Our study provides a comprehensive synthesis of previous research on the optimal deployment of geodetic networks. Over the past decade, the number of publications has fluctuated considerably (Figure 3a). The lowest output was recorded in 2016 (~35 publications), followed by a steady increase between 2017 and 2020. In 2024, a significant rise occurred again, reaching 65 publications. This trend clearly reflects the growing interest in the topic.

In terms of geographical distribution, the highest number of studies was published in China (102 papers), followed by the United States (73) and Germany (45) (Figure 3b). Regarding the host journals, the largest share of contributions appeared in the Journal of Geodesy and the Journal of Surveying Engineering (~30 publications each) (Figure 3c). In addition, leading high-impact journals such as Remote Sensing, Survey Review, GPS Solutions and Geophysical Journal International also ranked among the most active outlets. This indicates that research on CORS network optimisation is primarily disseminated within the domains of geodesy and remote sensing.

Among funding sources, the National Natural Science Foundation of China (NSFC) is the clear leading contributor, supporting more than 70 publications (Figure 3d). The gap between the first and second most active funding bodies is substantial, with a difference of nearly 50 papers, highlighting NSFC’s dominant role in advancing research on CORS network optimisation.

Regarding authorship, prominent contributors include Klein, Matsuoka and Rofatto, each credited with 6-14 publications (Figure 3e). In terms of institutional productivity, Wuhan University (China) is the most active institution, with over 20 articles, followed by leading centres such as CNRS - Centre National de la Recherche Scientifique (France), GFZ – Helmholtz Centre for Geosciences (Germany), and the Chinese Academy of Sciences (China) (Figure 3f).

Disciplinary distribution indicates that the majority of research is concentrated in Earth and Planetary Sciences (39%), followed by Engineering (24%), Environmental Sciences (8%), Computer Science (7%), and Physics and Astronomy (6%), while other areas collectively account for 16% (Figure 3j). This distribution underscores the multidisciplinary nature of CORS optimisation research, demonstrating strong links not only with geodesy, but also with computer science, engineering and environmental studies.

The bibliometric evidence confirms that the field is strongly supported by national grants and international research programmes. Leading institutions and researchers are increasingly applying Voronoi polygons, geodetic buffering, multi-GNSS integration and artificial intelligence approaches to improving network efficiency. The results suggest that the field is evolving in a multidisciplinary direction, with future research expected to focus on 3D coverage modelling, IoT integration and smart city infrastructure applications.

From a critical perspective, the bibliometric results show that the field is advancing rapidly but remains unevenly developed. The dominance of a limited number of countries and institutions indicates that current optimisation approaches are strongly shaped by region-specific technological capacity, data availability and funding structures. This limits the direct transferability of these methods to data-constrained regions such as Central Asia. In addition, although advanced optimisation algorithms and multi-GNSS approaches are increasingly represented in the literature, many studies remain simulation-based and provide limited validation using actual observational GNSS data. Therefore, a key research gap lies in connecting theoretical network optimisation with practical, region-specific performance evaluation. This gap justifies the use of simplified GIS-based spatial assessment as an initial planning tool, while also highlighting the need for future studies based on empirical GNSS observations, network geometry metrics and accuracy validation.

Overall, the bibliometric results indicate a growing and increasingly multidisciplinary research field, with strong contributions from leading countries, institutions and journals. These general patterns provide a foundation for a more detailed thematic analysis of methodological approaches, which is presented in the following section (3.2).

The bibliometric analysis reveals a clear and sustained increase in scientific research activity related to the optimisation of CORS networks. Based on the reviewed literature, a marked growth in publications has been observed, particularly after 2015, reflecting the rising global demand for high-precision geodetic infrastructures. This trend is closely associated with the rapid advances in GNSS technologies, the increasing reliance on real-time positioning systems, and the broader development of the digital economy.

The analysis further indicates that studies can be broadly categorised into several major methodological groups. The first includes studies grounded in classical geodetic network design, primarily based on least squares adjustment techniques. The second group encompasses research employing optimisation algorithms, in which methods such as genetic algorithms, particle swarm optimization (PSO), and simulated annealing are widely applied. The third consists of GIS-based and spatial analysis approaches, which are mainly used for visualisation purposes and preliminary network assessment.

Despite the methodological diversity, an important trend emerges from the bibliometric findings: while a significant portion of the literature relies on advanced mathematical and optimisation-based models, their practical implementation remains limited, particularly in developing countries. This limitation is often attributed to the lack of high-quality observational data, insufficient technical infrastructure, and the challenges associated with applying complex models under real-world conditions.

In addition, GIS-based approaches appear to be relatively underrepresented and are typically employed as simplified tools for initial evaluation. This indicates the presence of a gap between theoretically rigorous optimisation models and their practical, region-specific application. Such a discrepancy represents a critical challenge to the effective planning and deployment of CORS networks.

This study specifically addresses this research gap by integrating insights derived from bibliometric analysis with a simplified GIS-based spatial evaluation framework. The proposed approach does not aim to replace rigorous geodetic modelling, but rather to provide a practical and scalable conceptual tool that supports preliminary planning and decision-making processes, particularly in data-constrained environments.

The Uzbekistan case study is presented to demonstrate the application of simplified spatial analysis methods in a data-constrained national context. The planning and expansion of the CORS network in Uzbekistan are influenced by several local factors, including pronounced relief variability, extensive desert and semi-desert areas, uneven telecommunication and power infrastructure, and cross-border coordination requirements with neighbouring countries. Seasonal and cyclic ionospheric variations may also affect long-baseline RTK and VRS solutions. However, these factors are discussed here as contextual constraints only; the analysis evaluates spatial coverage and does not quantify positioning accuracy or GNSS network performance.

In the republic, very few scientific studies have been conducted in this field, which can be attributed to the scarcity of open data. Among them, Ruziev and et al. studied GNSS (global navigation satellite system) technology in the context of constructing a geodetic network in the capital of Uzbekistan and showed that no systematic work had previously been conducted on its reconstruction and development (Ruziev et al., 2024). In addition, this system involves appropriate organisation of the geodetic satellite network, as well as the development of a unified innovative system for state cadastres and real estate registration (Madimarova et al., 2024). The present GNSS network and its data constitute the main subject of Fozilova’s research, which emphasises that efforts are underway to establish a modern network in Uzbekistan’s geodesy sector. The primary-level networks and data transmission stations were nearly completed (Fazilova, 2017). Furthermore, as a result of research conducted by relevant state institutions, a web geoportal has been developed (Ndirangu et al., 2020).

Available open materials and practical reports often assess coverage based on Euclidean distance circle buffers, while coverage maps at the regional level, inter-station baseline length statistics, or PDOP scenarios are rarely presented (Prochniewicz et al., 2020). In addition, the placement problem is frequently addressed using simple statistical analysis approaches based on expert judgment. As a result, investment decisions face difficulties in providing a precise, quantitative answer to the question: “In which area will the installation of a certain number of stations yield the greatest benefit in terms of coverage and service quality?” Moreover, in recent years, ArcGIS has been successfully applied in geovisual monitoring and decision-making, owing to its capability to integrate real-time data and interactive maps (E.S.R.I., 2023).

The objective of our study is to evaluate the coverage of the existing CORS network in Uzbekistan using a geodetic approach (based on the WGS84 ellipsoid) through data collection, pre-processing, spatial analysis, visualisation, and the creation of a management geoportal. The study further aims to analyse the spatial distribution and service coverage of CORS stations; identify overlapping areas and uncovered zones based on 40 km buffer zones; and develop an interactive web map to present the results.

In conducting the research, the GIS (ArcGIS Pro 3.3.0) environment was utilised. Initially, the state boundary of the Republic of Uzbekistan, the provincial boundaries, and the boundary of the Republic of Karakalpakstan were included. A database of 132 CORS stations was compiled, of which 50 represented older (previously installed) stations and 82 were newly established ones. Separate layers (feature classes) were created for old and new CORS stations (Figure 4).

To distinguish between old stations (shown in blue) and new stations (red), thematic symbology was applied (Figure 4). Buffer zones were displayed in transparent colours to represent service coverage. National boundaries were added as a layer, enabling the analysis of CORS coverage in relation to both the country’s borders and administrative regions.

To evaluate the service coverage of the CORS network, circular buffer zones with radii of 40 km, 50 km and 60 km were generated around each station (Figure 5). These distances represent commonly used baseline lengths in GNSS network design. Shorter baselines are generally associated with more stable positioning performance due to the higher spatial correlation of atmospheric errors, whereas larger distances increase coverage, but may lead to reduced positioning reliability (Wanninger, 2002).

The buffer analysis was performed in the ArcGIS Pro environment (E.S.R.I., 2022), in which coverage areas were delineated for each scenario and the total coverage area was calculated (Table 1). In addition, the degree of overlap between buffer zones was analysed to identify redundant coverage and uncovered regions. These uncovered areas were considered priority zones for the installation of new stations.

At the regional level (Table 1), the analysis shows that areas with a high density of stations, such as Tashkent city, the Fergana Valley (Andijan, Fergana, Namangan), and the Syrdarya corridor, as well as large parts of the Samarkand, Jizzakh and Surkhandarya regions, remain well covered across all scenarios. For example, in the Jizzakh region, only 1.18 × 10³ km² remains uncovered out of a total area of 19.8 × 10³ km².

In contrast, the largest uncovered areas are in the Republic of Karakalpakstan (≈112.2 × 10³ km²), the Navoi region (≈54.3 × 10³ km²) and the Bukhara region (≈11.1 × 10³ km²). Meanwhile, coverage gaps in Kashkadarya (≈6.9 km²), Fergana (≈15.4 km²), Khorezm (≈67 km²) and Surkhandarya are minimal. These results indicate that the CORS network is relatively dense in populated and economically active regions, while coverage remains sparse in desert and semidesert zones such as the Ustyurt Plateau and the Kyzylkum Desert.

The comparative analysis demonstrates a clear relationship between buffer distance and coverage extent (Table 2). Under the 40 km scenario, the total coverage area is approximately 218,133.9 km², corresponding to 48.70% of the country’s territory. Increasing the radius to 50 km expands the coverage to 251,746.7 km² (56.21%), while the 60 km scenario reaches 277,077.7 km² (61.86%). These values should be interpreted strictly as spatial coverage indicators. They do not represent positioning accuracy, network reliability or GNSS performance, as no observational GNSS validation or accuracy assessment was conducted for the study.

The spatial variability of coverage is further illustrated in Figure 6, which presents the coverage area and coverage efficiency for each region based on the 40 km buffer scenario. The results show that regions such as Navoi, Karakalpakstan and Kashkadarya have large coverage areas but relatively lower efficiency, reflecting sparse station distribution over extensive territories. In contrast, regions with compact territories and dense station networks, such as Tashkent city, Andijan and Sirdarya, exhibit high coverage efficiency despite their smaller absolute coverage areas. This pattern highlights the uneven spatial performance of the network and emphasises the importance of considering both coverage extent and efficiency in network evaluation.

Furthermore, the analysis of station characteristics shows that older stations are mainly concentrated in the south-eastern regions of the country and have been in operation for many years. In contrast, newly installed stations were deployed more evenly across the territory during the modernisation process, thereby expanding overall coverage. Statistical analysis indicates that the average elevation of older stations is 604.5 m, whereas new stations are located at an average elevation of 885.5 m. The number and spatial distribution of new stations have contributed to a more balanced national network configuration.

Overall, the results (Table 2) highlight a trade-off between coverage and baseline distance. Larger buffer radii increase spatial coverage; however, they do not directly correspond to improved GNSS positioning accuracy, which also depends on additional factors such as atmospheric conditions, signal quality and network geometry. In this context, the range of 40–50 km may be considered a balanced compromise between coverage and potential positioning performance, although this conclusion should be interpreted with caution.

It is also important to note that most of the uncovered areas correspond to desert regions. Currently, these lands are gradually being developed and integrated into the national economy. Therefore, in the near future, the establishment of additional CORS stations in these regions will become necessary.

Finally, it should be emphasised that the analysis is based on a simplified geometric buffer approach and does not account for key GNSS error sources such as ionospheric and tropospheric effects, multipath, signal quality or network geometry (e.g., PDOP). Therefore, the results should be interpreted as a preliminary spatial assessment rather than a comprehensive geodetic optimisation.

In recent years, the demand for high-precision data in geodesy, remote sensing and monitoring technologies has increased significantly. This growing need has stimulated the rapid development of GNSS, CORS and InSAR technologies. Research conducted across various domains ((Benoit et al., 2015); (Bevly et al., 2016); (Yu et al., 2018)) has broadly highlighted the potential applications of these technologies and demonstrated their importance at the intersections of multiple scientific disciplines.

Building on the considerations outlined in the introduction, the study results support the view that inter-station spacing remains a critical factor in CORS network performance. The spatial analysis indicates that moderate station spacing provides adequate territorial coverage in many regions of Uzbekistan; however, noticeable gaps persist, particularly in the western parts of the country. This suggests that, while commonly recommended spacing ranges can serve as a useful reference, their practical effectiveness is strongly influenced by regional conditions such as terrain, atmospheric variability and infrastructure availability. Therefore, optimisation strategies should be adapted to local characteristics rather than relying on uniform spacing assumptions. Recent studies in Turkey and Malaysia and global analyses emphasise the importance of considering atmospheric effects, network geometry and technological advancements when determining the most effective CORS configuration ((Feng & Li, 2008); (Gökdaş & Özlüdemir, 2022); (Pipitone et al., 2023); (Rahmani et al., 2023); (Yu et al., 2020)).

Accordingly, our thematic analysis examined 53 selected articles, with closer scrutiny of the five most highly cited. This citation-weighted focus is deliberate: the most cited works define the methodological baseline against which regional adaptations are tested, making them the appropriate reference points for assessing transferability to Uzbekistan's specific topographic, atmospheric, and infrastructural conditions.

These works converge on a single conclusion: high-precision geodetic monitoring is not constrained by any one limitation but by a trilemma of accuracy, infrastructure, and cost, where optimising one dimension typically presses on the others. The generic atmospheric correction model of (Yu et al., 2018) defines the accuracy frontier, i.e., adapting tropospheric correction to terrain and atmospheric conditions yields strong results for deformation monitoring in seismically active regions, but precisely because it is tuned to local atmospheric behaviour, it resists direct transfer and would demand region-specific recalibration across Uzbekistan's arid lowlands and steep mountain gradients. (Bevly et al., 2016) shift the binding constraint from algorithm to infrastructure: their real-time GNSS/CORS work for connected and automated vehicles matters less as a transport result than as a demonstration that the value of geodetic precision is realised only through a dense, continuously operating reference network. (Benoit et al., 2015) then address the dimension that determines whether either of the preceding strengths is deployable at all: their low-cost, high-frequency Geocube network attacks the affordability barrier, converting high-precision monitoring from a capability that exists in principle into one feasible in resource-constrained settings.

The overall conclusion of these three studies is that GNSS, CORS, and InSAR technologies extend beyond geodesy and are increasingly applied in the fields of transportation, environmental monitoring, construction and safety. Future research should focus on integrating current methods, merging various sensor networks, and improving real-time data flow. Introducing such integration in Uzbekistan and advancing research in this direction will form the foundation for a modern CORS system in the country.

The findings of this case study should be interpreted within the scope of several limitations. First, access to data from government institutions is restricted, as most domain-specific information is considered confidential. This limits the completeness and transparency of the analysis. Second, due to data confidentiality, many studies have lost their relevance and are not directly aimed at addressing the current challenges. Third, practical constraints of real-time monitoring systems, such as network latency, reliability of data transmission channels, and the long-term stability of equipment, have not been sufficiently explored in a number of experimental studies. Fourth, the number of scientific works specifically focusing on the selected region remains limited, while fifth, the findings highlight a common limitation in practice, where simplified approaches such as buffer analysis are used instead of rigorous geodetic network evaluation methods.

Based on the identified limitations, several practical and scientific priorities can be outlined for future research. Strengthening collaboration between academic researchers and industry stakeholders is essential to improve data accessibility and facilitate more efficient data exchange. In addition, expanding national research efforts on geodetic networks, particularly CORS stations, is important for advancing scientific understanding in the Republic of Uzbekistan. Furthermore, gradual expansion of the territorial coverage of CORS stations should be considered, with the long-term goal of achieving full national coverage. At the same time, studying international best practices for the efficient deployment of CORS stations and adapting them to local conditions will support more effective network development. Overall, these directions will contribute to improving the reliability, efficiency and sustainability of geodetic infrastructure in data-limited environments.