A Bibliometric Analysis and Visualization of In-Vehicle Communication Protocols

Abstract

This research examined the domain of intelligent transportation systems (ITS) by analyzing the impact of scholarly work and thematic prevalence, as well as focusing attention on vehicles, their technologies, cybersecurity, and related scholarly technologies. This was performed by examining the scientific literature indexed in the Scopus database. This study analysed 2919 documents published between 2018 and 2025. The findings indicated that the highest and most significant journal was derived from IEEE Transactions on Vehicular Technology, with significant standing to the growth of communication and computing on vehicles with edge computing and AI optimization of vehicular systems. In addition, important PST research conferences highlighted the growing interest in academic research in cybersecurity for vehicle networks. Sensor networks, pose forensics, and privacy-preserving communication frameworks were some of the significant contributing fields marking the significance of the interdisciplinary nature of this research. Employing bibliometric analysis, the literature illustrated the multiple channels integrating knowledge creation and innovation in ITS through citation analysis. The outcome suggested an increasingly sophisticated research area, weighing technical progress and increasing concern about security and privacy measures. Further studies must investigate edge computing integrated with AI, advanced privacy-preserving linguistic protocols, and new vehicular network intrusion detection systems.

1. Introduction

The automotive industry’s growth has been heavily influenced by incorporating electronic components and communication interfaces, which are fundamental in creating contemporary intelligent transportation systems [1]. Basic safety functions, as well as advanced driver assistance systems (ADAS), can be performed, and in-vehicle communication protocols allow for the interfacing of numerous electronic control units (ECUs) within vehicles [2,3]. These functionalities have become essential with the progressive steps towards automation, increased electrification, and improved connectivity of these vehicles [4].

Communication protocols significantly impact these systems’ performance, reliability, and efficiency [5]. Of the numerous in-vehicle protocols developed for communication, the most widely adopted include Controller Area Network (CAN), Local Interconnect Network (LIN), FlexRay, Media Oriented Systems Transport (MOST), and most recently, Automotive Ethernet [6]. Each protocol has its technical and functional strengths. Robustness is a prominent attribute of CAN, making it commonplace in powertrain and chassis systems. LIN provides cost-effective solutions to low-speed networks, aiding window and mirror controls [7]. FlexRay offers high-speed and deterministic communication for safety-critical applications, and MOST informs systems that require high-capacity bandwidth. In context, the use of Automotive Ethernet marks a paradigm shift toward a standard capable of sustaining the massive data transfer requirements posed by autonomous driving technologies [8].

The growing intricacy and the incorporation of vehicular systems sharpened professional focus in academia and industry toward intensive research into these protocols [9]. Bibliometric analysis stands out for its ability to chart the temporal progression of scholarly interest, significant publications, key authors, and emerging research focus areas. Such reviews are beneficial for developing insights into knowledge silos, areas where a given technology is overly focused, or insight gaps for future research.

This bibliometric review synthesizes the research literature on in-vehicle communication protocols into a coherent structure. This study uses citation and publication analytics combined with thematic grouping to chart the scientific discourse on CAN, LIN, FlexRay, MOST, and Automotive Ethernet. The impact of in-vehicle communication protocols was evaluated using bibliometric analysis, which identified remarkable works, developing trends, and gaps in the research. Bibliometric analysis was needed to provide an organized, evidentiary summary of the domain’s scientific advancement and collaboration activities.

This study aims to identify the most substantial publications, authors, and journals in terms of in-vehicle communication protocols for co-authorship, institutional, and keyword networks to be visualized and to showcase the development of emerging themes and trends in technology, such as AI, security, and protocol innovation. These goals contribute to the analysis process and this study’s contribution to scholarship. To enhance clarity and align with scholarly review standards, the objectives of this study have been explicitly stated in a structured format. The objectives of this study are as follows:

• To identify the most influential publications, authors, and journals in the domain of in-vehicle communication protocols, including CAN, LIN, FlexRay, MOST, and Automotive Ethernet.
• To map and analyze co-authorship, institutional collaborations, and country-level contributions within the field.
• To uncover key thematic areas, keyword trends, and emerging research directions using co-occurrence and co-citation analysis.
• To explore the integration of modern technologies such as AI and cybersecurity into vehicular networks based on publication trends and citation metrics.

Despite the growing number of reviews and studies on intelligent transportation systems (ITS), few have employed a comprehensive bibliometric approach that synthesizes multiple dimensions—such as co-citation analysis, institutional collaboration networks, and co-occurrence keyword mapping—over a robust and updated dataset spanning 2018 to early 2025. This study distinguishes itself by analyzing 2919 scholarly works and identifying not only dominant trends but also emerging under-represented themes, including Time-Sensitive Networking (TSN), CAN FD, and the evolving role of AI-driven intrusion detection systems (IDS). Additionally, unlike earlier reviews that focus on a single aspect or timeframe, our approach integrates diverse bibliometric perspectives and applies VOSviewer-based visualization to reveal latent structures within the scholarly landscape.

2. Materials and Methods

2.1. Search Strategy and Scope

A systematic bibliometric search was based on the Scopus database for gathering scholarly articles on in-vehicle communication protocols, such as CAN, LIN, FlexRay, MOST, and Automotive Ethernet. To expand the scope of the research, the search using the search word construction with the help of keywords, plus other terms such as “vehicular communication”, “automotive networking”, and “in-vehicle systems” was utilized. The pool was restricted to academic papers, conference publications, and peer-reviewed articles in English. Using this strategy, the research obtained authoritative and relevant academic work discussing the evolution of vehicular network technologies.

The search string used was “Controller Area Network” OR “CAN Bus” OR “Local Interconnect Network” OR “LIN Bus” OR “FlexRay” OR “Media Oriented Systems Transport” OR “MOST protocol” OR “Automotive Ethernet” OR “Time-Sensitive Networking” OR “TSN” OR “CAN FD” OR “in-vehicle network” OR “vehicular communication protocol”.

2.2. Inclusion and Exclusion Criteria

The final sample included 2919 publications from January 2018 until March 2025. For this study, we decided to concentrate on this interval to include studies that reflect modern trends and the emerging role of digitization in automotive networks. Documents were eliminated because of a lack of cornerstone peer review, duplication, or focus on communication technologies within vehicles. Setting unambiguous inclusion and exclusion rules provided greater transparency in research and eliminated problems associated with the range and scope of data explored in the previous versions of this study. The detailed explanation of these variables increases the reliability and repeatability of the bibliometric analysis.

Considering the wide coverage and sound indexing of the Scopus database, a systematic bibliometric investigation has been conducted to make the analysis open and repeatable. Citation metrics utilized the complete counting approach, giving full credit to each co-author and reference in co-authorship and citation measurements. Networks based on co-authorship, keyword co-occurrence, and co-citation were produced using VOSviewer, applying the threshold of at least 10 citations or occurrences, eliminating smaller data, and improving interpretability. This mode, a norm in bibliometric studies, utilizes descriptive methods. However, complete documentation of the search and data handling steps was also conducted to ensure transparency and replicability of findings.

2.3. Sources of Information

Using the Scopus database maintained by Elsevier, this study follows the methodological approaches provided in [10]. Scopus, which was launched in 2004, is now one of the largest bibliographic databases in the world, with more than 77.8 million entries compiled since 1969. Its coverage comprises more than 23,000 peer-reviewed journals, 294 trade journals, over 852 book series, and proceedings from more than 120,000 international academic conferences. This makes Scopus an authoritative source for accessing the global research landscape. As [11] indicates, the database is comprehensive on a global scale and comes equipped with sophisticated bibliometric research tools. These tools allow for advanced filtering and segmentation by the year of publication, document type, author names, institutional affiliation, and citation counts, enabling more precise analysis of scholarly trends.

In combination with these functions, this study takes a systematic approach to finding and analyzing the literature about in-vehicle communication protocols. Scopus was chosen because of its credibility, data trustworthiness, and capability to conduct detailed scientometric analyses. The database provided relevant publications for analysis, such as CAN and LIN, FlexRay, MOST, and Automotive Ethernet, which aided in exploring existing research gaps, publication patterns, and significant contributions in this research domain.

2.4. Study Design

In analyzing the scholarly literature on communication technologies for vehicles, this study responds to the increasing academic interest in the architecture, functionality, and evolution of in-vehicle communication protocols, including CAN, LIN, FlexRay, MOST, and Automotive Ethernet. These contributions span comparative evaluations, safety and security assessments, and integration into intelligent transportation systems. Building upon prior bibliometric approaches, our research aims to map both the volume and direction of scholarly work in these domains, especially in underexplored areas.

To build our dataset, we used a combination of broad and targeted keyword searches (e.g., “Controller Area Network”, “CAN Bus”, “Automotive Ethernet”, etc.) that ensured coverage of both foundational technologies and recent innovations. The final dataset captures 2919 peer-reviewed publications and represents a broad, interdisciplinary spectrum of research.

For bibliometric mapping, we employed VOSviewer (version 1.6.20), which supports advanced network visualizations of bibliographic data. The following methodological configurations were applied:

• Clustering layout: LinLog modularity-based layout to enhance the visibility of thematic clusters.
• Minimum threshold for inclusion: At least 10 occurrences for keywords and 10 citations for co-cited references to focus on the most significant patterns while reducing noise from sparsely cited items.
• Counting method:

  ○

  Full counting was used for co-authorship networks to give equal weight to all contributors.

  ○

  Fractional counting was applied to co-citation analysis to normalize influence across highly connected items.

Visual outputs—such as co-authorship, keyword co-occurrence, and journal co-citation networks—were created using VOSviewer (version 1.6.20) and Microsoft Excel (version Excel 2024). These tools allowed for complementary visual and statistical analysis of relationships among scholars, institutions, and thematic areas.

Additionally, only documents classified as peer-reviewed journal articles, conference proceedings, review articles, or book chapters were included. Editorials, non-peer-reviewed reports, abstracts, and trade magazine articles were excluded to ensure analytical rigor.

3. Results

This bibliometric review was conducted on a dataset comprising 2919 scholarly documents on in-vehicle communication protocols, including CAN, FlexRay, LIN, MOST, and Automotive Ethernet. Almost half of the documents in the dataset were conference papers, suggesting an actively developing field, with 1465 entries (50.19%) for the conference papers. The journal articles group represented a close second with 1324 documents (45.36%), suggesting that academics are considerably engaged with peer-review processes endorsing research validation. Book chapters, although minimal, enhanced the dataset with 86 entries (2.95%), demonstrating their contribution towards comprehensive discourse or niche analysis of specific topics. Review articles were the least frequent, at a total of 44 (1.51%), indicating that while the field is rich in empirical and technical work, there is scant synthesis or consolidation of knowledge available. The observed document distribution indicates a strong emphasis on research and innovation in the technical area of focus.

Table 1. Type of document retrieved.

Type of Document	Frequency	Proportion
Article	1324	45.36%
Book chapter	86	2.95%
Conference paper	1465	50.19%
Review	44	1.51%
Grand Total	2919	100.00%

shows the types of documents retrieved.

3.1. Current Status of In-Vehicle Communication Protocol

3.1.1. Annual Trends in Publication

The “Recent Development of Communication Protocols in Vehicles” graph showcases publication tendencies concerned with in-vehicle communication protocols from 2018 during the first quarter of 2025. Increasing scientific output from 2018 to 2023 depicts/surging interest in research and development activities focusing on the subject area. Publication numbers started slightly below 300 in 2018 and increased steadily to just above 650 in 2023. This phenomenon, also referred to as a surge, can be attributed to the swift improvements in the field of automotive technologies, which include autonomous systems, vehicular ad hoc network (VANET) systems, Automotive Ethernet systems, and IoT as a constituent of smart transport systems [12]. Nonetheless, a steep decline in numbers from 2024, especially in the early months of 2025, is highly observable. This sharp drop from the norm is most likely not the result of an actual cutback in research activity but signifies an unfinished dataset for 2025, which is still pending. These indicators are commonplace in bibliometric studies where data capture for the current year’s data remains incomplete. From a high-level perspective, this trend captures the sustained and intense increase in activity within the research domain for the last couple of years, further underlining the increasing importance of the research attention of the field. Figure 1, which shows annual trends in publication, was generated using MS Excel.

3.1.2. Distribution of Organisations

This study of institutional contributions to in-vehicle communication protocols indicates that a significant concentration comes from Chinese universities. Figure 2 shows both the absolute and relative publication outputs for the top institutions. For example, Xidian University has 85 publications, accounting for approximately 2.91% of the total dataset. The Beijing University of Posts and Telecommunications follows closely with 84 (2.88%), and Tongji University with 82 (2.81%). These institutions are well-known for their strong research profiles in vehicular networks, intelligent transportation, and embedded communication systems. With 77 publications, Hunan University is joined by Beihang University with 74 publications, demonstrating considerable output. The two focus on research on autonomous vehicles and more advanced communication protocols such as CAN, FlexRay, and Automotive Ethernet. This dual representation of publication volume and proportion allows for a fairer comparison across institutions of varying sizes and outputs, highlighting both volume and relative research emphasis.

Intelligent transportation systems and modern vehicles require real-time responses and support for increased data throughput. All these features are catered for by CAN FD (Flexible Data-rate) [13]. This laterally improves the traditional CAN protocol by increasing payload size and data communication rate. Despite its growing significance, how CAN FD is cited in the literature and its bibliometric clusters suggest it is lacking in scrutinized scholarly research, especially regarding secure high-speed vehicular networks.

In addition to that, TSN remains relatively unexplored in co-citation networks or thematic clusters, where bibliometric mapping shows most of its attention should be directed towards [14]. This is especially true considering TSN’s potential to bridge the gap between old systems and modern automotive architectures. Because of this, Time-Sensitive Networking (TSN)—an IEEE standard designed for data streaming over Ethernet with deterministic latency, in contrast to TSN’s focus on delivering data in a controlled fashion—has yet to explore its complete potential in autonomous and connected vehicles [15]. In addition to ADAS, V2X and infotainment systems also require highly accurate timing, making TSN vital for these applications.

Further illustrating China’s extensive contribution in this field are Beijing Jiaotong University (65) and Shanghai Jiao Tong University (61), which have specialized work in V2X (Vehicle-to-Everything) communications and cyber–physical systems for vehicles. Outside of China, the most remarkable achievement is from the South Korean college Soonchunhyang University, which has 51 publications, showing the growing interest and research in smart transportation in the region. Coming in last is Tsinghua University, the most noted research center in the world, with 50 publications. The concentration of research in East Asia precisely illustrates the region’s dominance in intelligent mobility innovations and the growing global importance of research on in-vehicle communication protocols [16].

3.1.3. Distribution of Published Journals

The bibliometric study reveals a strong dominance of IEEE-affiliated sources in the in-vehicle communication protocols field, detailing the major journals and publications of significant value. Figure 3 presents the top journals in terms of absolute publication counts and their relative proportions. The IEEE Transactions on Vehicular Technology leads with 118 publications (4.04% of the total), followed by IEEE Access with 108 (3.70%) and IEEE Transactions on Intelligent Transportation Systems with 99 (3.39%).

Including percentage values enhances the analysis by contextualizing journal influence relative to the overall scholarly output in the field. This method ensures that smaller yet high-impact journals are not overlooked in volume-based comparisons.

As for the rest of the publications, the IEEE Transactions on Intelligent Transportation Systems (99 publications) stands out with profound studies on advanced traffic systems, V2V/V2X communication, and autonomous vehicle communication frameworks. The IEEE Vehicular Technology Conference (67 papers) remains one of the top showcases for new research-directed developments in the ecosystem of vehicular technology. Very useful in connecting the ‘ride’ industry and academic knowledge regarding automotive engineering standards are the SAE Technical Papers (59 publications). Significant Sensors (50) and Electronics (Switzerland) (44) publications that focus on communication hardware, sensor fusion, and performance of hardware and software in real-time systems. IEEE publications such as the Internet of Things Journal and the ACM International Conference Proceedings Series showcase the intersection of IoT and vehicle communication systems.

3.1.4. Distribution of Published Conference Papers

The distribution of published conference papers and journal articles demonstrates how impactful research continues to be concentrated in a limited number of publication outlets, particularly in the connected vehicles cybersecurity domain. Figure 4, generated using MS Excel, shows the distribution of published conference papers. Most strikingly, IEEE Transactions on Vehicular Technology is cited three times among the top-cited papers, with 556, 494, and 336 citation figures. This underscores its reputation as a leading outlet for research and innovation in vehicular networks, communication systems, and AI optimization engineering. Also noteworthy are the Privacy, Security, and Trust (PST) conferences held in 2018, showing an increased focus on cybersecurity for vehicular communication systems. This keen interest is equally shared in journals such as Sensors, IEEE Communications Surveys and Tutorials, and IEEE Transactions on Information Forensics and Security, which stand out for integrating sensor systems, secure computing, and literature surveys. The two recently established journals, Vehicular Communications and IEEE Transactions on Intelligent Transportation Systems, highlight the growing importance of specialized transportation and communication disciplines in the field. This distribution showcases reliance on specialized and multidisciplinary sources to disseminate research on intelligent transportation systems and vehicular cybersecurity, as conducted in [17].

3.1.5. Most Cited Countries

When analyzing the most cited nations in research on in-vehicle communication systems, distinct industry influence, and academic leadership disparities emerge. MS Excel was used to generate Figure 5, which shows the most cited countries. China leads the world as the highest contributor with 8746 publications; however, in average citations per publication, China’s impact factor (12.8) is moderate. China still lags significantly behind other leading countries. The United States follows China with 2845 published papers, demonstrating a significantly higher impact with average citations at 20.9 per publication. This indicates the higher quality and global relevance of US research. In this regard, South Korea, with 2542 publications and an average of 17.4 citations, demonstrates substantial academic and industrial investment in automotive communication technologies. While Australia published only 1193 papers, it had the highest citation impact, averaging 54.2 citations per publication. This showcases the exceptional quality and international visibility of their research. Malaysia also demonstrates a notable impact at 31.7 citations, surpassing larger contributors such as India (5.5) and Japan (9.2). Active members with balanced research quantity and quality include the United Kingdom (14.2), Italy (16.2), and Canada (23). These figures demonstrate the global balance of impactful research on vehicle communication systems.

3.2. Keyword Analysis of Research Hotspots

The co-occurrence column in the keyword dataset shows the extent to which terms appear together in a body of research, demonstrating thematic grouping and research hotspots involving in-vehicle communication and mobile communication systems. Figure 6, generated using VOSviewer, shows co-occurrence keywords. The keyword “5G mobile communication systems” has the most outstanding co-occurrence value (607), confirming its dominance and centrality in recent studies. “Accidents” (489) and “accident prevention” (307) also follow closely behind, reinforcing the growing focus of research on vehicular safety and crash mitigation.

Moreover, “Current” has a high co-occurrence (189), which may indicate foundational or baseline studies about vehicle technologies. The high value for “Access control” (178) emphasizes increasing attention to security and authentication issues about vehicular networks. Likewise, “adaptive control systems” (194) and “6G” (126) indicate the presence of next-generation intelligent network protocol tendencies.

The emerging themes, such as “accident detections” (61) and “adaptive boosting” (78), reflect the growing interest in incorporating machine learning and predictive analytics into vehicular environments. The lower co-occurrence term of “4G mobile communication systems” (63) suggests a shift in focus towards newer 5G and 6G technologies. These trends suggest that safety, connectivity, and advanced communication protocols are the primary focus areas for ongoing research, as mentioned in [18].

One emerging theme in bibliometric mapping is the increasing application of artificial intelligence (AI) and machine learning (ML) to in-vehicle communication systems. Core phrases ‘vehicular networks’ and ‘automotive cybersecurity’ are regularly accompanied by deep learning, reinforcement learning, and AI-based intrusion detection. Such development depicts a dependency on models of AI for advanced predictive algorithms, maintenance, traffic, and real-time threat monitoring within in-vehicle networks. The use of publications in [18,19], focusing on resource optimization using deep reinforcement learning and mobility-aware edge computing, highlights the influence that AI is having on the system design and efficiency of modern vehicular systems.

3.3. Co-Authorship Analysis

Conducting a research project is meticulous, often requiring collaboration among specialists from different fields to achieve reliable and precise results [20]. Following on from [21], studying the co-authorship networks can be considered a very narrow yet efficient approach for measuring the creativity of a particular research area, given that the analysis of co-authors’ countries and affiliated institutions is relatively simple.

3.3.1. Country Co-Authorship Network

The bibliometric analysis of publications by countries highlights the differences in research activity and its impact on productivity. Figure 7, generated using VOSviewer, shows the country’s co-authorship network. Nevertheless, its mean citations per publication (12.39) and a normalized citation rate (score) of 1.25 indicate that China does produce a great deal of work; however, the impact on individual papers is not very high relative to other countries. On the other hand, Australia and Canada tend to strike an equilibrium between volume and impact. Australia quenches its publication thirst with 70 publications but receives a remarkable 2153 citations, averaging over 30 citations per article and a normalized score of 2.22, the highest in the dataset. Similarly, Canada records 97 publications with 2082 citations, bolstering the average to 2.24, a high citation rate, emphasizing their sustained research impact. Numerous European and other developing nations meaningfully impact the research scope as well. Germany and France have solid outputs of 157 and 77 documents, though their average citation impacts of 8.21 and 10.24 still lag behind the global leaders. At the same time, however, Brazil, Egypt, and Denmark boast a substantial citation impact for their modest publication numbers, which imply focused, high-caliber work. Overall, countries that manage a balance of high output and high average citation scores tend to have the most substantial impact on the evolution of the discipline [22].

3.3.2. Highly Cited Publications

The major publications in connected vehicle technologies and cybersecurity highlight dominant communications systems, artificial intelligence, and intrusion detection technologies. Table 2 shows highly cited publications. One area of focus is optimizing systems through AI, examined by Liu et al. (2019) [19] and Hu (2018) [23] in deep reinforcement learning to concern edge computing, offloading, and resource allocation, which seeks to meet the increasing data processing needs in real-time vehicular networks. These papers attest to the academic discourse while advancing the development of connected vehicle ecosystems.

Another important cybersecurity topic is in-vehicle intrusion detection systems. Scholars have shown great concern regarding the protection of vehicular networks as cyber threats are on the rise. Seo et al. (2018) [24] developed novel IDS methods based on GANs and remote frame analysis in this context. Moreover, Choi et al. (2018) [25] proposed VoltageIDS, which focuses on communications at the lower levels. In addition, Wu et al.’s (2019) and El-Rewini et al.’s (2020) work on survey ID in cybersecurity, with no explicit focus, serve as other reference works for primary researchers by bringing together knowledge of different ID approaches and challenges in cybersecurity [2]. Documenting these gaps, which stem from the existing literature, shines a light on focusing on system performance in parallel to robust cybersecurity, marking the fact that there is a gap that the publications jointly depict.

Regarding the most cited countries in the field, China, Korea, and the United States, in particular, focus on applying artificial intelligence (AI) and machine learning (ML) to vehicles. This is clear in highly cited papers from these countries dealing with deep reinforcement learning for node–edge computation on vehicles, intrusion detection systems (IDS), and mobility-aware caching. For instance, substantial research performed by authors from China and the United States showcases advanced ML-based dynamic resource allocation and cybersecurity on vehicles. These countries have been strategically directing their research efforts towards AI-augmented vehicle innovation. It is no longer news that countries are rushing towards developing smart vehicles and adaptive vehicular networks, and these countries have marked their position as frontrunners in transportation research.

Table 2. Highly cited publications.

Article Name	Journal/Conference Name	Author	Year	Number of Citations
IEEE 802.11ad-Based Radar: An Approach to Joint Vehicular Communication-Radar System	IEEE Transactions on Vehicular Technology	[26]	2018	556
Deep Reinforcement Learning for Offloading and Resource Allocation in Vehicle Edge Computing and Networks	IEEE Transactions on Vehicular Technology	[19]	2019	494
Sensor technologies for intelligent transportation systems	Sensors (Switzerland)	[27]	2018	485
GIDS: GAN-based Intrusion Detection System for In-Vehicle Network	2018 16th Annual Conference on Privacy, Security and Trust, PST 2018	[24]	2018	396
Mobility-aware edge caching and computing in-vehicle networks: A deep reinforcement learning	IEEE Transactions on Vehicular Technology	[23]	2018	336
Challenges and Solutions for Cellular-Based V2X Communications	IEEE Communications Surveys and Tutorials	[28]	2020	312
VoltageIDS: Low-level communication characteristics for an automotive intrusion detection system	IEEE Transactions on Information Forensics and Security	[25]	2018	271
A survey of intrusion detection for in-vehicle networks	IEEE Transactions on Intelligent Transportation Systems	[29]	2019	260
Cybersecurity challenges in vehicular communications	Vehicular Communications	[9]	2020	240
PA-CRT: Chinese Remainder Theorem-Based Conditional Privacy-Preserving Authentication Scheme in Vehicular Ad-Hoc Networks	IEEE Transactions on Dependable and Secure Computing	[30]	2019	240

Table 2. Highly cited publications.

Article Name	Journal/Conference Name	Author	Year	Number of Citations
IEEE 802.11ad-Based Radar: An Approach to Joint Vehicular Communication-Radar System	IEEE Transactions on Vehicular Technology	[26]	2018	556
Deep Reinforcement Learning for Offloading and Resource Allocation in Vehicle Edge Computing and Networks	IEEE Transactions on Vehicular Technology	[19]	2019	494
Sensor technologies for intelligent transportation systems	Sensors (Switzerland)	[27]	2018	485
GIDS: GAN-based Intrusion Detection System for In-Vehicle Network	2018 16th Annual Conference on Privacy, Security and Trust, PST 2018	[24]	2018	396
Mobility-aware edge caching and computing in-vehicle networks: A deep reinforcement learning	IEEE Transactions on Vehicular Technology	[23]	2018	336
Challenges and Solutions for Cellular-Based V2X Communications	IEEE Communications Surveys and Tutorials	[28]	2020	312
VoltageIDS: Low-level communication characteristics for an automotive intrusion detection system	IEEE Transactions on Information Forensics and Security	[25]	2018	271
A survey of intrusion detection for in-vehicle networks	IEEE Transactions on Intelligent Transportation Systems	[29]	2019	260
Cybersecurity challenges in vehicular communications	Vehicular Communications	[9]	2020	240
PA-CRT: Chinese Remainder Theorem-Based Conditional Privacy-Preserving Authentication Scheme in Vehicular Ad-Hoc Networks	IEEE Transactions on Dependable and Secure Computing	[30]	2019	240

3.4. Co-Citation Analysis

3.4.1. Reference Co-Citations Analysis

The co-citation evaluation provides a detailed picture of authoritative figures often cited in other works, indicating their impact on the vehicle security research ecosystem [31]. Figure 8, generated using VOSviewer, shows the analysis of reference co-citations. It is also the case that Alazab M. dominates this dataset with a staggering 940 co-citations and 9640 total citations, significantly outpacing others’ contributions, including Emad A.’s, who follows at 4211 and Al-Kadri M.O. with 4054, suggesting their pivotal contributions to such works indeed lie in security protocols within vehicular networks. Such numbers indicate that his research is one of the primary engines driving innovation in their industry. Abboud K. is also notable, as are Ahmed E. and Ahmed M., who each accumulate these impressive figures of over 600 co-citations, hinting at broad domains of influence. Ai B. stands out with an unparalleled total citation count of 6101 paired with 613 co-citations, underscoring the dire importance of some of his single-authored publications. These authors often serve as a cornerstone of models, frameworks, or experimental works that are customarily cited. These researchers’ strong ties in co-citation imply that they construct a focal point or center of gravity in the in-vehicle networks and cybersecurity paradigm, shaping ideas and practical interventions for connected car technologies.

3.4.2. Journal Co-Citation Cited Analysis

The analysis of co-citation data in Figure 9, generated using VOSviewer, indicates that Ad Hoc Networks is one of the key trademark feeds in the domain, given its citations ad hoc netw. and ad hoc networks. These two sources have high total link strengths (3422 and 2159) and citations (92 and 78, respectively), which indicates their primary role in cornerstone research-level works in the area of connected vehicles, wireless communication, and vehicular ad hoc networks (VANETs). The need for such co-citation also signifies that, despite the differences in writing order of names, the network community extensively cites this journal for reportage on specialized network systems and defenses about automotive control systems. This merging proves the dominance of dependent constituents in these systems in technologies associated with connected vehicles. Further, “Accident Analysis & Prevention”, even Nancy’s version, strengthens this argument with 1939 link strength and 129 citations, indicating the growing need to merge safety evaluation and filtering relevance from cyberspace contexts. These sources emphasize two aspects of the literature: the thorough preparatory work involving construction and proactive defense strategy steps in intelligent transportation systems.

Table 3. Top core journals.

Journal/Conference	Subject	Number of Citations
IEEE Transactions on Vehicular Technology	Vehicular Technology	4175
IEEE Access	General Engineering/Technology	2664
IEEE Transactions on Intelligent Transportation Systems	Intelligent Transportation	2628
Sensors	Sensor Technology	1132
Vehicular Communications	Vehicular Communications	809
IEEE Internet of Things Journal	Internet of Things (IoT)	679
IEEE Communications Surveys and Tutorials	Communications/Surveys	474

shows the top core journals.

4. Discussion

The co-citation analysis reveals the scholarly context of the field, particularly emphasizing the area of cybersecurity related to vehicles, networks, and wireless communications. M. Alazab, A. H. Abdullah, I. Ahmad, and S. H. Ahmed emerge as leading contributors, exuding strong influence as critical hubs in the citation matrix by receiving numerous citations combined with high link values. To illustrate, M. Alazab dominates the domain with 9640 citations and 940 co-citation links, which showcase his crowned influence in the scholarly network, facing an ever-increasing need for novel insights. These authors focus on intrusion detection systems IDS, VANETs, and intelligent transportation security, which are border challenges in the smart mobility era.

The co-citation network is representative of active partnerships among researchers dealing with the technical and applied dimensions of connected vehicles, from machine learning-based threat detection to cryptographic techniques for CAN (Controller Area Network) [31]. Further analysis shows clusters centered around data privacy, in-vehicle communication security, and real-time vehicular network analytics domains. Notably, newer contributors such as B. Ai and E. Aliwa also show strong co-citation links, suggesting authoritative influence within the area. The blend of well-established experts alongside relatively new contributors indicates that not only is the field mature but also that it is highly versatile, fostering innovative behavior and interdisciplinary exploration.

This study of co-cited journals and conferences shows that the impact clusters of research are distributed among security and vehicle technology publications. Journals such as Ad Hoc Networks and their derivatives appear repeatedly and vigorously. For example, the journal “Ad Hoc Networks” alone has over 120 citations and over 3400 link strengths in its most cited version. This citation count marks it as the central journal for wireless vehicular communication system-related publications. Likewise, ACM Computing Surveys and Accident Analysis & Prevention serve as a theoretical base and practical safety concern regarding connected vehicles. The latter journal strengthens the trend towards risk mitigation with a normalized citation of 129 and 1939 total link strength, placing cybersecurity not only as a technical issue but as a public safety danger [32].

They act as linkages across automotive engineering, network science, and computer security and serve as interdisciplinary collaboration centers. This varies across publication channels and fosters effective diffusion of knowledge to augment scholarly participation in vehicle interconnectivity’s practical and theoretical aspects.

The analysis conducted at the country level synthesizes data at different scales. It reveals that the research on connected vehicle systems and their cybersecurity has a high level of globalization and, at the same time, is concentrated in the more economically developed areas. China provides adequate documents (n = 8746). Well-known publications such as “A Practical Wireless Attack on the Connected Car” elucidate the former by analyzing various weaknesses in the vehicle’s in-vehicle network (IVN) and suggesting pragmatic security solutions. These studies highlight the ever-increasing challenge of creating impenetrable systems, especially with the advanced availability of fully automated and semi-automated vehicles. At the same time, the literature published in Accident Analysis & Prevention is increasingly linking cybersecurity with road safety outcomes, symbolizing the intersection of transportation engineering and cybersecurity [2]. Another prominent and significant development is the use of AI in-vehicle systems, not only for driving but also for real-time threat monitoring and detection, as well as for self-adaptive communication systems. M. Alazab and S. H. Ahmed have been increasingly focused on AI-based systems that monitor and neutralize cyber threats in real time, demonstrating an emerging trend of proactive rather than reactive security. The growing rate of co-citation of publications concentrating on data governance, ethics, and privacy affects user trust. It signals a shift toward an emerging mature field concerned with technological viability and social consequences. This data curation procedure suggests that future studies will focus more on in-depth tiered frameworks that include a combination of technical accuracy, dynamic functionality, compliance with standards, system usability design, and human-centered design. Overall, the findings indicate that there is a field with coordinated and clear direction, which is complex, evolving, and still developing opportunities, balanced by established research, new insights, and pressing real-world needs.

One of the core contributions of this study is its broad and multifaceted bibliometric scope. Unlike prior reviews—such as Muslam [18], which centers on V2V security protocols, or Qasim et al. [33], which focuses narrowly on handover techniques—our study provides an integrated analysis of technical, thematic, and geographical trends across a larger corpus of the literature. Notably, this work brings attention to relatively underexplored subdomains such as TSN and CAN FD, where bibliometric clustering reveals gaps in security-focused research despite their technological significance. These insights provide actionable direction for scholars and practitioners aiming to address next-generation vehicular communication challenges. To further clarify how this work differentiates itself from the recent literature reviews in the field,

Table 4. Comparison of this study with recent bibliometric and review papers on in-vehicle communication protocols and intelligent transportation systems. Our work demonstrates broader coverage, integration of multiple bibliometric methods, and focus on emerging research directions.

Study	Focus	Time Range	Methodology	Dataset Size	Novelty Compared to Our Work
Muslam (2024) [18]	Security in V2V Communication	~2010–2023	Qualitative Review	~150 papers	Narrow focus, no co-citation or network analysis
Qasim et al. (2024) [33]	Handover Management in Vehicular Comms	~2015–2023	Survey	~100 papers	Specialized focus, no bibliometric scope
This Study	Comprehensive review of in-vehicle protocols	2018–2025	Bibliometric: co-citation, co-authorship, keyword clustering	2919 papers	Broad scope, visual network analysis, emerging trend detection

compares the scope, dataset, and methodological approaches of our study with other representative works.

This study advances the existing body of research by presenting the first comprehensive bibliometric analysis that integrates co-authorship, keyword co-occurrence, and co-citation networks across the five dominant in-vehicle communication protocols (CAN, LIN, FlexRay, MOST, and Automotive Ethernet), covering the period from 2018 to early 2025. Unlike prior works that focused narrowly on single protocols or limited metrics, our approach reveals emerging trends, such as the integration of Time-Sensitive Networking (TSN) and CAN FD within security-sensitive automotive architectures. Moreover, Table 4 compares this study to recent reviews, emphasizing our broader scope, methodological richness, and identification of underexplored but impactful subdomains. These insights position the current work as a distinctive contribution to the literature on vehicular communications and cybersecurity.

5. Conclusions

This paper offers a detailed bibliometric and co-citation analysis of the existing literature on connected vehicles, vehicular networks, and the intersections of cybersecurity. The results point to a deep and developing research field, noting the foundational work of important figures such as M. Alazab and S. H. Ahmed. The author and source clustering showcase particular focus areas such as disruption detection, in-vehicle communication protocols, vehicular machine learning, and telematics security. Critical research is conducted in these fields, and Ad Hoc Networks, ACM Computing Surveys, and Accident Analysis & Prevention serve as platforms for new theories that are empirically grounded and valuable to practice. Such journals note the impact of their publications. The country analysis illustrates the prominence of China, the US, and Australia, which indicates the global interest and collaboration in the security issues of connected mobility. The dynamic and interdisciplinary nature of the discipline is further evident through the caliber of some of its other conferences, including the IEEE Symposium on Security and Privacy and the IEEE Intelligent Vehicles Symposium.

The findings show a notable change in scholarly attention from the more traditional aspects of concern, such as security protocols and architectures, to more comprehensive AI-driven and user-centric methods. This change aligns with the industrial evolution of connected vehicle technologies and the more sophisticated security challenges [24].

In contrast with recent narrative reviews, this study not only confirms dominant trends but also visualizes and exposes emerging, under-represented themes in the ITS research ecosystem.

In general, this study illuminates the particular aspects of the discourse within the field, which are likely of interest to researchers, policymakers, and technology developers. Mapping described theories, contributors, and knowledge clusters highlight strengths while laying the groundwork for future research. Connected vehicles are expected to be integrated into the innovative city framework; therefore, concentrated academic attention will be needed to ensure that the vehicles are secure, efficient, and ethically used within the transport system.

Limitations and Future Research Directions

Although this study makes a substantial contribution to understanding trends in research on in-vehicle communication, several limitations should be acknowledged. First, the analysis relies exclusively on the Scopus database. While Scopus offers extensive global coverage and consistent indexing, it does not include content from other prominent repositories such as Web of Science, IEEE Xplore, or Google Scholar. This limitation may affect the completeness of author profiles, citation metrics, and journal representation—particularly in engineering-centric or regional domains.

Second, the bibliometric techniques applied here—such as co-citation analysis, publication volume counts, and link strength measures—tend to favor highly cited and established scholars. As a result, the analysis may under-represent contributions from early-career researchers, emerging authors, and under-resourced institutions. This inherent bias in citation-based metrics could obscure novel or interdisciplinary work that has not yet accrued significant citation counts.

To address these limitations, future studies could:

• Incorporate alternative impact metrics, such as altmetrics, emerging author indices, and social media or preprint citation signals, which offer complementary perspectives on scholarly influence.
• Apply temporal citation normalization or field-weighted metrics to better highlight recent and impactful publications across disciplines.
• Use content-based approaches—such as full-text analysis, keyword embedding, and topic modeling—to uncover latent themes and early-stage research clusters beyond citation frequency.

Additionally, broadening the data collection procedure to integrate multiple databases would yield a more holistic representation of global research activity. Combining scientometric methods with content analysis or natural language processing (NLP) techniques could enhance the depth and nuance of thematic interpretation. Exploring white papers, industry reports, and patent data may also provide valuable insight into how academic findings translate into real-world applications.

Finally, future research could benefit from examining the policy landscape shaping connected vehicle technologies—for instance, how national cybersecurity regulations and smart mobility frameworks influence scholarly output and funding. Together, these enhancements would further elevate the richness and inclusiveness of bibliometric inquiry in this fast-evolving domain.