Next Article in Journal
Layer Thickness Impact on Shock-Accelerated Interfacial Instabilities in Single-Mode Stratifications
Previous Article in Journal
Mapping 18F-FDG Positron Emission Tomography Uptake in the Aortic Wall and Thrombus: Validation and Reproducibility
Previous Article in Special Issue
A GNSS–Cellular Network Hybridization Strategy for Robust Positioning
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Data Sources for Traffic Analysis in Urban Canyons—The Comprehensive Literature Review

Faculty of Civil Engineering, Wrocław University of Science and Technology (Politechnika Wrocławska), 50-370 Wrocław, Poland
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(19), 10686; https://doi.org/10.3390/app151910686
Submission received: 22 August 2025 / Revised: 23 September 2025 / Accepted: 1 October 2025 / Published: 3 October 2025
(This article belongs to the Special Issue Mapping and Localization for Intelligent Vehicles in Urban Canyons)

Abstract

We propose a comprehensive literature review based on big data and V2X research to find promising tools to detect vehicles for traffic research and provide safe autonomous vehicle (AV) traffic. Presented data sources can provide real-time data for V2X systems and offline databases from VATnets for micro- and macro-modeling in traffic research. The authors want to present a set of sources that are not based on GNSS and other systems that could be interrupted by high-rise buildings and dense smart city infrastructure, as well as review of big data sources in traffic modeling that can be useful in future traffic research. Both reviews findings are summarized in tables at the end of the review sections of the paper. The authors added propositions in the form of two hypotheses on how traffic models can obtain data in the urban canyon connected environment scenario. The first hypothesis uses Roadside Units (RSUs) to retrieve data in similar ways to cellular data in traffic research and proves that this source is data rich. The second one acknowledges Bluetooth/Wi-Fi scanners’ research potential in V2X environments.

1. Introduction

Today, dense urban areas are recognized as tall buildings with minimal distance between the frontages. Such a way of urban planning reminds one of canyons. Buildings can interfere with the coverage of GNSS and BTSs (Base Transceiver Stations) of streets. However, as smart cities develop with IoT devices, the devices used by the city infrastructure can disturb itself on the same frequencies or spectrums, which the researchers consequently acknowledge. A wide variety of V2X and IoT-based systems used in cities can provide valuable data for transport modeling. The pinpoint positioning of autonomous vehicles can provide sufficient data for micro-modeling. Therefore, models will be used to further improve the capacity of the infrastructure. In Section 5 we proposed two hypotheses based on the review. Both hypotheses consider the usage of a connected environment.
In the review, the recently commonly used IoT technologies described and proposed in the literature are reviewed, such as Bluetooth, DSRC, IEEE 802.11p WAVE, Cellular V2X (considering usage of LTE, 4G, 5G), and Wi-Fi, as well as state-of-the-art technologies that are not in use in transportation modeling nowadays, however, with research potential such as ZigBee and LoRa, proposed in state-of-the-art literature. The technology selection does not include every technology but considers technologies with promising capabilities for big data research in transport modeling.
The purpose of the review in Section 4 is to find patterns in big data research for traffic modeling purposes; the data sources reviewed are Bluetooth, Wi-Fi, Cellular Data, GNSS data, Taxi data, E-scooter fleets, and Shared Bike fleet data.
V2X can consider many forms of communication, including Vehicle-to-Grid (V2G), Vehicle-to-Home (V2H), Vehicle-to-Building (V2B), Vehicle-to-Infrastructure (V2I), Vehicle-to-Load (V2L), and Vehicle-to-Vehicle (V2V) [1]. The focus of this paper is placed mainly on V2X itself, V2V, and V2I as promising sources for traffic engineering and transport modeling. Due to the number of abbreviations used in the paper, the abbreviations used in the paper are given directly in the text and also listed in the Abbreviations section at the end of the paper.

2. The Review Methodology and Structure

2.1. Meaning of Nomenclature in the Review

Firstly, a precise definition of the nomenclature used for the review should be made. Some terms used could be wrongly understood from a different point of view, as the review considers multidisciplinary research. An Autonomous Vehicle (AV) refers to a fully automated vehicle, according to the Society of Automotive Engineers (SAE), including level 5 [2], German BASt (Bundesanstalt für Straßenwesen—Federal Road Office)—level 5 [3], for NHTSA (National Highway Traffic Safety Administration), also corresponding level 5 [4]. As a traffic model, the micro and macro models are under consideration.

2.2. Selection Procedure and Classification of Articles

As mentioned in Section 1, not all of the potential technologies are listed in this paper. As traffic engineers, the authors focused on technologies known to have potential in traffic engineering or that are promising data sources for transport modeling in the future.
Keywords used for the review in Section 3:
  • v2x Bluetooth
  • v2i Bluetooth
  • v2x ZigBee
  • v2x urban canyon
  • Wi-fi urban canyon v2x
  • LoRa urban canyon v2x
  • DSRC v2x
  • IEEE 802.11p urban canyon v2x
  • Cellular v2x
  • LoRa v2x
Keywords used for the review in Section 4:
  • Cellular data record big data transportation
  • Cellular data record OD matrix
  • Bluetooth big data transportation
  • Bluetooth OD matrix
  • Wi-Fi big data transportation
  • Wi-Fi OD matrix
  • Wi-Fi big data transportation
  • Wi-Fi OD matrix
  • GPS big data transportation
  • GPS OD matrix
  • Taxi big data transportation
  • Taxi OD matrix
  • E-scooter big data transportation
  • E-scooter OD matrix
  • Bike big data transportation
  • Bike OD matrix

2.3. Publication by Year

The paper includes publications from 2009 to 2025. Figure 1 shows a chart showing the number of articles by year included in the review. The authors’ goal was to achieve a review of state-of-the-art technologies, and the structure of the publication years indicates that. The most publications were from the year 2024—20 positions, and 2020—19. The approximately two-and-a-half-year period, as at the time of writing this paragraph, is August 2025 up to 2023, consisting of 30.23% of the articles. The almost five-year publishing period from 2020 to 2025 consists of 62,02% of the articles. Figure 1 shows publication number in the horizontal axis and the years published in the vertical axis.

2.4. Paper Classification by Research Method

In this subsection, research is divided into four groups: experimental papers that concern real-world text, papers that use simulations to evaluate proposed V2X-related methods, papers that are related to big data research in traffic engineering, and the fourth group—others that are not suitable for classification to previous groups, such as databases, law documents, other types of documents that are not scientific work, web pages, and one study applicable to the discussed technology, without V2X purpose. For the first group, “Experimental V2X” articles with field tests among the simulations presented are included. The numbers of all groups are presented in Figure 2. The highest publication number is in the third group that includes big data traffic research, with a number of 52. However, the first two groups concerning V2X research consist of 71 papers.

2.5. The Review Methodology

The review methodology was based on keywords listed in Section 2.2; the search engine used was mostly Google Scholar as it gave much better results than the authors’ university finding tools. The parameters used was “Sort by relevance”, “Any type”, “Include citations”. However, in case of older papers found, the option “Since 2021” was used. Paper language was restricted to only English-language publications.

2.6. Structure of the Paper

The paper has six sections. The comprehensive review evolves from V2X technology solutions and data acquisition by big data traffic research in cities to form hypotheses about the future of traffic modeling in urban canyon connected environments using data-rich V2X systems. Section 1 is an introduction about the research area. Section 2 covers the methodology and structure of the review and papers. Section 3 is a review about IoT technologies used, and proposed, in V2X systems with focus on urban canyon application and obstacles. Section 4 is a review about traffic modeling big data solutions. Both Section 3 and Section 4 end with tables summarizing the technologies. Section 5 covers two hypotheses about traffic modeling future applications supported by V2X system data analysis. Section 6 is the conclusion of the review.

3. Recently Used IoT Technologies Capable of V2X Systems in Urban Canyons

Historical trends in the automotive industry have predicted rapid growth in the percentage of AVs in traffic [5]. Therefore, to ensure safe and fluent traffic, many technologies are proposed in the literature for V2X purposes. In this review, and reviews in the following sections of the paper, the focus is put not only on types of communication and system characteristics, but also on how the technologies perform in the urban canyons. For example, a real-world simulation performed in [6] prove that GNSS-based systems are not suitable for urban canyons. The study recognized a drop in the Position Dilution of Precision, higher error, and multipath bias in urban canyon scenarios, sharply reducing the effectiveness of V2X. Ref. [7] has the same conclusion, also supported by RTK (Real-Time Kinematic) tests, by using it in open road scenarios—the mean horizontal error was 0.49 m and maximum horizontal error 0.83 m—and in urban canyon scenarios, where values increased to 1.46 m and 3.68 m. That could be enough for lane-level accuracy. However, e.g., under the viaducts, the position error could be several times higher, making a GNSS-only-based solution inadequate for urban canyons. In [8], ray-tracing-based 3DMA GNSS and factor graph optimization reduced positioning error from >30 m to <10 m in urban canyons; however, it states that standard double difference methods cannot eliminate location-specific multipath/NLOS errors. The PhD dissertation in [9] proposes Ultra-Wideband ranging for GNSS, resulting in a horizontal error of 0.6 m to 1.1 m in an urban canyon scenario. Finally, ref. [10] summarizes that GNSS technologies are not suitable for urban canyons and proposes other solutions, such as V2X-based localization, waveform-based positioning, sharing one’s own in-vehicle GNSS measurements, inertial data, and V2X-observed measurements for improving position accuracy or usage of passive RF (Radio Frequency) backscatter tags on infrastructure. At the end of the section, findings are summarized in Table 1.
The first technology searched in the review is Bluetooth. It is proposed as a safe and proven low-range communication technology [11]. Bluetooth version 5 allowed consideration of the use of V2X. According to [12], the version provided twice the speed, four times the higher range, eight times the advertising capacity, and was optimized for low energy. The low range of Bluetooth could be a disadvantage of the proposed communication system. However, in urban canyons, long-range signals could be disturbed by objects and tall buildings. The disadvantage is less serious, and Bluetooth can be considered. Decentralized, ad hoc networks known as piconets, which operate across 79 distinct frequencies, are resistant to interference from dense urban radio signals. According to [13], because of Adaptive Frequency Hopping, which avoids interfered channels, only half of the 79 channels are used in classical Bluetooth, and the authors proposed a method reducing occupancy by up to 50%. Bluetooth (BT) on its blog [14] shows readiness for applications in the V2X environment. State-of-the-art literature tested BT in V2X applications and estimated a range of 300 m used statically and in real-world highway scenarios around 1 km while driving [15]; however, this was done with degradation of throughput values, from 992.5 kbps at 0 m to 128.9 kbps at 100 m. In a further study [16], effective speed for sending and receiving messages is only 20 kmph using a stationary roadside device located 1 m above the ground, with a mobile device inside the vehicle. In [17], experimental results from dynamic testing demonstrate that reliable data reception is achievable at velocities up to 70 km/h, primarily due to the high frequency of advertising transmissions. Bluetooth is also proposed as a part of V2X systems [18] as a way of transferring messages from OBU to apps in smartphones of road users, where for safety-critical messages the DSRC is proposed. An open-source simulation platform using DSRC based on IEEE 802.11bd is proposed in [19]. The platform recognizes urban canyons in the detailed high-frequency model 3GPP NLOS. Another form of DSRC communication is between vehicles and pedestrians’ smartphones. In a simulated scenario [20], the system provides a Collision Detection Algorithm with safe delays (under 100 ms) even under congested scenarios.
Initial research proposed 5G as a novel technological standard for cellular data transmission. Currently, 5G is widely developed in many countries, with significant improvement over the previous 4G standard for LTE technologies. An interesting finding is in study [21], where LTE and 5G were compared in an intersection; 5G had a lower packet delivery ratio in longer distances—over 800 m, due to signal strength issues. With sufficient infrastructure, LTE can match 5G performance. However, ref. [22] shows that NR-V2X performs with a 20% better packet reception ratio and has up to 95% more power consumption reduction. The dissertation in [23], aiming to optimize 5G usage, proposed a K-means clustering algorithm for a novel distance metric for segregating road users, an allocation scheme called Vehicular Frequency Reuse that significantly reduces handover rates (over 99% reduction) and enhances link reliability for high-speed road users. Also, the dissertation proposes AI-based systems to improve uplink transmission while minimizing power consumption. One advantage of 5G technology is its ability to operate in underutilized frequency bands, such as those around 3 GHz, which are less congested compared to bands commonly used by Bluetooth, Wi-Fi, and ZigBee. Study [24] proposes antennas operating at 400 MHz channel bandwidth at a practical range of ~200 m at 28/38 GHz. The proposed system can support up to 1 Mbps downlink and 25 Mbps uplink for vehicle speeds up to 250 km/h. Guidelines and architecture for 5G-based V2X in [25] proposed a requirement of 10 ms latency (only 5 ms for remote driving). Although 5G is reliable in urban canyon scenarios, it needs up to an additional 3 dB extra transmitter power [26]. Real-world tests conducted in [27] showed blockages impact 5G signal (at 28 GHz) with up to 7.5 dB loss for one car blockage, 2 dB loss through clear windows, but 15 dB through sun-protective film; reflection from walls adds only about 3 dB loss, and diffraction and foliage can cause loss over 20 dB. Three-dimensional ray-tracing matches measurement within ~4 dB. NLOS can be effectively detected using a high-accuracy deep learning model [28]. The state-of-the-art model achieved 93.5% accuracy in distinguishing between NLOS and LOS conditions, surpassing the performance of earlier approaches. Additionally, 5G technology can complement GNSS-based positioning in urban canyon environments. Paper [29] introduces a hybrid approach incorporating Kalman filters, multi-rate data fusion, and mmWave technology. The study concludes that for SAE Level 3 and higher, the deployment of private 5G networks is essential.
In [30], researchers focus on Cellular V2X strictly in an urban canyon scenario, forming them by containers. The empirical test showed severe packet loss occurring below −110 dBm received power in tested scenarios—vehicle behind a container, and containers representing an open-topped alleyway of the urban canyon. A deep learning approach is used in [31], with K-means for addressing variable fading channel conditions and a Genetic Algorithm for optimizing two parameters: number of reflection points and reflection coefficient to match the model for actual field measurements. A comparative analysis between low and high urban canyon environments revealed that the low urban canyon exhibits a greater number of reflection points and a higher reflection coefficient. In [32], urban canyons were classified in four groups: low—with a building height of 1 to 20 m, medium—from 20 to 60 m, high—over 60 m, and tall—over 100 m buildings. The simulation proves that the higher the buildings, the more severe the multipath effects, signal reflections, and greater shadowing are. This finding is also present in [33], with in-chamber tests supported by field measurements. The study provided a 3D geometric-based irregular channel model that reflects urban conditions. An independent GNSS-based CV2X-LOCA system is proposed in [34], maintaining less than 4 m accuracy in real-world scenarios even in tunnels, beside skyscrapers, and under elevated roads. The system is based on the RSU every 120–150 m. The simulation experiment showed accuracy up to 2.5 m, even in the high-speed (125 km/h) scenario. An example of a C-V2X application could be the Multi-Sensor LiDAR [35], where the system aggregates and fuses point clouds from LiDAR sensors in multiple vehicles using machine learning for pedestrian detection. For the urban canyon scenario, a comparison study [36] proposes a combination of C-V2X and DSRC. The review of existing technologies in this paper notes that most commercial devices support only one communication technology, and from the urban canyon point of consideration, all use GNSS.
For V2I in [37], Cellular V2X (4G and 5G) was used, and for V2V, DSRC was used in urban canyon scenarios. The study employed a fusion of Cellular V2I, V2V, and INS as an alternative to GPS, achieving robust and accurate positioning. The proposed method demonstrated an 18.7% improvement in accuracy compared to the state-of-the-art solutions available at the time of publication. An interesting finding in [38] concludes that due to expected strong line-of-sight (high K-factor—7.46 dB in the case), long delay spread (of 63.8 ns), and frequency diversity, the urban canyon scenario could potentially lead to an improvement in performance of the system. This finding is also supported by [39], where an LTE link was tested. In the urban canyon scenario, LTE had stronger line-of-sight and less frequency dispersion in comparison to suburban and highway environments. However, as noted in [40], a testbed study of LTE V2V communication in urban canyon environments with varying building heights revealed that denser and taller canyons exhibit increased signal reflectivity. Consequently, signal attenuation and path loss are more pronounced, negatively impacting V2V performance metrics.
For the IEEE 802.11p WAVE standard simulation test, the researchers in [41] tested two scenarios, WAVE-CH07 (Urban Canyon, LOS) and WAVE-CH08 (Urban Canyon, NLOS), and proposed a new channel estimator (LSA-DD) that improves robustness by up to 3 dB SNR compared to prior methods. Reference [42] concluded research in three cities, mainly on the 60 GHz band; the study found that urban canyons yields high RMS delay spreads (up to 80 ns) due to strong, persistent reflections, where open streets with trees caused much less reverberation. In [43], the digital twin of the city is provided based on real-world measurements to test V2X systems based on the IEEE 802.11p WAVE standard. The 3D model incorporates building reflection and penetration losses using realistic ray-tracing techniques and empirically validated building attenuation models. A total of 60 GHz IEEE 802.11ad mmWave usage for V2I in urban canyon scenarios [44] has found that an RSU located at 5 m has an 8.5% blockage rate, but increasing the height by 10 m can eliminate over 99% of the blockages. For full coverage in an urban canyon, at least 2× more the current RSU density is needed at around 20 RSUs per square kilometer. As demonstrated in [45], mmWave performance in urban canyon environments can be further enhanced through machine learning-based beam alignment. Dynamic beam adjustments are particularly critical in scenarios where mmWave signals are obstructed by large vehicles or other transient obstacles. As a result, the beam search overhead is reduced by 70%. A framework for V2X combining several technologies, such as DSRC, LTE, 5G, NB-IoT, Bluetooth, GPRS, and mmWave, was proposed in [46], and in an urban canyon scenario, NLOS path loss can be extremely large after just one corner, especially for mmWave. The simulation performed in [47] states that RSUs significantly enhance object detection, reduce blind spots, and improve cooperative perception in ADAS systems. For transport modeling, RSUs can be a valuable source for travel data. Therefore, the need for RSUs in systems based on IEEE 802.11p shows promising insight for transport engineers.
In [48], Wi-Fi was selected over ZigBee as the V2X communication technology for autonomous golf carts designed for elderly passengers. The decision was driven by ZigBee’s insufficient bandwidth, which was inadequate for supporting video calls—an integral part of the system. In a similar scenario—using electric golf carts [49]—Wi-Fi was used; when the signal did not provide sufficient signal, 4G was successfully used as a backup. In [50], Wi-Fi is proposed as GNSS support along with IMU (Inertial Measurement Unit) in urban canyons with data fusion based on a Kalman filter, with a result of 1 to 10 m in the urban canyon scenario. State-of-the-art applications in five intersections in Michigan also used Wi-Fi. However, this was only for OBU-OBU communication [51]. In the NLOS scenario in the V2P usage proposed in [52], Wi-Fi performance is severely reduced with the maximum connection distance lowered to a maximum of 40 m due to long CET and signal attenuation. The beacon stuffing method was proposed to address connection issues, which allows reception of at least one BSM at 100 m. The presence of Wi-Fi, as a widely deployed technology in urban areas, can degrade the performance of C-V2X communication due to potential interference and spectrum congestion. A simulation performed in [53] indicates that the reliability range can be lowered up to 90% (20 m). Study [54] quantified the increased deployment costs resulting from Wi-Fi interference with ITS systems, which necessitate denser infrastructure. A similar issue may arise with the installation of Roadside Units (RSUs) in congested urban environments. However, from a transport modeling point of view, the denser the data sources, the more valuable detailed data is.
ZigBee operates on a 2.4 GHz frequency band. The band is the same for Bluetooth and Wi-Fi described in previous paragraphs of this paper. Reference [55] conducted a comprehensive real-world test of the ZigBee XBee S2C module, concluding 3 dBm is enough antenna power for applications, as more power does not give much better results but consumes more energy. Also, the study states that multi-hop extends coverage/RSSI but at the cost of latency, similarly to Advanced Encryption Standard; however, it keeps energy consumption reasonable. The maximum reliable range and antenna type were set in [56], where the estimated reliable range is 600 m. However, ref. [57] indicates RSSI drops after a 300 m distance. However, the system based on Arduino and XBee successfully gave warnings for risks such as close stops (less than 3 m), intersection approach, and unfavorable environmental conditions. Ref. [58] proposed vehicle collision warning with speed prediction with less than 0.3 m/s error and location with less than 0.3 m error. However, for location data, GPS or other satellite-based systems are proposed, which could be interrupted in an urban canyon scenario, like another example using GPS and ZigBee for V2V in a low-speed scenario [59]. An algorithm was proposed to optimize speed to reduce stops and idling. At the same time, for crossing intersections at a green light [60], ZigBee was used for communication between OBU and RSU; however, this was done with GPS usage for speed and location estimation. Hybrid approaches are also possible; for example, using a combination of GNSS data, IMU, speedometer, and 3D maps with Kalman filter usage [61], achieving accuracy of 1.5 m in an urban canyon with a 40% error rate. The scenario presented in [62] is suitable for urban canyons. The study presents a system based on ZigBee for short-range and LoRa for long-range communications. The framework enables RSUs to track vehicle location without GNSS-based data. Another study proposed a ZigBee/Wi-Fi/BT RSSI-based location system, allowing for obtaining precise (less than 1m) location data without using a GPS node-to-node [63]. Therefore, the ZigBee-based system can be used in urban canyons. Another concern could be 2.4 GHz band usage in cities, widely used for Bluetooth and Wi-Fi. Study [64] proposed 900 MHz as an alternative with a range benefit over 2.4 GHz for V2V uses.
LoRa’s official range is 500 m with line-of-sight. However, in [65], researchers tested the LoRa obstacle environment, and the range dropped to 160 m during V2V and V2I communication tests. On the other side, ref. [66] states that LoRa provides excellent penetration through urban obstacles, maintaining connectivity with packet error rates below 10%. Also, the study highlights low cost can power consumption over DSRC and C-V2X. A system based on LoRa, achieving 100% packet delivery, was proposed in [67]. High delivery in real-world tests was achieved through device-to-device direct communication supported by a Kalman filter. A stable connection was provided in the urban canyon scenario, but with affected line-of-sight and RSSI stability. Worth noting is study [68], which proposes guidelines on how to operate with bandwidth and spreadfactor to resist Doppler and fast fading. In the NLOS canyon scenario, LoRa is viable for non-safety V2X applications up to 58 km/h [69]. Higher speeds do not correspond with the transport policy of modern cities focused on lowering accident casualties; therefore, SF-12, with lower RSSI at higher speeds, tested in the paper with comparison to SF-7, should be sufficient for the urban canyon scenario. Moreover, SF-12 performs better in NLOS scenarios, with better detectability of weaker signals. It is important to mention LoRa’s far superior capabilities in UAV applications in comparison to Bluetooth, Zigbee, SigFox, and 802.11p, managing to provide V2V communication over 10km range in urban areas [70].
Data from V2X systems will be possible to acquire from VANets. In [71], a Software-Defined Networking (SDN) approach to VANets is proposed. SDN improves programmability, scalability, and flexibility. The study features real-time and historical data collection for analysis. The data collected are location, speed, and connectivity status. Ref. [72] indicates that VANets can provide GPS coordinates. Ref. [71] adds that VANets can also collect trajectory, traffic density, and traffic flow patterns. VANets data is also applicable to urban planning and infrastructure planning usage. For this case, the Infrastructure Domain type of VANet architecture is more useful, as it communicates between OBUs and RSUs. RSUs will provide data from OBUs to the central servers in the case. An ad hoc domain connecting OBU to OBU will not provide sufficient information for traffic modeling. Traffic-oriented VANet architecture is also present in [73], where Congestion-Free Path (CFP), and Optimized CFP are added to SDN VANet. A real-world test of SDN is presented in [74], using Raspberry Pi and OpenFlow switches. OpenDaylight controller, in this case, allows for data collection via standardized OpenFlow messages. Another SDN-based solution, enhanced with a 3-stage fuzzy decision tree model, is proposed to aid traffic signal optimization in [75]. In this case, data is gathered from the vehicles via VANet to RSU. RSUs standardize and preprocess the data. Data is processed by central or fog-based fuzzy decision models. Machine learning approaches in VANet architecture can also result in a rich data source for traffic modeling, as it requires data acquisition for the learning process. In [76], vehicles and RSUs collect the data, the blockchain layer verifies the identity and reputation of the data source, then the Neuro-Fuzzy server aggregates data and performs fuzzy logic-based decision making, and a distributed deep neural network needs the data collected to train.

4. Traffic Modeling—How V2X Technologies Provide Data in Urban Canyons, Based on Actual Big Data Research

To anticipate how traffic modeling will evolve in future urban canyons, this section summarizes recent research on the use of big data sources in traffic modeling. Based on this review, a hypothesis is formulated in the following section.
Traffic modeling uses various types of data. This section is a review of data collection and usage by transport engineers. The technologies mentioned are already used/proposed as V2X, and advanced research consists of Bluetooth/Wi-Fi usage. This data source is independent of specific industries, as the information is collected using Bluetooth and Wi-Fi scanners. Spatiotemporal data enables the initial acquisition of origin–destination matrices for travel patterns. Based on the detected locations, it is possible to reconstruct the traveler’s path; by analyzing the time intervals between points, travel speed can be determined, which in turn allows for the identification of the mode of transportation. With census data and detection times, travel origins and destinations can be classified as home, work, or other to acquire travel motivation. With such data gained, transportation models and analysis based on the models are possible. BT/Wi-Fi scanners located on roads search for MAC addresses of Bluetooth devices in cars or phones with visibility mode on. Therefore, even when Wi-Fi or Bluetooth is used solely for OBU-to-OBU communication, it is still possible to obtain the necessary data. The least advanced BT scanner usage was traffic measurements with an average speed measure in the road section [77]. Alternatively, Wi-Fi scanners could be used to obtain MAC addresses. However, despite the high hopes in previous work, with usage of a Bluetooth scanner [78] has already demonstrated significantly lower data usability than address detection using Bluetooth. The effectiveness of Bluetooth technology used for this purpose increases with version 5.0 [79]. Experiments conducted in three cities [80] shows that in comparison with RADAR detection, Bluetooth scanners have only 62% detection accuracy. An interesting finding in the study was that Bluetooth detection performed better in dense urban areas. That finding is crucial, referring to urban canyon usage. State-of-the-art research [81] allows for the grouping of detected travelers, which may be crucial in using scanner data to study travelers’ transportation behaviors. The use of Bluetooth scanners in crowd research was investigated in [82]. In the case of determining travel matrices, which are still not precise, the problem was addressed in [83]. Paper [84] studied a dedicated group of students. The specific research group allowed the focus to be placed on detecting the means of transport using Bluetooth/Wi-Fi scanners. The study differentiated between five means of transport. In [85], it was shown how to use Bluetooth to determine the distribution of trips on a transport network. Report [86] showed that traditional methods are more economically advantageous for continuous measurements throughout the year. When congestion measurements are planned for a longer period, the use of Bluetooth scanners is justified, even in an indoor scenario [87], with a novel method for collecting and analyzing Bluetooth traffic during university exams using the Ubertooth One device. The method captured 263 unique Bluetooth MAC addresses across five sessions, classified traffic confidence levels, filtered out noise using RSSI thresholds, and localized devices using trilateration and heatmaps. In the urban area where the use of Bluetooth scanners was commissioned and a series of studies were being conducted is Brisbane, Australia, ref. [88] mentions a network of 400 Bluetooth scanners there. In [89], the Brisbane scanner network use was to combine scanner data with data from smartcards. The Brisbane scanners arrangement is shown in studies such as [90,91,92]. The possible bias of data based on socioeconomic factors is mentioned in [93]. A paper that used Bluetooth scanners and GPS data to verify travel distribution is [94]. Despite arguably low usage of GPS in urban canyons, multiple papers in the section above, including experimental ones, made attempts to use GPS data in urban canyons, with successful results. Therefore, studies combining both V2X urban canyon sources is promising for traffic modeling in urban canyons.
Depending on the public or private RSU owners, adequate data can be similar as from the BTS (Base Transceiver Station). However, in urban canyons, the Transport Analysis Zones (TAZ), according to the review in previous sections, will be smaller, with a denser RSU grid (even around every 120 m) to ensure proper wave propagation and connection stability. With 120 m between RSUs, a transportation zone grid will consist of thousands of transport zones, where an example study using mobile data in Stokholm had only 32 zones of the entire city [95]; for example, much smaller Wrocław (Poland) in its Comprehensive Traffic Research [96] is divided into 475 transport zones, where mobile phone data was used as support to classical methods—surveys and traffic measurements. Wrocław’s Comprehensive Traffic Research was analyzed further in [97]. The establishment of TAZ borders using K-means clustering was described in [98]. Using the mobile phone network in [99], trips could be divided into classic types related to home, work, and other activities. Ref. [100] proposed software that combines data from mobile phone networks with traditional data in the form of surveys, censuses, and publicly available geolocation data. The frequency of updates for the area in which the traveler is located depends on the database used. In [101], attention was drawn to the existence of two types of databases. The first is a billing database, which is biased because it depends on the frequency of mobile phone calls. This will generate a bias correlated, for example, with the age of the traveler. Accordingly, a signal database was utilized to record changes in the base station coverage area in which the traveler is located. The location is verified every six hours, or every three hours in the case of newer systems. Analysis of this type of data allows the traveler’s route to be estimated. The ping-pong effect and the bias associated with the socio-demographic characteristics of mobile phone users are also discussed there. Data from mobile network operators were also compared with traditional survey data. In [102], the same research area as in [101] was taken up, namely the Paris agglomeration, with particular attention paid to the problem of detecting travelers moving by metro. For an urban canyon V2X scenario, blind spots similar to that would be solved just by using RSUs that are not GNSS-based; proposed solutions are in the Section above. The comprehensive review [103] summarizes eight publicly available databases from mobile phones with five uses, including transportation in the form of obtaining OD matrixes, traffic congestion detection, urban planning, passenger hub analysis, and transport mode detection by using rule-based heuristics, fuzzy logic, Partitioning Around Medoids, Random Forest, Pipe-based trip inference, and the classical Speed-based Classification. These methods can be used to detect the transport mode in case of partially available databases from RSU operators in urban canyons, as sometimes law regulations require anonymizing data (e.g., in the EU by RODO [104]). The usage of mobile phone databases requires including bias related to age, as with age fewer people use smartphones [105]. With the use of RSUs in urban canyons by the same method, the bias should be redefined for the use of autonomous vehicles.
The study in [106] addressed the issue of problems related to measuring passengers underground. The main purpose was to compare data from travelers’ mobile phones’ GPS with data from AVL (automatic vehicle location) systems located in motor vehicles. However, this was also based on GPS. GNSS-based data in an urban scenario, according to [107], has an up to 51% success rate in identifying travelers’ home locations, and according to the study, still has better coverage and accuracy than traditional methods, such as surveys. A Bayesian network can be used to estimate the O-D matrix, ensuring GPS-based trajectory data and traffic counts [108]. The study also states that connected AV and Floating Car Data in the future can provide a sufficient big data source for traffic modeling. The study in [109] focused on determining how accurate GPS data must be to estimate vehicle speeds. Vehicle speeds can also be determined at the level of entire transport networks [110]. Bluetooth scanners were used as a reference, with speed measurements being their first application in traffic engineering, as mentioned above in the review. In general, GPS data can be detailed enough to be used in micro-level modeling; in [111], this data was used to assess the variability of drivers’ driving styles. Even with a small amount of data, for example, in the absence of connections to different base stations, GIS-based methods have been developed that increase the accuracy of determining a traveler’s location [112]. Using a neural network, it is possible to accurately determine the type of vehicle [112]. Vehicle GPS data are already used in transportation engineering. In dense urban environments, ref. [113] proposes the Conditional Wasserstein Generative Adversarial Network with Gradient Penalty (CWGAN-GP), predicting ride-sourcing OD matrices using GPS data obtained from taxi fleets in Chengdu (China) and New York City (USA). Another example of an advanced method—Randomized K-Nearest Neighbor Regression with Spherical Distance—is in [114] using New York City, Porto (Portugal), and San Francisco (USA) databases for real-time analysis, predicting taxi travel time and detecting and scoring speeding behavior, enhancing passenger safety. Ref. [115] analyzed GPS-based data from the Chicago Data Portal, including trip duration, distance, fare, and pickup/drop-off locations aggregated to census tracts. The study proposed a variety of recommendations for improving taxi policy by sociodemographic analysis of the passengers. The study [116] found that by using taxi data, this transportation node plays a vital role in serving disadvantaged populations and enhancing multimodal connectivity, especially for short, first/last-mile trips. Over one billion trips were analyzed in [117] in Manhattan in New York City, which is comparable to an urban canyon, and GNSS data was suitable in a directed weighted network, along with identification of key zones with introducing OD rank. The study was supported with a Monte Carlo simulation for verification.
Multiple data sources can reduce the impact of biases on results. For example, in [118], ANPR (Automatic Number Plate Recognition) data was used to reduce bias using data delivered from a taxi company. Ref. [119] focuses on the problem of bias in data from large databases. Although the publication primarily addresses traffic safety, it utilizes the same type of data, and the challenges encountered in traffic model development are comparable. The chapter also explores the application of specific data sources in modeling. The research methodology involved interviews with experts from academic, commercial, and governmental sectors.
In the case of bias in demographic data, the authors of [119] suggested that traditional research should be taken into account as a solution. The problem of data bias was also discussed in the book [120]. The largest source number found during the review was four in [121]—Bluetooth, loop detectors, navigation systems, and classical Census data. The study effectively addresses the urban canyon problem through multisource integration and quality weighting.
The review authors proposed the Personal Transporter [122] as a vehicle form in connected V2X environments in urban scenarios. Comparable vehicles would be e-scooters and bikes. Therefore, research including big data databases from e-scooters, such as [123], where the influence of the built environment and socioeconomic factors on e-scooter ridership was studied. The most interesting finding from an urban canyon research point of view is the correlation with higher ridership in a dense urban area. Therefore, that will be a valuable data source. Ref. [124] uses GPS data analysis and statistical models to uncover temporal patterns and policy implications for cities, comparing city bike fleets with e-scooter dockless fleets, by analyzing data by real-time API scraping from multiple vendors, gas prices, and weather data. This finding is interesting—a 1% gas price increase results in a 3.13% increase in e-scooter trips. To continue the bike fleets topic, ref. [125] used GPS data from three different bike fleets to analyze modal shift in Delft (Netherlands). User behavior was analyzed in [126] by a model combining local movements and long-range Lévy flight-like transitions, validated through Monte Carlo simulations. The study used two databases from Chicago and New York City (USA), containing over 40 million trips. While the study used end-to-end locations, a smaller study with bike usage in a university campus used GPS to retrieve point-to-point travel paths [127]. The paper identified areas for system expansion and suggested operational improvements, such as 6 h rebalancing, dynamic relocation, and incentive programs, as university bike sharing differs from city systems. With technological progress in a connected environment, dock stations for Personal Transporters can be virtual. The study in [128] analyzed travel patterns in Beijing (China) with a proposed clustering algorithm. The algorithm outperformed classical methods—K-means and MILP. The report in [129] indicates that shared bike fleets have “massive potential for expansion”, with electric bikes generating 30% of trips while taking up 21% of fleets; therefore, Personal Transporters connected to V2X systems will have the potential to research travel patterns with this micromobility transport mode. The study was possible to conduct by the use of data from Fluctuo, an aggregator of shared mobility data, data collected via open APIs, city partnerships, and operator agreements. In an urban canyon scenario, data would be collected by RSUs. The urban canyon will need sufficient transportation. Therefore, relying only on autonomous vehicles wouldn’t be effective as public transportation nodes. Big data GNSS data analysis [130] shows that bike fleets can support public transport, improving modal split in large cities. The more public transport connections, the more bike-shared fleet trips.

5. Traffic Modeling—A Hypothesis Based on the Review of Traffic Research in the Age of Connected Environment in Urban Canyons

5.1. RSU-Based Big Data Traffic Modeling

The review highlights technologies and systems that show promise for enabling safe and efficient V2X communication, even in complex urban canyon environments. This section integrates insights from both research domains to formulate a hypothesis regarding traffic modeling in the era of VANETs. From a traffic modeling point of view, urban canyons cause problems in efficiently planning and managing traffic flows, generated by the high-rise density grid of buildings, causing a high population density that must obtain nodes of transport with sufficient capacity. With limited space between buildings, it would not be possible to ensure the capacity of the road for private autonomous vehicles. Therefore, it is needed to support AVs by correctly planned and cost-efficient public transportation nodes. Transport models are necessary for planning transportation in urban areas. The review showed that big data used for transportation models results in precise and cost-efficient models. With connected V2X environments, more data sources will be available. However, as the review on V2X communication systems shows, urban canyon data sources can provide much more valuable and less biased data. An example can be a comparison of reviewed cellular data and data from RSUs. The methodology would be similar. However, RSUs will provide a more precise location. Figure 3 shows the TAZ Voronoi Diagram based on BTS locations in the Powstańców Śląskich district in Wrocław (Poland). An example case study street—also called Powstańców Śląskich (the district is named after the main street)—is shown as green lines in Figure 3. The street is a four-lane road with a tram track in the middle, with bike paths and pavements on both sides of the road, and trees. The street was chosen because of the high-rise buildings, up to over 200 m. The historical street buildings were destroyed during World War II, then rebuilt in Soviet residential style in the 1960s. In the recent 30 years, this street has been intensively densified with offices and high-rise apartment buildings. The BTSs are shown as red triangles. This representation varies from actual transmission ranges, as high-rise buildings interrupt signal. However, big data research uses zones according to the Voronoi Diagrams. Therefore, the example and figure are adequate.
In Wrocław’s Comprehensive Traffic Research, zones are not comparable to the BTS transmission range. The classical TAZs are set according to the natural barriers (rivers, lakes, hills), streets, railroads, and historical district borders. Figure 4 shows classical zones in Wrocław’s Comprehensive Traffic Research 2018. In Figure 4, the red rectangle shows the borders of Figure 3; the green line represents Powstańców Śląskich street’s section shown in Figure 3.
RSUs in a connected environment can provide the same type of data as BTSs. According to the review, the displacement of RSUs has to be a few times denser. Based on the review in the hypothetical case study, 120 m is assumed. RSUs are placed in two rows every 240 m. Therefore, from one to another is around 120 m. In Figure 5, the layout of the RSU is proposed on the street of the case study, with the Voronoi Diagram range for each RSU. The coverage in such a scenario is much more precise. Consequently, the data exhibits reduced bias, improved accuracy, and greater usability. TAZs derived from RSUs can be aligned with traditional zoning schemes, as illustrated in Figure 4, due to their significantly smaller size—typically a 120 m trapezoid, compared to conventional polygons that may span several dozen meters to over a kilometer. Furthermore, the increased spatial resolution of these zones enables the classification of trip origins and destinations based on land use, even within TAZs that serve mixed functions. For example, this approach allows for differentiation in areas such as the Powstańców Śląskich district, which includes both residential and commercial zones, as demonstrated in the case study. In the much smaller zones, it is possible to identify, e.g., the 8 am origin of the trip as work due to the location of the last connected RSU by the commuter, and the destination as home, due to a zone with only residential buildings.

5.2. 2.4 GHz Technologies Scanner-Based Big Data Traffic Modeling

The second example based on the review is the usage of Bluetooth/Wi-Fi scanners. During the review, numerous papers using at least one of these two technologies were found, mostly non-related to safety messages, V2P applications, etc. Therefore, vehicle or traveler detection can occur more frequently than nowadays. When fewer phone users have visibility mode on, and only cars with Bluetooth car kits are detected, detection of devices without visibility mode is possible. However, due to legal constraints, this aspect falls outside the scope of the present study. One key advantage of the approach based on Roadside Units (RSUs), as opposed to Base Transceiver Stations (BTSs), is its independence from private data owners. Consequently, research can be conducted without the need to purchase datasets. However, it comes with the cost of acquiring scanners. Therefore, the scanner layout will be much more spread than BTSs shown in Figure 5.
According to the literature examples from the review, an example BT/Wi-Fi scanner placement in the case scenario street is proposed in Figure 6. The scanner placement is shown as a red hexagon. The trips are recorded by scanning for identical MAC addresses of devices. For example, a trip along Powstańców Śląskich street would result in consecutive detections from scanners C-D-F-H, and a trip with an origin in the south-east of Figure 6 up to the northern end of the case study’s street would be an I-E-C-D-F-H trip. The trip with the destination in the middle of the street section would be C-D. Depending on the connected environment architecture, other technologies can be scanned. ZigBee is a promising source, as it operates similarly to BT and Wi-Fi at 2.4 GHz frequency.

6. Conclusions

In the comprehensive literature review, 131 sources are listed, mostly including research papers. The review methodology and structure of the papers are analyzed in Section 2. The purpose of the review was to introduce traffic research possibilities in urban canyon connected environment scenarios. For this purpose, a two-part review has been performed. The first review in Section 3 was to obtain a state-of-the-art point of view of V2X systems, VANet architecture, and data acquisition. The Section 4 review was performed to acquire state-of-the-art big data modeling knowledge. Findings from both of the sections are summarized in Table 1 and Table 2. Section 5 summarizes the obtained knowledge based on the literature review and, based on this, proposes two hypotheses on how big data research will develop in an urban canyon connected environment. The first hypothesis proposes using RSUs as data sources, with their placement determined by insights from the literature review and proposed in the case study. The second hypothesis proposes how Bluetooth/Wi-Fi scanners can be used in the urban canyon scenario, with the proposition of scanner locations in the case study street. Both hypotheses and the review prove that V2X systems in urban canyons will provide rich data sources for traffic modeling. The hypothesis from Section 5.1 shows that with a dense, around 120 m RSU grid, the data will be at a high level of accuracy and usability in traffic modeling. Research gaps and trends found in Section 3 review are as follows: BT integration with smartphones, GNSS support with Wi-Fi, GNSS-free localization in case of ZigBee usage, NLOS applications for 5G/NR-V2X and LoRa, RSU density optimization for C-V2X systems, and mmWave integration, beamforming, and digital twin modeling for DSRC. The research gap for traffic modeling is to obtain and base a case study on data delivered by existing connected vehicle environments and testbeds.

Author Contributions

Conceptualization, M.Z. and M.K.; methodology, M.Z.; software, M.Z.; validation, M.Z. and M.K.; formal analysis, M.K.; investigation, M.Z.; resources, M.Z.; data curation, M.Z.; writing—original draft preparation, M.Z.; writing—review and editing, M.Z. and M.K.; visualization, M.Z.; supervision, M.K.; project administration, M.Z.; funding acquisition, M.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
V2XVehicle-to-Everything
V2IVehicle-to-Infrastructure
V2VVehicle-to-Vehicle
V2PVehicle-to-Pedestrian
V2HVehicle-to-Home
V2BVehicle-to-Building
V2LVehicle-to-Load
V2GVehicle-to-Grid
GNSSGlobal Navigation Satellite System
GPSGlobal Positioning System
IoTInternet of Things
SAESociety of Automotive Engineers
DSRCDedicated Short-Range Communications
BTBluetooth
NHTSANational Highway Traffic Safety Administration
BAStBundesanstalt für Straßenwesen—Federal Road Office
ODOrigin–Destination
RTKReal-Time Kinematic
3DMA3D Map Aided
NLOSNon-Line-of-Sight
LOSLine-of-Sight
RFRadio Frequency
AFHAdaptive Frequency Hopping
OBUOn-Board Unit
RSURoad-Side Unit
NR-V2XNew Radio Vehicle-to-Everything
INSInternal Navigation System
NB-IoTNarrowband Internet of Things
ADASAdvanced Driver Assistance Systems
IMUInertial Measurement Unit
RSSIReceived Signal Strength Indication
BSMBasic Safety Message
UAVUnmanned Aerial System
BTSBase Transceiver Station
AVLAutomatic Vehicle Location
TAZTransport Analysis Zone
PTPersonal Transporter
MILPMixed-Integer Linear Programming
APIApplication Programming Interface
ANPRAutomatic Number Plate Recognition

References

  1. Kumar, G.; Mikkili, S. Critical Review of Vehicle-to-Everything (V2X) Topologies: Communication, Power Flow Characteristics, Challenges, and Opportunities. CPSS Trans. Power Electron. Appl. 2024, 9, 10–26. [Google Scholar] [CrossRef]
  2. J3016_202104: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles—SAE International. Available online: https://www.sae.org/standards/content/j3016_202104/ (accessed on 14 July 2025).
  3. Gasser, T.M.; Westhoff, D. BASt-Study: Definitions of Automation and Legal Issues in Germany. In Proceedings of the Road Vehicle Automation Workshop, Irvine, CA, USA, 25 July 2012. [Google Scholar]
  4. Levels of Automation. National Highway Traffic Safety Administration 2022. Available online: https://www.nhtsa.gov/sites/nhtsa.gov/files/2022-05/Level-of-Automation-052522-tag.pdf (accessed on 2 October 2025).
  5. Litman, T. Autonomous Vehicle Implementation Predictions: Implications for Transport Planning; Victoria Transport Policy Institute: Victoria, BC, Canada, 2019. [Google Scholar]
  6. Ramesh, R.; Prasanna Kumar, B.Y.N.; Manjanagowda, A.; Kumar, S.; Jain, R. Impact of GNSS Accuracy on V2X Applications. In Proceedings of the 2024 International Conference on Vehicular Technology and Transportation Systems (ICVTTS), Bangalore, India, 27–28 September 2024; Volume 1, pp. 1–6. [Google Scholar]
  7. Liao, Z.; Yu, B.; Kang, C.; Ge, Y.; Gong, Z. Research on GNSS and RTK Performance for V2X Application Scenarios. In Proceedings of the 2022 IEEE 8th International Conference on Computer and Communications (ICCC), Chengdu, China, 9–12 December 2022; pp. 499–504. [Google Scholar]
  8. Zhang, G.; Ng, H.-F.; Wen, W.; Hsu, L.-T. 3D Mapping Database Aided GNSS Based Collaborative Positioning Using Factor Graph Optimization. IEEE Trans. Intell. Transp. Syst. 2021, 22, 6175–6187. [Google Scholar] [CrossRef]
  9. Wang, D. Cooperative V2X Relative Navigation Using Tight-Integration of DGPS and V2X UWB Range and Simulated Bearing. Ph.D. Thesis, University of Calgary, Calgary, AB, Canada, 2015. [Google Scholar]
  10. Ko, S.-W.; Chae, H.; Han, K.; Lee, S.; Seo, D.-W.; Huang, K. V2X-Based Vehicular Positioning: Opportunities, Challenges, and Future Directions. IEEE Wirel. Commun. 2021, 28, 144–151. [Google Scholar] [CrossRef]
  11. Gáspár, P.; Szalay, Z.; Aradi, S. Highly Automated Vehicle Systems; BME MOGI: Budapest, Hungary, 2014. [Google Scholar]
  12. Nguyen, P.N.; Tran Truong, T.P.; Nguyen, T.T.L.; Ha, S.; Khanh, V.H.; Nguyen, T.A.; Le, K.B.; Nguyen, M.T. Acb: Authority Control Bluetooth Devices in Smarthome Environment System 2024. Available online: http://dx.doi.org/10.2139/ssrn.5064588 (accessed on 2 October 2025).
  13. Yang, G.; Shin, H.; Kim, H. Frequency Hopping as Diffusion: A New Perspective on Mitigating Bluetooth Packet Collisions in Dense Deployments. In Proceedings of the 2024 IEEE 21st International Conference on Mobile Ad-Hoc and Smart Systems (MASS), Seoul, Republic of Korea, 23–25 September 2024; pp. 19–28. [Google Scholar]
  14. Dittmer, C. Bluetooth Use Cases for V2X Automotive ApplicationsV2X. 2024. Available online: https://www.bluetooth.com/blog/bluetooth-use-cases-for-v2x-automotive-applications/ (accessed on 2 October 2025).
  15. Aza, A.; Melendi, D.; García, R.; Pañeda, X.G.; Pozueco, L.; Corcoba, V. Bluetooth 5 Performance Analysis for Inter-Vehicular Communications. Wirel. Netw. 2022, 28, 137–159. [Google Scholar] [CrossRef]
  16. Iordache, V.; Minea, M.; Gheorghiu, R.A.; Cormos, A.C.; Badau, F. Testing Bluetooth Low Energy v5 Capabilities for Connected Vehicles Applications. In Proceedings of the 2024 47th International Conference on Telecommunications and Signal Processing (TSP), Online, 1–12 July 2024; pp. 130–134. [Google Scholar]
  17. Perez-Yuste, A.; Pitarch-Blasco, J.; Falcon-Darias, F.A.; Nuñez, N. Design of New BLE GAP Roles for Vehicular Communications. Sensors 2024, 24, 4835. [Google Scholar] [CrossRef]
  18. Kumar, N.; Shukla, H.; Rajalakhsmi, P. V2X Enabled Emergency Vehicle Alert System. arXiv 2024, arXiv:2403.19402. [Google Scholar] [CrossRef]
  19. Vladyko, A.; Plotnikov, P.; Tambovtsev, G. Simulation-Based Evaluation of V2X System with Variable Computational Infrastructure. Network 2025, 5, 4. [Google Scholar] [CrossRef]
  20. Rabieh, K.; Samir, R.; Azer, M.A. Empowering Pedestrian Safety: Unveiling a Lightweight Scheme for Improved Vehicle-Pedestrian Safety. Information 2024, 15, 160. [Google Scholar] [CrossRef]
  21. Petrov, T.; Pocta, P.; Kovacikova, T. Benchmarking 4G and 5G-Based Cellular-V2X for Vehicle-to-Infrastructure Communication and Urban Scenarios in Cooperative Intelligent Transportation Systems. Appl. Sci. 2022, 12, 9677. [Google Scholar] [CrossRef]
  22. Shin, C.; Farag, E.; Ryu, H.; Zhou, M.; Kim, Y. Vehicle-to-Everything (V2X) Evolution From 4G to 5G in 3GPP: Focusing on Resource Allocation Aspects. IEEE Access 2023, 11, 18689–18703. [Google Scholar] [CrossRef]
  23. Raeisi Ziarani, M. Resource Allocation in V2I Link Between Connected Autonomous Vehicles and 5G Mm-Wave Band Small-Cells Using Machine Learning. Ph.D. Thesis, University of Calgary, Calgary, AB, Canada, 2024. [Google Scholar]
  24. Bansal, A.; Formosa, N. Navigating the Road to Connectivity: Use Cases and Design Considerations for V2X Networks Using Millimeter-Wave 5G Beam-Steering Antennas. In Proceedings of the 2023 IEEE Microwaves, Antennas, and Propagation Conference (MAPCON), Ahmedabad, India, 10–14 December 2023; pp. 1–6. [Google Scholar]
  25. Campolo, C.; Molinaro, A.; Iera, A.; Fontes, R.R.; Rothenberg, C.E. Towards 5G Network Slicing for the V2X Ecosystem. In Proceedings of the 2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft), Montreal, QC, Canada, 25–29 June 2018; pp. 400–405. [Google Scholar]
  26. Chatzoulis, D.; Chaikalis, C.; Xenakis, A.; Kosmanos, D.; Anagnostou, K.E. Channel Coding QoS Analysis in 5G V2X Dynamic Propagation Models. IEEE Access 2025, 13, 80149–80173. [Google Scholar] [CrossRef]
  27. Solomitckii, D.; Semkin, V.; Karttunen, A.; Petrov, V.; Nguyen, S.L.H.; Nikopour, H.; Haneda, K.; Andreev, S.; Talwar, S.; Koucheryavy, Y. Characterizing Radio Wave Propagation in Urban Street Canyon with Vehicular Blockage at 28 GHz. IEEE Trans. Veh. Technol. 2020, 69, 1227–1236. [Google Scholar] [CrossRef]
  28. Ma, Z.; Deng, Z.; Tian, Z.; Zhang, Y.; Wang, J.; Guo, J. A Line of Sight/Non-Line of Sight Recognition Method Based on the Dynamic Multi-Level Optimization of Comprehensive Features. Sensors 2025, 25, 304. [Google Scholar] [CrossRef]
  29. Mendes, B.; Araújo, M.; Goes, A.; Corujo, D.; Oliveira, A.S.R. Exploring V2X in 5G Networks: A Comprehensive Survey of Location-Based Services in Hybrid Scenarios. Veh. Commun. 2025, 52, 100878. [Google Scholar] [CrossRef]
  30. Kleinow, T.; Lakshmanan, S.; Richardson, P.; Elangovan, V.; Schmidt, S.; Locke, J.; Crowder, M. A Validated Model for Non-Line-of-Sight V2X Communications. In Proceedings of the 2020 Antenna Measurement Techniques Association Symposium (AMTA), Newport, RI, USA, 2–5 November 2020; pp. 1–6. [Google Scholar]
  31. Ye, L.; Li, J.; Zeng, L.; Han, Q.; Liu, E. 3D Channel Model Parameters Optimizing for C-V2X Communication Scenarios with K-Means++ and Genetic Algorithm. In Proceedings of the 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, 24–28 September 2023; pp. 42–47. [Google Scholar]
  32. Lei, J.; Chen, S.; Zeng, L.; Liu, F.; Zhu, K.; Liu, J. In-Chamber V2X Oriented Test Scheme for Connected Vehicles. In Proceedings of the 2019 IEEE Intelligent Vehicles Symposium (IV), Paris, France, 9–12 June 2019; pp. 1–6. [Google Scholar]
  33. Ye, L.; Liu, E.; Han, Q.; Sun, Q.; Zeng, L.; Liu, F. Scenarios Oriented V2V 3D Channel Model Optimization. In Proceedings of the 2021 IEEE International Intelligent Transportation Systems Conference (ITSC), Indianapolis, IN USA, 19–22 September 2021; pp. 3204–3209. [Google Scholar]
  34. Huang, Z.; Chen, S.; Pian, Y.; Sheng, Z.; Ahn, S.; Noyce, D.A. Toward C-V2X Enabled Connected Transportation System: RSU-Based Cooperative Localization Framework for Autonomous Vehicles. IEEE Trans. Intell. Transp. Syst. 2024, 25, 13417–13431. [Google Scholar] [CrossRef]
  35. Giuliano, R.; Vegni, A.M.; Loscrí, V.; Innocenti, E.; Vizzarri, A.; Mazzenga, F. MuSLi: A Multi Sensor LiDAR Detection for C-V2X Networks. Comput. Netw. 2023, 221, 109514. [Google Scholar] [CrossRef]
  36. Kiela, K.; Barzdenas, V.; Jurgo, M.; Macaitis, V.; Rafanavicius, J.; Vasjanov, A.; Kladovscikov, L.; Navickas, R. Review of V2X–IoT Standards and Frameworks for ITS Applications. Appl. Sci. 2020, 10, 4314. [Google Scholar] [CrossRef]
  37. Liu, Q.; Liu, R.; Wang, Z.; Han, L.; Thompson, J.S. A V2X-Integrated Positioning Methodology in Ultradense Networks. IEEE Internet Things J. 2021, 8, 17014–17028. [Google Scholar] [CrossRef]
  38. Ivan, I.; Besnier, P.; Bunlon, X.; Le Danvic, L.; Crussière, M.; Drissi, M. Influence of Propagation Channel Modeling on V2X Physical Layer Performance. In Proceedings of the Fourth European Conference on Antennas and Propagation, Barcelona, Spain, 12–16 April 2010; pp. 1–5. [Google Scholar]
  39. Möller, A.; Kürner, T. LTE Link Level Performance Evaluation Using Stochastic Channel Models for V2X Communication. In Proceedings of the 2015 IEEE Vehicular Networking Conference (VNC), Kyoto, Japan, 16–18 December 2015; pp. 25–31. [Google Scholar]
  40. Han, Q.; Yuan, X.; Zeng, L.; Zu, H.; Ye, L.; Lin, L. Scenario Oriented V2V Field Test Scheme with Dense Node Array. In Proceedings of the 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), Online, 20–23 September 2020; pp. 1–6. [Google Scholar]
  41. Seo, J.; Park, K.; Jeon, W.; Kwak, J.; Kim, D.K. Performance Evaluation of V2X Communications in Practical Small-Scale Fading Models. In Proceedings of the 2009 IEEE 20th International Symposium on Personal, Indoor and Mobile Radio Communications, Tokyo, Japan, 13–16 September 2009; pp. 2434–2438. [Google Scholar]
  42. Molisch, A.F.; Mecklenbräuker, C.F.; Zemen, T.; Prokes, A.; Hofer, M.; Pasic, F.; Hammoud, H. Millimeter-Wave V2X Channel Measurements in Urban Environments. IEEE Open J. Veh. Technol. 2025, 6, 520–541. [Google Scholar] [CrossRef]
  43. Ding, C.; Ho, I.W.-H. Digital-Twin-Enabled City-Model-Aware Deep Learning for Dynamic Channel Estimation in Urban Vehicular Environments. IEEE Trans. Green Commun. Netw. 2022, 6, 1604–1612. [Google Scholar] [CrossRef]
  44. Wang, S.; Huang, J.; Zhang, X. Demystifying Millimeter-Wave V2X: Towards Robust and Efficient Directional Connectivity under High Mobility. In Proceedings of the 26th Annual International Conference on Mobile Computing and Networking ACM, London, UK, 21–25 September 2020; pp. 1–14. [Google Scholar]
  45. Wang, Y.; Narasimha, M.; Heath, R.W. Towards Robustness: Machine Learning for MmWave V2X with Situational Awareness. In Proceedings of the 2018 52nd Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, USA, 28–31 October 2018; pp. 1577–1581. [Google Scholar]
  46. Zhang, D. Distributed Cache Enabled V2X Networks: Proposals, Research Trends and Challenging Issues. arXiv 2018, arXiv:1803.06059. [Google Scholar] [CrossRef]
  47. Hejazi, H.; Wippelhauser, A.; Bokor, L. Extending Artery/OMNeT++ for Evaluating V2X Sensor Fusion Algorithms with Infrastructure Support in Cooperative Advanced Driver Assistance Systems. IEEE Access 2025, 13, 128492–128517. [Google Scholar] [CrossRef]
  48. Hughes, E.; El-Tawab, S. Software Defined V2X Mesh Networking for Autonomous Vehicles in a Controlled Environment. In Proceedings of the 2023 IEEE Global Conference on Artificial Intelligence and Internet of Things (GCAIoT), Dubai, United Arab Emirates, 10–11 December 2023; pp. 204–210. [Google Scholar]
  49. Abidi, B.; Moreno, F.M.; Haziti, M.E.; Hussein, A.; Kaff, A.A.; Gomez, D.M. Hybrid V2X Communication Approach Using WiFi and 4G Connections. In Proceedings of the 2018 IEEE International Conference on Vehicular Electronics and Safety (ICVES), Madrid, Spain, 12–14 September 2018; pp. 1–5. [Google Scholar]
  50. Adegoke, E.I.; Zidane, J.; Kampert, E.; Ford, C.R.; Birrell, S.A.; Higgins, M.D. Infrastructure Wi-Fi for Connected Autonomous Vehicle Positioning: A Review of the State-of-the-Art. Veh. Commun. 2019, 20, 100185. [Google Scholar] [CrossRef]
  51. Ward, J.; Tracz, T. P3Mobility Leading in Economically Sustainable Safety with V2X Technology in Oakland County Michigan; Road Commission for Oakland County, P3Mobilit: Toronto, ON, Canada, 2025. [Google Scholar]
  52. de Almeida, T.T.; Ribeiro Júnior, J.G.; Campista, M.E.M.; Costa, L.H.M.K. Wi-Fi Direct Performance Evaluation for V2P Communications. J. Sens. Actuator Netw. 2020, 9, 28. [Google Scholar] [CrossRef]
  53. Naik, G.; Jerry Park, J.-M. Impact of Wi-Fi Transmissions on C-V2X Performance. In Proceedings of the 2019 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN), Newark, NJ, USA, 11 November 2019; pp. 1–10. [Google Scholar]
  54. Ivanov, V.; Grigoriev, V.; Mekhrafzuni, A. Wi-Fi 5.8 GHz Interference Density Impact on Intelligent Transportation System Link. In Proceedings of the 2020 IEEE International Conference on Electrical Engineering and Photonics (EExPolytech), St. Petersburg, Russia, 15–16 October 2020; pp. 18–21. [Google Scholar]
  55. Haque, K.F.; Abdelgawad, A.; Yelamarthi, K. Comprehensive Performance Analysis of Zigbee Communication: An Experimental Approach with XBee S2C Module. Sensors 2022, 22, 3245. [Google Scholar] [CrossRef] [PubMed]
  56. Iordache, V.; Minea, M.; Gheorghiu, R.A.; Badau, F.; Stoica, V.; Stan, V.A. Assessing Performance of Long-Range ZigBee for Road Infrastructure Communications. In Proceedings of the 2024 16th International Conference on Electronics, Computers and Artificial Intelligence (ECAI), Iasi, Romania, 27 June 2024; pp. 1–5. [Google Scholar]
  57. Sarango, A.; Vásquez, G.; Torres, R. A Vehicle-to-Vehicle Communication System Using ZigBee. IOP Conf. Ser. Earth Environ. Sci. 2024, 1370, 012008. [Google Scholar] [CrossRef]
  58. Li, Z.; Yu, S. Design and Research of a Vehicle Approaching Reminder Device Based on ZigBee Wireless Sensor Network. Sci. Program. 2022, 2022, 8797953. [Google Scholar] [CrossRef]
  59. Lei, Y.; Wang, T.; Wu, J. Vehicles Relative Positioning Based on ZigBee and GPS Technology. In Proceedings of the 2016 6th International Conference on Electronics Information and Emergency Communication (ICEIEC), Beijing, China, 17–19 June 2016; pp. 59–62. [Google Scholar]
  60. Dong, C.; Chen, X.; Dong, H.; Yang, K.; Guo, J.; Bai, Y. Research on Intelligent Vehicle Infrastructure Cooperative System Based on Zigbee. In Proceedings of the 2019 IEEE 5th International Conference on Transportation Information and Safety (ICTIS), Liverpool, UK, 14–17 July 2019; pp. 1337–1343. [Google Scholar]
  61. Gu, Y.; Hsu, L.-T.; Kamijo, S. GNSS/Onboard Inertial Sensor Integration with the Aid of 3-D Building Map for Lane-Level Vehicle Self-Localization in Urban Canyon. IEEE Trans. Veh. Technol. 2016, 65, 4274–4287. [Google Scholar] [CrossRef]
  62. Bansal, M.; Shrivastava, P.; Harshan, J. A Prototype on the Feasibility of Learning Spatial Provenance in XBee and LoRa Networks. In Proceedings of the 2024 16th International Conference on COMmunication Systems & NETworkS (COMSNETS), Bangalore, India, 3–7 January 2024; pp. 264–266. [Google Scholar]
  63. Javed, I.; Tang, X.; Saleem, M.A.; Javed, A.; Zia, M.A.; Shoukat, I.A. Localization for V2X Communication with Noisy Distance Measurement. Int. J. Intell. Netw. 2023, 4, 355–360. [Google Scholar] [CrossRef]
  64. Dudhe, R.; Matias, N.; Jowkar, A.M.; Dsouza, A.A. An Affordable Solution to V2V Communication Using XBee. Int. J. Innov. Technol. Explor. Eng. 2019, 8, 3931–3935. [Google Scholar] [CrossRef]
  65. Haque, K.F.; Abdelgawad, A.; Yanambaka, V.P.; Yelamarthi, K. A LoRa Based Reliable and Low Power Vehicle to Everything (V2X) Communication Architecture. In Proceedings of the 2020 IEEE International Symposium on Smart Electronic Systems (iSES) (Formerly iNiS), Chennai, India, 14–16 December 2020; pp. 177–182. [Google Scholar]
  66. Zadobrischi, E.; Havriliuc, Ș. Enhancing Scalability of C-V2X and DSRC Vehicular Communication Protocols with LoRa 2.4 GHz in the Scenario of Urban Traffic Systems. Electronics 2024, 13, 2845. [Google Scholar] [CrossRef]
  67. Haque, K.F.; Abdelgawad, A.; Yanambaka, V.P.; Yelamarthi, K. LoRa Architecture for V2X Communication: An Experimental Evaluation with Vehicles on the Move. Sensors 2020, 20, 6876. [Google Scholar] [CrossRef]
  68. Li, Y.; Han, S.; Yang, L.; Wang, F.-Y.; Zhang, H. LoRa on the Move: Performance Evaluation of LoRa in V2X Communications. In Proceedings of the 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu, China, 26–30 June 2018; pp. 1107–1111. [Google Scholar]
  69. Turan, B.; Narmanlioglu, O.; Koc, O.N.; Kar, E.; Coleri, S.; Uysal, M. Measurement Based Non-Line-of-Sight Vehicular Visible Light Communication Channel Characterization. IEEE Trans. Veh. Technol. 2022, 71, 10110–10114. [Google Scholar] [CrossRef]
  70. Godoy, J.A.; Cabrera, F.; Araña, V.; Sánchez, D.; Alonso, I.; Molina, N. A New Approach of V2X Communications for Long Range Applications in UAVs. In Proceedings of the 2018 2nd URSI Atlantic Radio Science Meeting (AT-RASC), Gran Canaria, Spain, 28 May–1 June 2018; pp. 1–4. [Google Scholar]
  71. Karne, R.K.; Sreeja, T.K. Review on Vanet Architecture and Applications. Turk. J. Comp. Math. Educ. 2021, 12, 1745–1749. [Google Scholar]
  72. Zhou, R. VANET Architecture Analysis and Protocols. Int. J. Comput. Appl. 2022, 184, 44–54. [Google Scholar] [CrossRef]
  73. Ibrar, M.; Adbeb, T.; Di, W. Software-Defined Networking (SDN) Based VANET Architecture: Mitigation of Traffic Congestion. Int. J. Adv. Comput. Sci. Appl. 2020, 11, 706–714. [Google Scholar] [CrossRef]
  74. Secinti, G.; Canberk, B.; Duong, T.Q.; Shu, L. Software Defined Architecture for VANET: A Testbed Implementation with Wireless Access Management. IEEE Commun. Mag. 2017, 55, 135–141. [Google Scholar] [CrossRef]
  75. Balta, M.; Özçeli̇k, İ. A 3-Stage Fuzzy-Decision Tree Model for Traffic Signal Optimization in Urban City via a SDN Based VANET Architecture. Future Gener. Comput. Syst. 2020, 104, 142–158. [Google Scholar] [CrossRef]
  76. Ghosh, J.; Kumar, N.; Al-Utaibi, K.A.; Sait, S.M.; Vo, V.N.; So-In, C. Reliable Data Transmission for a VANET-IoIT Architecture: A DNN Approach. Internet Things 2024, 25, 101129. [Google Scholar] [CrossRef]
  77. Luber, A.; Bauer, S.; Junghans, M.; Schulz, J. On Measuring Traffic with Wi-Fi and Bluetooth; Institute of Transportation Systems German Aerospace Center: Berlin, Germany, 2011. [Google Scholar]
  78. Abbott-Jard, M.; Shah, H.; Bhaskar, A. Empirical Evaluation of Bluetooth and Wifi Scanning for Road Transport. In Proceedings of the Australasian Transport Research Forum 2013, Brisbane, QLD, Australia, 2–4 October 2013. [Google Scholar]
  79. Chiva, I.-C.; Minea, M.; Minea, V.-L.; Semenescu, A. Anonymous Detection of Traveler Flows Employing Bluetooth Technologies. Proceedings 2020, 63, 61. [Google Scholar] [CrossRef]
  80. Gheorghiu, R.A.; Iordache, V.; Cormoș, A.C. Analysis of the Possibility to Detect Road Vehicles via Bluetooth Technology. Sensors 2021, 21, 7281. [Google Scholar] [CrossRef]
  81. Crawford, F.; Watling, D.P.; Connors, R.D. Identifying Road User Classes Based on Repeated Trip Behaviour Using Bluetooth Data. Transp. Res. Part A Policy Pract. 2018, 113, 55–74. [Google Scholar] [CrossRef]
  82. Abedi, N.; Bhaskar, A.; Chung, E. Bluetooth and Wi-Fi MAC Address Based Crowd Data Collection and Monitoring: Benefits, Challenges and Enhancement. In Proceedings of the Australasian Transport Research Forum 2013, Brisbane, QLD, Australia, 2–4 October 2013. [Google Scholar]
  83. Tsubota, T.; Yoshii, T. An Analysis of the Detection Probability of MAC Address from a Moving Bluetooth Device. Transp. Res. Procedia 2017, 21, 251–256. [Google Scholar] [CrossRef]
  84. Bjerre-Nielsen, A.; Minor, K.; Sapieżyński, P.; Lehmann, S.; Lassen, D.D. Inferring Transportation Mode from Smartphone Sensors: Evaluating the Potential of Wi-Fi and Bluetooth. PLoS ONE 2020, 15, e0234003. [Google Scholar] [CrossRef] [PubMed]
  85. Hainen, A.M.; Wasson, J.S.; Hubbard, S.M.L.; Remias, S.M.; Farnsworth, G.D.; Bullock, D.M. Estimating Route Choice and Travel Time Reliability with Field Observations of Bluetooth Probe Vehicles. Transp. Res. Rec. J. Transp. Res. Board 2011, 2256, 43–50. [Google Scholar] [CrossRef]
  86. Gudishala, R.; Wilmot, C.; Mokkapatti, A. Travel Time Estimation Using Bluetooth; Department of Civil and Environmental Engineering Louisiana State University: Baton Rouge, LA, USA, 2016. [Google Scholar]
  87. Bouaru, R.; Peculea, A.; Iancu, B.; Buzura, S.; Cebuc, E.; Dadarlat, V. Analysis of a Bluetooth Traffic Dataset Obtained during University Examination Sessions. Data 2024, 9, 49. [Google Scholar] [CrossRef]
  88. Bhaskar, A.; Chung, E. Fundamental Understanding on the Use of Bluetooth Scanner as a Complementary Transport Data. Transp. Res. Part C Emerg. Technol. 2013, 37, 42–72. [Google Scholar] [CrossRef]
  89. Hussain, E.; Behara, K.N.S.; Bhaskar, A.; Chung, E. A Framework to Quantify the Differences between Multi-Modal Travel Demand: A Case Study on Brisbane Network Using Bluetooth and Smartcard Data. IEEE Trans. Intell. Transp. Syst. 2019, 23, 8126–8135. [Google Scholar] [CrossRef]
  90. Bhaskar, A.; Kieu, L.M.; Qu, M.; Nantes, A.; Chung, E. On the Use of Bluetooth MAC Scanners for Live Reporting of the Transport Network. In Proceedings of the Eastern Asia Society for Transportation Studies, Taipei, Taiwan, 9–11 September 2013; Volume 9. [Google Scholar]
  91. Gunda, P.K. Network-Wide Traffic Congestion Visual Analytics: A Case Study for Brisbane Bluetooth MAC Scanner Data. Master’s Thesis, Queensland University of Technology, Brisbane, QLD, Australia, 2021. [Google Scholar]
  92. Michau, G.; Nantes, A.; Chung, E.; Abry, P.; Borgnat, P. Retrieving Trip Information from a Discrete Detectors Network: The Case of Brisbane Bluetooth Detectors. In Proceedings of the 32nd Conference of Australian Institutes of Transport Research, Sydney, NSW, Australia, 17–18 February 2014. [Google Scholar]
  93. Malinovskiy, Y.; Saunier, N.; Wang, Y. Analysis of Pedestrian Travel with Static Bluetooth Sensors. Transp. Res. Rec. J. Transp. Res. Board 2012, 2299, 137–149. [Google Scholar] [CrossRef]
  94. Araghi, B.N.; Hammershøj Olesen, J.; Krishnan, R.; Tørholm Christensen, L.; Lahrmann, H. Reliability of Bluetooth Technology for Travel Time Estimation. J. Intell. Transp. Syst. 2015, 19, 240–255. [Google Scholar] [CrossRef]
  95. Cebecauer, M.; Gundlegård, D.; Jenelius, E.; Burghout, W. Spatio-Temporal Public Transport Mode Share Estimation and Analysis Using Mobile Network and Smart Card Data. In Proceedings of the 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, 24–28 September 2023; pp. 2543–2548. [Google Scholar]
  96. Kompleksowe Badania Ruchu We Wrocławiu i Otoczeniu—KBR 2018—Kompleksowe Badania Ruchu—Biuletyn Informacji Publicznej Urzędu Miejskiego Wrocławia. Available online: https://bip.um.wroc.pl/artykul/565/37499/kompleksowe-badania-ruchu-we-wroclawiu-i-otoczeniu-kbr-2018 (accessed on 14 August 2025).
  97. Kruszyna, M. Investment Challenges Pertaining to the Achievement of the Goals of the Mobility Policy Based on the Analysis of the Results of Traffic Surveys in Wroclaw. Arch. Civ. Eng. 2021, 67, 505–523. [Google Scholar]
  98. Dong, H.; Wu, M.; Ding, X.; Chu, L.; Jia, L.; Qin, Y.; Zhou, X. Traffic Zone Division Based on Big Data from Mobile Phone Base Stations. Transp. Res. Part C Emerg. Technol. 2015, 58, 278–291. [Google Scholar] [CrossRef]
  99. Alexander, L.; Jiang, S.; Murga, M.; González, M.C. Origin–Destination Trips by Purpose and Time of Day Inferred from Mobile Phone Data. Transp. Res. Part C Emerg. Technol. 2015, 58, 240–250. [Google Scholar] [CrossRef]
  100. Toole, J.L.; Colak, S.; Sturt, B.; Alexander, L.P.; Evsukoff, A.; González, M.C. The Path Most Traveled: Travel Demand Estimation Using Big Data Resources. Transp. Res. Part C Emerg. Technol. 2015, 58, 162–177. [Google Scholar] [CrossRef]
  101. Bonnel, P.; Hombourger, E.; Olteanu-Raimond, A.-M.; Smoreda, Z. Passive Mobile Phone Dataset to Construct Origin-Destination Matrix: Potentials and Limitations. Transp. Res. Procedia 2015, 11, 381–398. [Google Scholar] [CrossRef]
  102. Larijani, A.N.; Olteanu-Raimond, A.-M.; Perret, J.; Brédif, M.; Ziemlicki, C. Investigating the Mobile Phone Data to Estimate the Origin Destination Flow and Analysis; Case Study: Paris Region. Transp. Res. Procedia 2015, 6, 64–78. [Google Scholar] [CrossRef]
  103. Okmi, M.; Por, L.Y.; Ang, T.F.; Ku, C.S. Mobile Phone Data: A Survey of Techniques, Features, and Applications. Sensors 2023, 23, 908. [Google Scholar] [CrossRef]
  104. Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation) (Text with EEA Relevance). Off. J. Eur. Union 2016, 119, L 119/1.
  105. Smartphone Ownership in the US by Age 2023. Available online: https://www.statista.com/statistics/489255/percentage-of-us-smartphone-owners-by-age-group/ (accessed on 14 August 2025).
  106. Carrel, A.; Lau, P.S.C.; Mishalani, R.G.; Sengupta, R.; Walker, J.L. Quantifying Transit Travel Experiences from the Users’ Perspective with High-Resolution Smartphone and Vehicle Location Data: Methodologies, Validation, and Example Analyses. Transp. Res. Part C Emerg. Technol. 2015, 58, 224–239. [Google Scholar] [CrossRef]
  107. Sinclair, M.; Maadi, S.; Zhao, Q.; Hong, J.; Ghermandi, A.; Bailey, N. Assessing the Socio-Demographic Representativeness of Mobile Phone Application Data. Appl. Geogr. 2023, 158, 102997. [Google Scholar] [CrossRef]
  108. Leurent, F.; Sun, D.; Xie, X. On Heterogenous Sampling Rates in Origin–Destination Matrix Estimation Based on Trajectory Data and Link Counts. Transp. Res. Rec. 2023, 2677, 1169–1180. [Google Scholar] [CrossRef]
  109. Patire, A.D.; Wright, M.; Prodhomme, B.; Bayen, A.M. How Much GPS Data Do We Need? Transp. Res. Part C Emerg. Technol. 2015, 58, 325–342. [Google Scholar] [CrossRef]
  110. Yu, J.; Stettler, M.E.J.; Angeloudis, P.; Hu, S.; Chen, X. (Michael) Urban Network-Wide Traffic Speed Estimation with Massive Ride-Sourcing GPS Traces. Transp. Res. Part C Emerg. Technol. 2020, 112, 136–152. [Google Scholar] [CrossRef]
  111. Wang, X.; Khattak, A.J.; Liu, J.; Masghati-Amoli, G.; Son, S. What Is the Level of Volatility in Instantaneous Driving Decisions? Transp. Res. Part C Emerg. Technol. 2015, 58, 413–427. [Google Scholar] [CrossRef]
  112. Forghani, M.; Karimipour, F.; Claramunt, C. From Cellular Positioning Data to Trajectories: Steps towards a More Accurate Mobility Exploration. Transp. Res. Part C Emerg. Technol. 2020, 117, 102666. [Google Scholar] [CrossRef]
  113. Li, C.; Zheng, L.; Jia, N. Network-Wide Ride-Sourcing Passenger Demand Origin-Destination Matrix Prediction with a Generative Adversarial Network. Transp. A Transp. Sci. 2024, 20, 2109774. [Google Scholar] [CrossRef]
  114. Putatunda, S.; Laha, A.K. Travel Time Prediction in Real Time for GPS Taxi Data Streams and Its Applications to Travel Safety. Hum.-Centric Intell. Syst. 2023, 3, 381–401. [Google Scholar] [CrossRef]
  115. Wang, Z.; Zhang, Y.; Jia, B.; Gao, Z. Comparative Analysis of Usage Patterns and Underlying Determinants for Ride-Hailing and Traditional Taxi Services: A Chicago Case Study. Transp. Res. Part A Policy Pract. 2024, 179, 103912. [Google Scholar] [CrossRef]
  116. Wang, F.; Ross, C.L. New Potential for Multimodal Connection: Exploring the Relationship between Taxi and Transit in New York City (NYC). Transportation 2019, 46, 1051–1072. [Google Scholar] [CrossRef]
  117. Riascos, A.P.; Mateos, J.L. Networks and Long-Range Mobility in Cities: A Study of More than One Billion Taxi Trips in New York City. Sci. Rep. 2020, 10, 4022. [Google Scholar] [CrossRef]
  118. Rahmani, M.; Jenelius, E.; Koutsopoulos, H.N. Non-Parametric Estimation of Route Travel Time Distributions from Low-Frequency Floating Car Data. Transp. Res. Part C Emerg. Technol. 2015, 58, 343–362. [Google Scholar] [CrossRef]
  119. Griffin, G.P.; Mulhall, M.; Simek, C.; Riggs, W.W. Mitigating Bias in Big Data for Transportation. J. Big Data Anal. Transp. 2020, 2, 49–59. [Google Scholar] [CrossRef]
  120. Oliveira, J.; Farias, C.M.; Pacitti, E.; Fortino, G. (Eds.) Big Social Data and Urban Computing: First Workshop, BiDU 2018, Rio de Janeiro, Brazil, 31 August, 2018, Revised Selected Papers, Proceedings of the BiDU 2018, Rio de Janeiro, Brazil, 31 August 2018; Communications in Computer and Information Science; Springer International Publishing: Cham, Switzerland, 2019; Volume 926, ISBN 978-3-030-11237-0. [Google Scholar]
  121. Sun, W.; Vij, A.; Kaliszewski, N. A Flexible and Scalable Single-Level Framework for OD Matrix Inference Using IoT Data. Transp. Res. Part A Policy Pract. 2023, 175, 103775. [Google Scholar] [CrossRef]
  122. Zawodny, M.; Kruszyna, M. Proposals for Using the Advanced Tools of Communication between Autonomous Vehicles and Infrastructure in Selected Cases. Energies 2022, 15, 6579. [Google Scholar] [CrossRef]
  123. Huo, J.; Yang, H.; Li, C.; Zheng, R.; Yang, L.; Wen, Y. Influence of the Built Environment on E-Scooter Sharing Ridership: A Tale of Five Cities. J. Transp. Geogr. 2021, 93, 103084. [Google Scholar] [CrossRef]
  124. Younes, H.; Zou, Z.; Wu, J.; Baiocchi, G. Comparing the Temporal Determinants of Dockless Scooter-Share and Station-Based Bike-Share in Washington, D.C. Transp. Res. Part A Policy Pract. 2020, 134, 308–320. [Google Scholar] [CrossRef]
  125. Ma, X.; Yuan, Y.; Van Oort, N.; Hoogendoorn, S. Bike-Sharing Systems’ Impact on Modal Shift: A Case Study in Delft, the Netherlands. J. Clean. Prod. 2020, 259, 120846. [Google Scholar] [CrossRef]
  126. Loaiza-Monsalve, D.; Riascos, A.P. Human Mobility in Bike-Sharing Systems: Structure of Local and Non-Local Dynamics. PLoS ONE 2019, 14, e0213106. [Google Scholar] [CrossRef]
  127. Aliari, S.; Nasri, A.; Motalleb Nejad, M.; Haghani, A. Toward Sustainable Travel: An Analysis of Campus Bikeshare Use. Transp. Res. Interdiscip. Perspect. 2020, 6, 100162. [Google Scholar] [CrossRef]
  128. Sun, Z.; Li, Y.; Zuo, Y. Optimizing the Location of Virtual Stations in Free-Floating Bike-Sharing Systems with the User Demand during Morning and Evening Rush Hours. J. Adv. Transp. 2019, 2019, 4308509. [Google Scholar] [CrossRef]
  129. Mayne, K.; Fried, L.; Ducellier, C.; Shared Ambition: The Potential for Bike Sharing in Europe: Benchmarking 148 Cities. Cycling Industries Europe. 2024. Available online: https://cyclingindustries.com/fileadmin/CIE_Bike_Sharing_2024_Report.pdf (accessed on 2 October 2025).
  130. Radzimski, A.; Dzięcielski, M. Exploring the Relationship between Bike-Sharing and Public Transport in Poznań, Poland. Transp. Res. Part A Policy Pract. 2021, 145, 189–202. [Google Scholar] [CrossRef]
  131. OpenStreetMap. Available online: https://www.openstreetmap.org/ (accessed on 19 August 2025).
Figure 1. Publications by year chart.
Figure 1. Publications by year chart.
Applsci 15 10686 g001
Figure 2. Publications by type chart.
Figure 2. Publications by type chart.
Applsci 15 10686 g002
Figure 3. TAZ grid in classical cellular data big data analysis; map data from Open Street Map [131].
Figure 3. TAZ grid in classical cellular data big data analysis; map data from Open Street Map [131].
Applsci 15 10686 g003
Figure 4. Classical TAZ in Wrocław’s Comprehensive Traffic Research [96]; map data from Open Street Map [131].
Figure 4. Classical TAZ in Wrocław’s Comprehensive Traffic Research [96]; map data from Open Street Map [131].
Applsci 15 10686 g004
Figure 5. TAZ in case study’s street with proposed RSU locations; map data from Open Street Map [131].
Figure 5. TAZ in case study’s street with proposed RSU locations; map data from Open Street Map [131].
Applsci 15 10686 g005
Figure 6. Location of Bluetooth/Wi-Fi scanners in case study’s street; map data from Open Street Map [131].
Figure 6. Location of Bluetooth/Wi-Fi scanners in case study’s street; map data from Open Street Map [131].
Applsci 15 10686 g006
Table 1. Summary of Section 3 findings about each technology.
Table 1. Summary of Section 3 findings about each technology.
TechnologyStrengthsWeaknessesRangeData RateLatencyCostScalabilityUrban Canyon SuitabilityMaturityResearch Trends/Gaps
BluetoothLow energy, adaptive frequency hopping, good short-range performanceLimited range, low throughput at distance~300 m static, ~1 km dynamicUp to ~992 kbps~100 msLowMediumModerate—short range avoids interferenceMatureIntegration with smartphones, hybrid V2X systems
Wi-FiHigh bandwidth, widely availablePoor NLOS performance, interference in urban areas~40–100 m in NLOSHighVariable, often >100 msMediumMediumLimited—signal attenuation, interferenceMatureBeacon stuffing, GNSS support, hybrid systems
ZigBeeLow power, multi-hop support, good short-range accuracyLow bandwidth, interference on 2.4 GHz~300–600 mLowModerateLowMediumModerate—suitable for short-range V2XMatureHybrid with LoRa, GNSS-free localization
LoRaLong range, low power, good NLOS performanceLow data rate, speed limitations~160–500 m (urban), >10 km (UAV)Very lowHighVery LowHighGood—especially for non-safety appsEmergingSF tuning for urban NLOS, UAV applications
LTE (4G)Reliable, widely deployedPerformance drops in dense urban canyons~800 m+Medium~10–50 msMediumHighModerate—LOS helps, but reflections degradeMatureFusion with INS, V2I/V2V hybrid systems
5G/NR-V2XHigh speed, low latency, AI optimization, mmWave supportNeeds dense infrastructure, blockage sensitivity~200 m (mmWave), variesUp to 25 Mbps uplink~5–10 msHighHighGood—with extra power and beamformingEmergingDeep learning for NLOS detection, private 5G for SAE L3+
C-V2XHigh reliability, GNSS-independent localization, LiDAR fusionSignal loss in deep urban canyons~150 m (RSU-based)High<10 msHighHighStrong—RSU-based systems perform wellEmergingRSU density optimization, hybrid with DSRC
DSRC/IEEE 802.11pLow latency, proven V2V/V2I useBlockage and reflection issues in urban canyons~300 mMedium<100 msMediumMediumModerate—needs RSU density increaseMaturemmWave integration, beam-forming, digital twin modeling
Table 2. Summary of Section 4 findings about each technology.
Table 2. Summary of Section 4 findings about each technology.
Technology/Data SourceWhat It MeasuresSpatial ResolutionTemporal ResolutionAdvantagesDisadvantagesTypical Applications
Bluetooth/Wi-Fi ScannersMAC addresses, travel paths, speed, transport modeMedium (scanner coverage area)High (real-time or near real-time)Low-cost, passive data collection, useful in urban canyonsLower detection rate (~62%), privacy concerns, device visibility requiredOD matrix estimation, transport mode detection, congestion analysis
Mobile Phone Data (Billing and Signal DBs)Location changes, trip purpose (home/work/other)Low to Medium (depends on BTS density)Low to Medium (3–6 h)Wide coverage, passive dataBiased by age and phone usage, ping-pong effect, privacy regulationsOD matrix, urban planning, transport mode detection
GPS Data (From Phones, Taxis, AVs)Precise location, speed, trip pathsHigh (point-level)High (real-time)High accuracy, suitable for micro-level modelingSignal loss in urban canyons, privacy, costSpeed estimation, driving behavior, OD matrix, AV modeling
Smartcards (Public Transport)Boarding/alighting data, trip frequencyHigh (station-level)~High (per transaction, check-in, check-out)Accurate for PT users, integrates with other dataLimited to PT users, no full trip pathPublic transport planning, multimodal analysis
ANPR (Automatic Number Plate Recognition)Vehicle counts, travel times, license platesHigh (camera coverage)High (real-time)Accurate vehicle tracking, useful for bias correctionInfrastructure cost, privacy concernsTraffic flow analysis, bias reduction
AVL (Automatic Vehicle Location)Vehicle GPS locationHigh (vehicle-level)High (real-time)Accurate fleet trackingLimited to equipped vehiclesFleet management, public transport monitoring
Loop DetectorsVehicle counts, speedPoint-based (road section)High (real-time)Proven technology, reliableFixed location, no OD infoTraffic volume, speed monitoring
Micromobility Data (E-scooters, Bikes)Trip paths, usage patternsHigh (GPS-level)High (real-time)Supports modal shift analysis, urban canyon-friendlyVendor-dependent, data access issuesFirst/last-mile analysis, urban mobility planning
Census and Survey DataTrip purpose, socio-demographicsLow (zone-level)Very Low (annual or less)Rich contextual dataOutdated quickly, expensiveModel calibration, bias correction
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.

Share and Cite

MDPI and ACS Style

Zawodny, M.; Kruszyna, M. Data Sources for Traffic Analysis in Urban Canyons—The Comprehensive Literature Review. Appl. Sci. 2025, 15, 10686. https://doi.org/10.3390/app151910686

AMA Style

Zawodny M, Kruszyna M. Data Sources for Traffic Analysis in Urban Canyons—The Comprehensive Literature Review. Applied Sciences. 2025; 15(19):10686. https://doi.org/10.3390/app151910686

Chicago/Turabian Style

Zawodny, Michał, and Maciej Kruszyna. 2025. "Data Sources for Traffic Analysis in Urban Canyons—The Comprehensive Literature Review" Applied Sciences 15, no. 19: 10686. https://doi.org/10.3390/app151910686

APA Style

Zawodny, M., & Kruszyna, M. (2025). Data Sources for Traffic Analysis in Urban Canyons—The Comprehensive Literature Review. Applied Sciences, 15(19), 10686. https://doi.org/10.3390/app151910686

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop