signal propagation issues.
Table 6 shows the tendency in maximum, minimum, and average delay for each case. The minimum delay in S1 varies from 0.28 to 0.48, while maximum delays range only from 7.20 s at 45 m to 7.23 s at 120 m. Notably, this indicates that the intrinsic topographical factors impact signal transmission more uniformly in the maximum delay. A similar pattern is observed for S2. In both scenarios, altitude selection is critical in shaping propagation delay. The choice of the best height must be skillfully balanced, considering aspects such as signal quality, coverage requirements, and the unique needs of the planned application. Higher altitudes have longer maximum delays, possibly expanding coverage. However, this comes at the expense of periodically increased average latency, which may affect real-time applications. Lower altitudes, on the other hand, may provide lower maximum delays but are better suited to applications requiring little delay, such as video streaming or UAV-based surveillance.
The final proposed RSS model obtained by fitting the curve, following Equation (2) at 75 m to S1, as given in Table 7, while Table 8 presents the same results but for S2. The models are suggested for more exploration in the future in the same type of environment. While this study focuses on simulation-based analysis without practical methods for real-time monitoring and adjustment, it is essential to note that simulations based on probabilistic models can effectively capture real-time behaviors in dynamic environments. Implementing continuous real-time adjustments could indeed enhance communication performance by improving accuracy, but it would also demand substantial processing resources. For practical UAV deployments, predictive modeling derived from simulations plays a crucial role in optimizing operations and mitigating risks. Future advancements in artificial intelligence hold promise for enabling real-time adaptive strategies in UAV deployment scenarios. However, the current study lays a foundational understanding through simulation-based probabilistic modeling, which is instrumental in informing initial deployment strategies and guiding future research toward more adaptive and efficient UAV communication systems.
4.4. Future Work and Considerations Beyond Our Current Scope
This paper primarily focuses on the deployment of UAVs in specific environmental settings, and while it acknowledges the importance of weather impacts on UAV performance, a detailed analysis of these factors is beyond the current scope. The impact of weather that was considered includes rain and gas, but not fog, humidity, wind, and temperature variations. Lower frequencies of 3.5 GHz and 6 GHz are generally less affected by weather conditions compared to higher frequencies of 28 GHz and 60 GHz, which suffer significant attenuation due to these factors. At lower altitudes of 30 m, UAVs are less exposed to severe weather but are more likely to encounter ground-level obstacles, while higher altitudes of 75 m and 120 m offer larger coverage areas but increase exposure to atmospheric conditions such as strong winds and temperature variations, leading to greater signal degradation. Therefore, higher frequencies and altitudes necessitate advanced adaptive techniques to mitigate weather-related impacts on UAV communications. Future work will include a detailed analysis and quantification of weather effects to provide a more comprehensive understanding of UAV performance under varying conditions.
The types and sources of interference are critical aspects that require further investigation beyond the scope of this study. While our research focuses on evaluating UAV applications using RT propagation modeling, understanding interference sources is essential for comprehensive system analysis. It is important to note that the current findings, which do not account for interference, may represent an upper performance bound. When interference is considered, the signal-to-noise and interference ratio impacts signal quality for all users, although trends at different altitudes are likely to follow similar patterns. Future studies should explore co-channel and adjacent channel interference to understand their effects better and represent a more practical limitation of the propagation environment.
The work herein provides valuable insights into the immediate effects of UAV communication in emergency scenarios using RT propagation modeling. It is important to acknowledge that long-term variations in signal propagation dynamics, such as seasonal changes, evolving environmental conditions, operational fluctuations, and new developments like buildings, can significantly impact real-world performance over time. Similarly, changes in vegetation can alter foliage losses, further influencing propagation characteristics. Future studies could benefit from analyzing these factors across various temporal and environmental conditions to provide a more comprehensive understanding of UAV communication systems’ reliability and performance stability. This approach is crucial, especially when new developments and environmental changes require further experimentation to validate and refine the model.
Finally, examining signal strength in S1 and S2 by utilizing quadratic polynomials and the subsequent derivation of coefficients yields significant findings regarding the correlation between signal intensity and distance and the transmission of signals in difficult topographical conditions. The box plots clearly illustrate a negative correlation between signal strength and frequency, indicating the necessity for customized approaches and infrastructure advancements to effectively address geographical challenges and enhance wireless communication systems in various settings. While the proposed approach is consistent with existing signal propagation theories, it must be validated by practical research such as the mobility of UEs and different arrival rates. Following establishing a baseline, the insertion of specialized antennas may be assessed to see whether they increase coverage and communication accuracy within their specific environmental niches. The iterative nature of this procedure ensures that the selected base station technique meets realistic performance standards and efficiently improves communication in various ecological settings.
5. Conclusions
In a typical mountainous region, terrestrial base stations are scarce as they are normally not considered cost-effective. The use of UAVs to compensate for the lack of connectivity is of interest, but few studies focus on propagation modeling in such an area. Thus, a Ray-Tracing-based channel characterization has been presented based on the distribution of distances between User Equipment and the transmitter attached to UAVs. The Ray-Tracing multiple-path propagation tool was used to generate received signal strength against distance and set to minimize the errors observed on mean square error quantification. For the frequency of 3.5 GHz, for example, the average losses in received power for weather, foliage, delay, and phase shift are below 1.5 dBm, 3 dBm, 2 s, and 3 degrees, respectively. In the height variation analysis, flying a UAV at a height of 75 m results in optimized conditions that maximize LOS conditions and minimize NLOS complete outage of no signal. A strategic location near the coverage area’s center optimizes signal propagation and reduces the impact of barriers. The study finally proposed four Received Signal Strength models that can be used in network planning based on different frequencies, including the mmWave range of 28 and 60 GHz. The final section highlighted several other opportunities for the study, opening up the opportunity for more research and exploration in the area.
The final suggested models are particularly relevant to the specific types of environments discussed in this study, such as mountainous and suburban areas. These conclusions provide a realistic representation of performance in near real-world scenarios, offering valuable insights for drone operators in such environments. For different types of terrain or metropolitan settings, further investigation using the same framework is required due to differences in obstacles and vegetation. However, the performance trends, especially those influenced by distance-dependent path loss, are expected to be similar. While the average RSS may vary, the model presented here can still provide useful trends and insights that can be adapted to other environments with additional validation.
Conceptualization, all authors; methodology, all authors; software, S.A.; validation, all authors; formal analysis, S.A.; investigation, S.A. and N.L.M.K.; resources, A.A.-S.; data curation, S.A., A.A.-S. and N.L.M.K.; writing—original draft preparation, S.A.; writing—review and editing, A.A.-S., N.F.A., and N.L.M.K.; visualization, S.A.; supervision, A.A.-S., N.F.A. and N.L.M.K.; project administration, A.A.-S. and N.F.A.; funding acquisition, A.A.-S. All authors have read and agreed to the published version of the manuscript.
The raw data supporting the conclusions of this article will be made available by the authors on request.
The authors declare no conflicts of interest.
The following abbreviations are used in this manuscript:
| 5G | Fifth Generation (of Mobile Communication) |
| 6G | Sixth Generation |
| 3D | Three Dimensional |
| GHz | Gigahertz |
| PL | Path losses |
| mmWave | Millimeter wave |
| A2G | Air-to-Ground |
| LOS | Line-of-Sight |
| MB | Multiple Bounce |
| MSE | Mean Square Error |
| NLOS | Non-Line-of-Sight |
| dB | Decibel |
| GBSMs | Geometry-based stochastic models |
| MIMO | Multiple-Input Multiple-Output |
| 3GPP | Third-generation partnership project |
| m | Metre |
| Km | Kilometre |
| SBR | Shooting, and bouncing ray |
| RSS | Received Signal Strength |
| PITI | The total propagation loss |
| MATLAB | Matrix laboratory |
| UE | User Equipment |
| AOA | Angle of Arrival |
| AOD | Angle of Departure |
| Rx | Receiver |
| Tx | Transmitter |
| RT | Ray tracing |
| SB | Single Bounce |
| URLLC | Ultra Reliable Low Latency Communications |
| UAV | Unmanned Aerial Vehicle |
| UKM | Universiti Kebangsaan Malaysia (The National University of Malaysia) |
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 5. Distance and signal strength relationship in S1 using 4 different frequencies.
Enlarge this image.Figure 6. Distance and signal strength relationship in S2 using 4 different frequencies.
Enlarge this image.Figure 9. Propagation delay at different altitudes of UAV at different distances of UE in S1.
Enlarge this image.Figure 10. Propagation delay at different altitudes of UAV at different distances of UE in S2.
Comparison of existing exponential decay foliage models.
| Model | A | B | C | F (GHz) | D (m) |
|---|---|---|---|---|---|
| CCIR [ | 0.2 | 0.3 | 0.6 | 0.2–95 | 0, 400 |
| Wiesberger_a [ | 1.33 | 0.248 | 0.588 | 0.23–95 | 14, 400 |
| Wiesberger_b [ | 0.45 | 0.248 | 1 | 0.23–95 | 0, 14 |
| Ko et al. [ | 0.805 | 0.261 | 0.277 | 28 | 20, 370 |
| Lv et al. [ | 2.143 | 0.078 | 0.650 | 38–40 | 2.9, 11.8 |
| FITU-R [ | 0.39 | 0.39 | 0.25 | 9–40 | 0, 200 |
| Cost235 [ | 15.6 | −0.009 | 0.26 | 9.6–57.6 | 0, 200 |
| Horak et al. [ | 0.39 | 0.39 | 0.25 | 9.6–57 | 0, 200 |
Literature review on outdoor mmWave deployment studies.
| Ref. | Freq GHz | Env. | Contributions |
|---|---|---|---|
| [ | 1–100 | Suburban | The research considers channel characteristics across spectra and scenarios, offering adaptability to simplify specific channel models through parameter adjustments. |
| [ | 60 | Suburban | Measurements carried out at the American College of Greece. It offers insights into excess loss, azimuth gain degradation, and temporal power fluctuations. |
| [ | mmWaves | Sea | It analyzes statistical properties and validates the model’s accuracy through comparison with measurement data, aiding system design and performance evaluation for UAV communication networks at sea. |
| [ | mmWaves | Urban | Stochastic model for UAV-to-vehicle communication. It validates the model’s time-variant statistical properties through simulations based on measured and RT data, enhancing understanding of UAV communication channels. |
| [ | mmWaves | Suburban | This article compares indoor and outdoor mmWave propagation using free space path loss models. Despite increased route loss outside, the approach delivers high packet delivery ratios, average throughput, and low latency. |
| [ | mmWaves | Urban | This study explores outdoor propagation, comparing real-world measurements with RT simulations to refine mmWaves wireless design. RT is proven accurate and aids in theoretical coverage area modeling. |
| This work | 3.6, 6, 28, 60 | S1, S2 | Provides insights into optimizing network design and UAV placement in 5G networks across diverse terrains, aiding wireless communication system performance in challenging environments. |
Simulation parameters used in this study.
| Parameter | Value |
|---|---|
| Carrier Frequency | 3.5, 6, 28 and 60 GHz |
| Tx Antenna Type | Omni directional |
| UE Antenna type | Omni directional |
| Transmitter Power | 5 W |
| Maximum range of Transmitter | 3 km |
| UAV altitude | 30–120 m |
| UEs Height | 1.5 m AGL |
| Receiver Sensitivity | −100 dBm |
| Coverage Area | 2 × 3 km² |
| Foliage Depth | 5 m |
| Material | Mountain, Building = concrete |
| Weather | Rain and Gas |
| Scenario 1 (S1) | Skardu, Pakistan |
| Scenario 2 (S2) | UKM, Malaysia |
Frequency, altitude, and environmental factor comparison analysis for S1 with max 10-meter-high buildings, tall mountain, and 5-meter foliage density.
| Parameters | RSS | Plus | Plus | Final | UEs | Final | |||
|---|---|---|---|---|---|---|---|---|---|
| Freq. (GHz) | Alt (m) | Average (dBm) | Nb | Max | Min | Avg | |||
| 30 | −62.9 | −63.0 | −66.2 | −101.3 | 5 | 6.3 | 0.0 | 3.2 | |
| 3.5 | 75 | −63.7 | −63.7 | −66.7 | −100.5 | 2 | 6.1 | 0.2 | 2.8 |
| 120 | −62.2 | −62.3 | −65.5 | −100.0 | 1 | 6.2 | 0.1 | 3.1 | |
| 30 | −68.3 | −68.4 | −72.1 | −106.1 | 6 | 6.2 | 0.2 | 3.4 | |
| 6 | 75 | −68.4 | −68.4 | −72.1 | −102.3 | 2 | 6.1 | 0.2 | 3.5 |
| 120 | −67.2 | −67.3 | −71.1 | −104.8 | 1 | 6.2 | 0.1 | 3.1 | |
| 30 | −81.1 | −85.1 | −90.9 | −123.5 | 8 | 6.2 | 0.1 | 3.1 | |
| 28 | 75 | −81.7 | −85.8 | −91.6 | −122.7 | 3 | 6.2 | 0.3 | 3.2 |
| 120 | −81.2 | −85.3 | −91.1 | −122.4 | 2 | 6.2 | 0.1 | 2.9 | |
| 30 | −89.4 | −110.2 | −117.4 | −148.5 | 9 | 6.2 | 0.1 | 3.0 | |
| 60 | 75 | −88.5 | −106.9 | −116.1 | −146.0 | 4 | 6.1 | 0.1 | 3.2 |
| 120 | −87.1 | −107.5 | −114.7 | −145.9 | 2 | 6.2 | 0.0 | 3.3 | |
Frequency, altitude, and environmental factor comparison analysis for S2 with max 20-meter-high buildings and 5-meter foliage density.
| Parameters | RSS | Plus | Plus | Final | UEs | Final | |||
|---|---|---|---|---|---|---|---|---|---|
| Freq. (GHz) | Alt (m) | Average (dBm) | Nb | Max | Min | Avg | |||
| 30 | −56.2 | −56.2 | −59.5 | −97.7 | 4 | 6.1 | 0.2 | 3.1 | |
| 3.5 | 75 | −57.6 | −57.5 | −60.7 | −97.9 | 1 | 6.1 | 0.0 | 2.9 |
| 120 | −58.1 | −57.9 | −61.2 | −98.5 | 0 | 6.2 | 0.2 | 3.2 | |
| 30 | −61.1 | −61.2 | −64.9 | −102.4 | 5 | 6.3 | 0.2 | 3.4 | |
| 6 | 75 | −61.9 | −61.9 | −65.8 | −102.6 | 1 | 6.2 | 0.1 | 3.2 |
| 120 | −63.2 | −63.1 | −66.8 | −102.8 | 0 | 6.2 | 0.1 | 3.2 | |
| 30 | −74.4 | −77.7 | −83.5 | −119.3 | 6 | 6.2 | 0.2 | 3.2 | |
| 28 | 75 | −75.4 | −79.0 | −84.8 | −119.5 | 2 | 6.2 | 0.2 | 3.2 |
| 120 | −76.3 | −79.9 | −85.7 | −119.9 | 1 | 6.2 | 0.1 | 3.1 | |
| 30 | −81.1 | −97.5 | −105.5 | −139.5 | 8 | 6.1 | 0.4 | 3.5 | |
| 60 | 75 | −81.8 | −98.6 | −105.8 | −140.0 | 3 | 6.0 | 0.1 | 2.9 |
| 120 | −82.6 | −99.7 | −106.9 | − 140.4 | 2 | 6.2 | 0.2 | 3.3 | |
Propagation delay analysis.
| Scenario | Altitude (m) | Max Delay ( | Min Delay ( | Average Delay ( |
|---|---|---|---|---|
| S1 | 45 | 7.20 | 0.28 | 2.09 |
| 75 | 7.21 | 0.32 | 2.08 | |
| 105 | 7.22 | 0.43 | 2.10 | |
| 120 | 7.23 | 0.48 | 2.11 | |
| S2 | 45 | 5.81 | 0.25 | 2.55 |
| 75 | 5.82 | 0.31 | 2.56 | |
| 105 | 5.83 | 0.40 | 2.58 | |
| 120 | 5.84 | 0.45 | 2.83 |
The quadratic regression coefficients for S1 found by varying the frequencies at 75 m.
| Frequency (GHz) | a | b | c |
|---|---|---|---|
| 3.5 | −4.72 | 8.89 | −49.75 |
| 6 | −7.27 | 21.29 | −69.94 |
| 28 | −8.02 | 21.14 | −84.48 |
| 60 | −32.56 | 128.96 | −213.8 |
The quadratic regression coefficient for S2 found by varying the frequencies at 75 m.
| Frequency (GHz) | a | b | c |
|---|---|---|---|
| 3.5 | −1.72 | −11.90 | −19.4 |
| 6 | 5.20 | −49.34 | 25.35 |
| 28 | −4.43 | −3.64 | −45.85 |
| 60 | −29.33 | 107.70 | −184.27 |
References
1. Hayat, S.; Yanmaz, E.; Muzaffar, R. Survey on unmanned aerial vehicle networks for civil applications: A communications viewpoint. IEEE Commun. Surv. Tutor.; 2016; 18, pp. 2624-2661. [DOI: https://dx.doi.org/10.1109/COMST.2016.2560343]
2. Saad, W.; Bennis, M.; Chen, M. A vision of 6G wireless systems: Applications, trends, technologies, and open research problems. IEEE Netw.; 2019; 34, pp. 134-142. [DOI: https://dx.doi.org/10.1109/MNET.001.1900287]
3. Zhang, C.; Zhang, W.; Wang, W.; Yang, L.; Zhang, W. Research challenges and opportunities of UAV millimeter-wave communications. IEEE Wirel. Commun.; 2019; 26, pp. 58-62. [DOI: https://dx.doi.org/10.1109/MWC.2018.1800214]
4. Zhou, T.; Tao, C.; Salous, S.; Liu, L. Measurements and analysis of angular characteristics and spatial correlation for high-speed railway channels. IEEE Trans. Intell. Transp. Syst.; 2017; 19, pp. 357-367. [DOI: https://dx.doi.org/10.1109/TITS.2017.2681112]
5. Jawad, A.M.; Nordin, R.; Jawad, H.M.; Gharghan, S.K.; Abu-Samah, A.; Abu-Alshaeer, M.J.; Abdullah, N.F. Wireless drone charging station using class-e power amplifier in vertical alignment and lateral misalignment conditions. Energies; 2022; 15, 1298. [DOI: https://dx.doi.org/10.3390/en15041298]
6. Alzahrani, B.; Oubbati, O.S.; Barnawi, A.; Atiquzzaman, M.; Alghazzawi, D. UAV assistance paradigm: State-of-the-art in applications and challenges. J. Netw. Comput. Appl.; 2020; 166, 102706. [DOI: https://dx.doi.org/10.1016/j.jnca.2020.102706]
7. Zhu, L.; Zhang, J.; Xiao, Z.; Cao, X.; Wu, D.O.; Xia, X.G. 3-D beamforming for flexible coverage in millimeter-wave UAV communications. IEEE Wirel. Commun. Lett.; 2019; 8, pp. 837-840. [DOI: https://dx.doi.org/10.1109/LWC.2019.2895597]
8. Liu, L.; Zhang, S.; Zhang, R. Multi-beam UAV communication in cellular uplink: Cooperative interference cancellation and sum-rate maximization. IEEE Trans. Wirel. Commun.; 2019; 18, pp. 4679-4691. [DOI: https://dx.doi.org/10.1109/TWC.2019.2926981]
9. Wang, Z.; Duan, L. Chase or wait: Dynamic UAV deployment to learn and catch time-varying user activities. IEEE Trans. Mob. Comput.; 2021; 22, pp. 1369-1383. [DOI: https://dx.doi.org/10.1109/TMC.2021.3107027]
10. Zakariaa, A.F.; Nasirb, M.F.M.; bin Mukhtar, A.; Ridzuand, M.N. Design Analysis of Open and Ducted Propellers in UAV Application. J. Kejuruter.; 2022; 34, pp. 119-125. [DOI: https://dx.doi.org/10.17576/jkukm-2022-si5(2)-13]
11. Mohd Kamal, N.; Sahwee, Z.; Norhashim, N.; Ahmad Shah, S.; Hashim, W. Throughput Performance of Cellular-Connected UAV Coexisting with Terrestrial Users. Unmanned Syst.; 2024; 12, pp. 535-543. [DOI: https://dx.doi.org/10.1142/S230138502442007X]
12. Helander, J.; Zhao, K.; Ying, Z.; Sjöberg, D. Performance analysis of millimeter-wave phased array antennas in cellular handsets. IEEE Antennas Wirel. Propag. Lett.; 2015; 15, pp. 504-507. [DOI: https://dx.doi.org/10.1109/LAWP.2015.2455040]
13. Alraih, S.; Nordin, R.; Shayea, I.; Abdullah, N.F.; Abu-Samah, A.; Alhammadi, A. Effectiveness of handover control parameters on handover performance in 5G and beyond mobile networks. Wirel. Commun. Mob. Comput.; 2022; 2022, 2266282. [DOI: https://dx.doi.org/10.1155/2022/2266282]
14. Zhang, J.; Ge, X.; Li, Q.; Guizani, M.; Zhang, Y. 5G millimeter-wave antenna array: Design and challenges. IEEE Wirel. Commun.; 2016; 24, pp. 106-112. [DOI: https://dx.doi.org/10.1109/MWC.2016.1400374RP]
15. Zeng, Y.; Wu, Q.; Zhang, R. Accessing from the sky: A tutorial on UAV communications for 5G and beyond. Proc. IEEE; 2019; 107, pp. 2327-2375. [DOI: https://dx.doi.org/10.1109/JPROC.2019.2952892]
16. Ancans, G.; Bobrovs, V.; Haidine, A.; Aqqal, A. Spectrum usage for 5G mobile communication systems and electromagnetic compatibility with existent technologies. Broadband Communications Networks-Recent Advances and Lessons from Practice; IntechOpen: London, UK, 2018; pp. 27-41.
17. Abdullah, N.F.; Berraki, D.; Ameen, A.; Armour, S.; Doufexi, A.; Nix, A.; Beach, M. Channel parameters and throughput predictions for mmWave and LTE-A networks in urban environments. Proceedings of the 2015 IEEE 81st Vehicular Technology Conference (VTC Spring); Glasgow, UK, 11–14 May 2015; IEEE: Piscataway, NJ, USA, 2015; pp. 1-5. [DOI: https://dx.doi.org/10.1109/VTCSpring.2015.7145737]
18. Rappaport, T.S.; Sun, S.; Mayzus, R.; Zhao, H.; Azar, Y.; Wang, K.; Wong, G.N.; Schulz, J.K.; Samimi, M.; Gutierrez, F. Millimeter wave mobile communications for 5G cellular: It will work!. IEEE Access; 2013; 1, pp. 335-349. [DOI: https://dx.doi.org/10.1109/ACCESS.2013.2260813]
19. Alobaidy, H.A.; Singh, M.J.; Behjati, M.; Nordin, R.; Abdullah, N.F. Wireless transmissions, propagation and channel modelling for iot technologies: Applications and challenges. IEEE Access; 2022; 10, pp. 24095-24131. [DOI: https://dx.doi.org/10.1109/ACCESS.2022.3151967]
20. Zhao, N.; Lu, W.; Sheng, M.; Chen, Y.; Tang, J.; Yu, F.R.; Wong, K.K. UAV-assisted emergency networks in disasters. IEEE Wirel. Commun.; 2019; 26, pp. 45-51. [DOI: https://dx.doi.org/10.1109/MWC.2018.1800160]
21. Wang, C.X.; Bian, J.; Sun, J.; Zhang, W.; Zhang, M. A survey of 5G channel measurements and models. IEEE Commun. Surv. Tutorials; 2018; 20, pp. 3142-3168. [DOI: https://dx.doi.org/10.1109/COMST.2018.2862141]
22. Wang, C.X.; Lv, Z.; Gao, X.; You, X.; Hao, Y.; Haas, H. Pervasive wireless channel modeling theory and applications to 6G GBSMs for all frequency bands and all scenarios. IEEE Trans. Veh. Technol.; 2022; 71, pp. 9159-9173. [DOI: https://dx.doi.org/10.1109/TVT.2022.3179695]
23. Cheng, X.; Huang, Z.; Chen, S. Vehicular communication channel measurement, modelling, and application for beyond 5G and 6G. IET Commun.; 2020; 14, pp. 3303-3311. [DOI: https://dx.doi.org/10.1049/iet-com.2020.0531]
24. Yu, C.; Liu, Y.; Chang, H.; Zhang, J.; Zhang, M.; Poechmueller, P.; Wang, C. AG channel measurements and characteristics analysis in hilly scenarios for 6G UAV communications. China Commun.; 2022; 19, pp. 32-46. [DOI: https://dx.doi.org/10.23919/JCC.2022.11.003]
25. Matolak, D.W.; Sun, R. Air–ground channel characterization for unmanned aircraft systems—Part III: The suburban and near-urban environments. IEEE Trans. Veh. Technol.y; 2017; 66, pp. 6607-6618. [DOI: https://dx.doi.org/10.1109/TVT.2017.2659651]
26. Khawaja, W.; Ozdemir, O.; Guvenc, I. UAV air-to-ground channel characterization for mmWave systems. Proceedings of the 2017 IEEE 86th Vehicular Technology Conference (VTC-Fall); Toronto, ON, Canada, 24–27 September 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 1-5.
27. Wheeb, A.H.; Nordin, R.; Samah, A.; Alsharif, M.H.; Khan, M.A. Topology-based routing protocols and mobility models for flying ad hoc networks: A contemporary review and future research directions. Drones; 2021; 6, 9. [DOI: https://dx.doi.org/10.3390/drones6010009]
28. Liu, Y.; Wang, C.X.; Chang, H.; He, Y.; Bian, J. A novel non-stationary 6G UAV channel model for maritime communications. IEEE J. Sel. Areas Commun.; 2021; 39, pp. 2992-3005. [DOI: https://dx.doi.org/10.1109/JSAC.2021.3088664]
29. Chang, H.; Wang, C.X.; Liu, Y.; Huang, J.; Sun, J.; Zhang, W.; Gao, X. A novel nonstationary 6G UAV-to-ground wireless channel model with 3-D arbitrary trajectory changes. IEEE Internet Things J.; 2020; 8, pp. 9865-9877. [DOI: https://dx.doi.org/10.1109/JIOT.2020.3018479]
30. Mao, K.; Zhu, Q.; Song, M.; Hua, B.; Zhong, W.; Ye, X. A geometry-based beamforming channel model for UAV mmWave communications. Sensors; 2020; 20, 6957. [DOI: https://dx.doi.org/10.3390/s20236957] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33291413]
31. Cui, Z.; Guan, K.; He, D.; Ai, B.; Zhong, Z. Propagation modeling for UAV air-to-ground channel over the simple mountain terrain. Proceedings of the 2019 IEEE International Conference on Communications Workshops (ICC Workshops); Shanghai, China, 20–24 May 2019; IEEE: Piscataway, NJ, USA, 2019; pp. 1-6.
32. Rappaport, T.S.; Heath Jr, R.W.; Daniels, R.C.; Murdock, J.N. Millimeter Wave Wireless Communications; Pearson Education: London, UK, 2015.
33. Akdeniz, M.R.; Liu, Y.; Samimi, M.K.; Sun, S.; Rangan, S.; Rappaport, T.S.; Erkip, E. Millimeter wave channel modeling and cellular capacity evaluation. IEEE J. Sel. Areas Commun.; 2014; 32, pp. 1164-1179. [DOI: https://dx.doi.org/10.1109/JSAC.2014.2328154]
34. Hemadeh, I.A.; Satyanarayana, K.; El-Hajjar, M.; Hanzo, L. Millimeter-wave communications: Physical channel models, design considerations, antenna constructions, and link-budget. IEEE Commun. Surv. Tutorials; 2017; 20, pp. 870-913. [DOI: https://dx.doi.org/10.1109/COMST.2017.2783541]
35. Azevedo, J.A.; Santos, F.E. A model to estimate the path loss in areas with foliage of trees. Aeu-Int. J. Electron. Commun.; 2017; 71, pp. 157-161. [DOI: https://dx.doi.org/10.1016/j.aeue.2016.10.018]
36. Castellanos, G.D.; Teuta, G. Urban-Vegetation ratio evaluation for Path Loss model in Amazonian region for Television Bands. Proceedings of the 2017 47th European Microwave Conference (EuMC); Nuremberg, Germany, 10–12 October 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 699-702.
37. Alraih, S.; Nordin, R.; Samah, A.A.; Shayea, I.; Abdullah, N.F. A survey on handover optimization in beyond 5G mobile networks: Challenges and solutions. IEEE Access; 2023; 11, pp. 59317-59345. [DOI: https://dx.doi.org/10.1109/ACCESS.2023.3284905]
38. ITU. Attenuation in Vegetation. ITU-R Recommendation P.833-4 2021.; Available online: https://www.itu.int/rec/R-REC-P.833-4-200304-S/en (accessed on 1 July 2023).
39. Hejselbæk, J.; Nielsen, J.Ø.; Fan, W.; Pedersen, G.F. Empirical study of near ground propagation in forest terrain for internet-of-things type device-to-device communication. IEEE Access; 2018; 6, pp. 54052-54063. [DOI: https://dx.doi.org/10.1109/ACCESS.2018.2871368]
40. Rougerie, S.; Israel, J.; Kan, T. Validation of ITU-R P. 833–9 tree attenuation model for Land Mobile Satellite propagation channel at Ku/Ka band. Proceedings of the 2021 15th European Conference on Antennas and Propagation (EuCAP); Dusseldorf, Germany, 22–26 March 2021; IEEE: Piscataway, NJ, USA, 2021; pp. 1-5.
41. Uwaechia, A.N.; Mahyuddin, N.M. A comprehensive survey on millimeter wave communications for fifth-generation wireless networks: Feasibility and challenges. IEEE Access; 2020; 8, pp. 62367-62414. [DOI: https://dx.doi.org/10.1109/ACCESS.2020.2984204]
42. Yun, Z.; Iskander, M.F. Ray tracing for radio propagation modeling: Principles and applications. IEEE Access; 2015; 3, pp. 1089-1100. [DOI: https://dx.doi.org/10.1109/ACCESS.2015.2453991]
43. Castellanos, G.; De Beelde, B.; Plets, D.; Martens, L.; Joseph, W.; Deruyck, M. Evaluating 60 GHz FWA Deployments for Urban and Rural Environments in Belgium. Sensors; 2023; 23, 1056. [DOI: https://dx.doi.org/10.3390/s23031056] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36772094]
44. De Beelde, B.; De Beelde, R.; Tanghe, E.; Plets, D.; Verheyen, K.; Joseph, W. Vegetation loss at D-band frequencies and new vegetation-dependent exponential decay model. IEEE Trans. Antennas Propag.; 2022; 70, pp. 12092-12103. [DOI: https://dx.doi.org/10.1109/TAP.2022.3209215]
45. Al-Nuaimi, M.; Stephens, R. Measurements and prediction model optimisation for signal attenuation in vegetation media at centimetre wave frequencies. IEE Proc.-Microwaves Antennas Propag.; 1998; 145, pp. 201-206. [DOI: https://dx.doi.org/10.1049/ip-map:19981883]
46. Weissberger, M. An Initial Critical Summary of Models for Predicting the Attenuation of Radio Waves by Trees [Final Report]. 1982; Available online: https://ui.adsabs.harvard.edu/abs/1982ecac.rept...W/abstract (accessed on 20 August 2023).
47. Zhang, P.; Yang, B.; Yi, C.; Wang, H.; You, X. Measurement-based 5G millimeter-wave propagation characterization in vegetated suburban macrocell environments. IEEE Trans. Antennas Propag.; 2020; 68, pp. 5556-5567. [DOI: https://dx.doi.org/10.1109/TAP.2020.2975365]
48. Zhang, Y.; Anderson, C.R.; Michelusi, N.; Love, D.J.; Baker, K.R.; Krogmeier, J.V. Propagation modeling through foliage in a coniferous forest at 28 GHz. IEEE Wirel. Commun. Lett.; 2019; 8, pp. 901-904. [DOI: https://dx.doi.org/10.1109/LWC.2019.2899299]
49. Van Dooren, G.; Govaerts, H.; Herben, M. COST 235: Radiowave Propagation Effects on Next-Generation Fixed-Services Terrestrial Telecommunications Systems. 1997; Available online: https://op.europa.eu/en/publication-detail/-/publication/1b803f14-c913-11e6-ad7c-01aa75ed71a1 (accessed on 11 May 2023).
50. Lagrone, A. Propagation of VHF and UHF electromagnetic waves over a grove of trees in full leaf. IEEE Trans. Antennas Propag.; 1977; 25, pp. 866-869. [DOI: https://dx.doi.org/10.1109/TAP.1977.1141708]
51. Leonor, N.R.; Fernandes, T.R.; Sánchez, M.G.; Caldeirinha, R.F. A 3-D model for millimeter-wave propagation through vegetation media using ray-tracing. IEEE Trans. Antennas Propag.; 2019; 67, pp. 4313-4318. [DOI: https://dx.doi.org/10.1109/TAP.2019.2905957]
52. Behjati, M.; Nordin, R.; Zulkifley, M.A.; Abdullah, N.F. 3D global path planning optimization for cellular-connected UAVs under link reliability constraint. Sensors; 2022; 22, 8957. [DOI: https://dx.doi.org/10.3390/s22228957] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36433554]
53. Singh, S.; Chawla, M. A Review on Millimeter Wave Communication and Effects on 5G System. Int. Adv. Res. J. Sci. Eng. Technol.; 2017; 4, [DOI: https://dx.doi.org/10.17148/IARJSET.2017.4705]
54. Rani, P.; Sharma, R. Intelligent transportation system performance analysis of indoor and outdoor internet of vehicle (iov) applications towards 5g. Tsinghua Sci. Technol.; 2024; 29, pp. 1785-1795. [DOI: https://dx.doi.org/10.26599/TST.2023.9010119]
55. Yamakura, H.; Ching, G.S.; Kishiki, Y.; Sekino, N.; Oshima, I.; Imai, T. Measurement and Ray Tracing Simulation with Urban Microcell Environments at 28GHz Band. Ieice Trans. Commun.; 2022; 105, pp. 748-756. [DOI: https://dx.doi.org/10.1587/transcom.2021ISP0006]
56. Sheikh, M.U.; Ruttik, K.; Saba, N.; Mutafungwa, E.; Jäntti, R.; Hämäläinen, J. Analysis of drone propagation with ray tracing from Sub-6 GHz upto terahertz frequencies in a real world urban environment. Proceedings of the 2021 International Conference on Information Networking (ICOIN); Jeju, Republic of Korea, 13–16 January 2021; IEEE: Piscataway, NJ, USA, 2021; pp. 169-174.
57. Tang, H.; Yang, N.; Zhang, Z.; Du, Z.; Shen, J. 5G NR and Enhancements: From R15 to R16; Elsevier: Amsterdam, The Netherlands, 2021.
58. Pagliari, E.; Davoli, L.; Cicioni, G.; Palazzi, V.; Ferrari, G. On UAV Terrestrial Connectivity Enhancement through Smart Selective Antennas. J. Phys. Conf. Ser.; 2024; 2716, 012057. [DOI: https://dx.doi.org/10.1088/1742-6596/2716/1/012057]
59. Norhashim, N.; Kamal, N.M.; Shah, S.A.; Sahwee, Z.; Ruzani, A.A. A review of unmanned aerial vehicle technology adoption for precision agriculture in Malaysia. Unmanned Syst.; 2023; pp. 1-19. [DOI: https://dx.doi.org/10.1142/S230138502450016X]
60. ITU, I.T.U. Effects of Building Materials and Structures on Radiowave Propagation above about 100 MHz. ITU-R Recommendation P.2040-3. 2023; Available online: https://www.itu.int/rec/R-REC-P.2040-3-202308-I/en (accessed on 10 October 2023).
61. ITU. Electrical Characteristics of the Surface of the Earth. ITU-R Recommendation P.527-6.; 2021; Available online: https://www.itu.int/rec/R-REC-P.527-6-202109-I/en (accessed on 8 September 2023).
62. ITU, I.T.U. Specific Attenuation Model for Rain for Use in Prediction Methods. ITU-R Recommendation P.838-3. 2005; Available online: https://www.itu.int/rec/R-REC-P.838-3-200503-I/en (accessed on 15 July 2023).
63. ITU, I.T.U. Attenuation by Atmospheric Gases and Related Effects. ITU-R Recommendation P.676-13. 2022; Available online: https://www.itu.int/rec/R-REC-P.676-13-202208-I/en (accessed on 12 July 2023).
64. Banafaa, M.; Pepeoğlu, Ö.; Shayea, I.; Alhammadi, A.; Shamsan, Z.; Razaz, M.A.; Alsagabi, M.; Al-Sowayan, S. A comprehensive survey on 5G-and-beyond networks with UAVs: Applications, emerging technologies, regulatory aspects, research trends and challenges. IEEE Access; 2024; 12, pp. 7786-7826. [DOI: https://dx.doi.org/10.1109/ACCESS.2023.3349208]
65. Stöcker, C.; Bennett, R.; Nex, F.; Gerke, M.; Zevenbergen, J. Review of the current state of UAV regulations. Remote. Sens.; 2017; 9, 459. [DOI: https://dx.doi.org/10.3390/rs9050459]
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Abu-Samah, Asma 1
; Abdullah, Nor Fadzilah 1
; Nadhiya Liyana Mohd Kamal 2
1 Department of Electrical, Electronic and Systems Engineering, Universiti Kebangsaan Malaysia, Bangi 43600, Selangor, Malaysia;
2 Unmanned Aerial System Research Laboratory, Malaysian Institute of Aviation Technology, Universiti Kuala Lumpur, Dengkil 43900, Selangor, Malaysia;