Real-life Deployment Research Articles

Similarity-driven adversarial testing of neural networks

Although Convolutional Neural Networks (CNNs) are among the most important algorithms of computer vision and the artificial intelligence-based systems, they are vulnerable to adversarial attacks. Such attacks can cause dangerous consequences in real-life deployments. Consequently, testing of the artificial intelligence-based systems from their perspective is crucial to reliably support human prediction and decision-making through computation techniques under varying conditions. While proposing new effective attacks is important for neural network testing, it is also crucial to design effective strategies that can be used to choose target labels for these attacks. That is why, in this paper we propose a novel similarity-driven adversarial testing methodology for target label choosing. Our motivation is that CNNs, similarly to humans, tend to make mistakes mostly among categories they perceive similar. Thus, the effort to make models predict a particular class is not equal for all classes. Motivated by this, we propose to use the most and least similar labels to the ground truth according to different similarity measures to choose the target label for an adversarial attack. They can be treated as best- and worst-case scenarios in practical and transparent testing methodologies. As similarity is one of the key components of human cognition and categorization, the approach presents a shift towards a more human-centered security testing of deep neural networks. The obtained numerical results show the superiority of the proposed methods to the existing strategies in the targeted and the non-targeted testing setups.

Multi-category sorting of plastic waste using Swin Transformer: A vision-based approach

Sorting out plastic waste (PW) from municipal solid waste (MSW) by material type is crucial for reutilization and pollution reduction. However, current automatic separation methods are costly and inefficient, necessitating an advanced sorting process to ensure high feedstock purity. This study introduces a Swin Transformer-based model for effectively detecting PW in real-world MSW streams, leveraging both morphological and material properties. And, a dataset comprising 3560 optical images and infrared spectra data was created to support this task. This vision-based system can localize and classify PW into five categories: polypropylene (PP), polyethylene (PE), polyethylene terephthalate (PET), polyvinyl chloride (PVC), and polystyrene (PS). Performance evaluations reveal an accuracy rate of 99.75% and a mean Average Precision (mAP50) exceeding 91%. Compared to popular convolutional neural network (CNN)-based models, this well-trained Swin Transformer-based model offers enhanced convenience and performance in five-category PW detection task, maintaining a mAP50 over 80% in the real-life deployment. The model's effectiveness is further supported by visualization of detection results on MSW streams and principal component analysis of classification scores. These results demonstrate the system's significant effectiveness in both lab-scale and real-life conditions, aligning with global regulations and strategies that promote innovative technologies for plastic recycling, thereby contributing to the development of a sustainable circular economy.

The impact of quantum computing on the development of algorithms and software

Introduction: There is a great potential that the quantum computing can change the way of algorithms and software development more than classical computers. Thus, this article will try to focus on how algorithm design and software development can be affected by quantum computing as well as what possibilities could appear when quantum principles are implemented into traditional paradigms. This paper aims at identifying the impact of quantum computing on algorithm and software advancement, through a discussion of essential quantum algorithms, quantum languages, as well as the opportunities and challenges of quantum technologies. Method: An extensive literature review and theoretical investigation was also performed to investigate the foundational concepts of quantum computing and subsequent effects on algorithm and software engineering. Some of the research questions included exploring the contrast between classical and quantum algorithms, reviewing current literature on quantum programming languages, and delving into examples of real-life deployments of quantum algorithms cross numerous domains. Results: This paper shows that quantum computing brings qualitatively new paradigms in the algorithm design and function while the quantum algorithms such as Shor’s and Grover’s perform exponentially faster certain problems. Software development for quantum has brought the need to devise new frameworks of coding in light of probability in quantum circuit. It is also comforting to note that there is still effort being made although in its most embryonic form to create quantum programming languages like Qiskit and Cirq. Some of challenges include quantum decoherence; limited number of quantum hardware; and need for strong error correction processes.Conclusion: While there are currently relatively few quantum algorithms it is believed that the findings in this field have the ability to revolutionize algorithm and software design and subjects like cryptography, optimization and AI. However, trends in quantum computing show that the constraints to computational capabilities are likely to be lifted to allow creativity to develop the most powerful software solutions

Reference-Based OCT Angiogram Super-Resolution With Learnable Texture Generation.

Optical coherence tomography angiography (OCTA) can visualize retinal microvasculature and is important to qualitatively and quantitatively identify potential biomarkers for different retinal diseases. However, the resolution of optical coherence tomography (OCT) angiograms inevitably decreases when increasing the field-of-view (FOV) given a fixed acquisition time. To address this issue, we propose a novel reference-based super-resolution (RefSR) framework to preserve the resolution of the OCT angiograms while increasing the scanning area. Specifically, textures from the normal RefSR pipeline are used to train a learnable texture generator (LTG), which is designed to generate textures according to the input. The key difference between the proposed method and traditional RefSR models is that the textures used during inference are generated by the LTG instead of being searched from a single reference (Ref) image. Since the LTG is optimized throughout the whole training process, the available texture space is significantly enlarged and no longer limited to a single Ref image, but extends to all textures contained in the training samples. Moreover, our proposed LTGNet does not require an Ref image at the inference phase, thereby becoming invulnerable to the selection of the Ref image. Both experimental and visual results show that LTGNet has competitive performance and robustness over state-of-the-art methods, indicating good reliability and promise in real-life deployment. The source code is available at https://github.com/RYY0722/LTGNet.

Teledrive: An Embodied AI Based Telepresence System

This article presents ‘Teledrive’, a telepresence robotic system with embodied AI features that empowers an operator to navigate the telerobot in any unknown remote place with minimal human intervention. We conceive Teledrive in the context of democratizing remote ‘care-giving’ for elderly citizens as well as for isolated patients, affected by contagious diseases. In particular, this paper focuses on the problem of navigating to a rough target area (like ‘bedroom’ or ‘kitchen’) rather than pre-specified point destinations. This ushers in a unique ‘AreaGoal’ based navigation feature, which has not been explored in depth in the contemporary solutions. Further, we describe an edge computing-based software system built on a WebRTC-based communication framework to realize the aforementioned scheme through an easy-to-use speech-based human-robot interaction. Moreover, to enhance the ease of operation for the remote caregiver, we incorporate a ‘person following’ feature, whereby a robot follows a person on the move in its premises as directed by the operator. Moreover, the system presented is loosely coupled with specific robot hardware, unlike the existing solutions. We have evaluated the efficacy of the proposed system through baseline experiments, user study, and real-life deployment.

Dynamic Sub-graph Distillation for Robust Semi-supervised Continual Learning

Continual learning (CL) has shown promising results and comparable performance to learning at once in a fully supervised manner. However, CL strategies typically require a large number of labeled samples, making their real-life deployment challenging. In this work, we focus on semi-supervised continual learning (SSCL), where the model progressively learns from partially labeled data with unknown categories. We provide a comprehensive analysis of SSCL and demonstrate that unreliable distributions of unlabeled data lead to unstable training and refinement of the progressing stages. This problem severely impacts the performance of SSCL. To address the limitations, we propose a novel approach called Dynamic Sub-Graph Distillation (DSGD) for semi-supervised continual learning, which leverages both semantic and structural information to achieve more stable knowledge distillation on unlabeled data and exhibit robustness against distribution bias. Firstly, we formalize a general model of structural distillation and design a dynamic graph construction for the continual learning progress. Next, we define a structure distillation vector and design a dynamic sub-graph distillation algorithm, which enables end-to-end training and adaptability to scale up tasks. The entire proposed method is adaptable to various CL methods and supervision settings. Finally, experiments conducted on three datasets CIFAR10, CIFAR100, and ImageNet-100, with varying supervision ratios, demonstrate the effectiveness of our proposed approach in mitigating the catastrophic forgetting problem in semi-supervised continual learning scenarios. Our code is available: https://github.com/fanyan0411/DSGD.

SnTe/SnSe Heterojunction Based Ammonia Sensors with Excellent Withstand to Ambient Humidities.

Non-invasive breath testing has gained increasing importance for early disease screening, spurring research into cheap sensors for detecting trace biomarkers such as ammonia. However, real-life deployment of ammonia sensors remains hindered by susceptibility to humidity-induced interference. The SnTe/SnSe heterojunction-based chemiresistive-type sensor demonstrates an excellent response/recovery to different concentrations of ammonia from 0.1 to 100ppm at room temperature. The improved sensing properties of the heterojunctions-based sensors compared to single-phased SnTe or SnSe can be attributed to the stronger NH3 adsorptions, more Te vacancies, and hydrophobic surface induced by the formed SnTe/SnSe heterojunctions. The sensing mechanisms are investigated in detail by using in situ techniques such as diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS), Kelvin probe, and a.c. impedance spectroscopy together with the Density-Function-Theory calculations. The formed heterojunctions boost the overall charge transfer efficiency between the ammonia and the sensing materials, thus leading to the desirable sensing features as well, with excellent resistance to ambient humidities.

Validation and interpretation of a multimodal drowsiness detection system using explainable machine learning

Background and objectiveDrowsiness behind the wheel is a major road safety issue with efforts focused on developing drowsy driving detection systems. However, most drowsy driving detection studies using physiological signals have focused on developing a 'black box' machine learning classifier, with much less focus on 'robustness' and 'explainability'—two crucial properties of a trustworthy machine learning model. Therefore, this study has focused on using multiple validation techniques to evaluate the overall performance of such a system using multiple supervised machine learning-based classifiers and then unbox the black box model using explainable machine learning. MethodsDriving was simulated via a 30-minute psychomotor vigilance task while the participants reported their level of subjective sleepiness with their physiological signals: electroencephalogram (EEG), electrooculogram (EOG) and electrocardiogram (ECG) being recorded. Six different techniques, comprising subject-dependent and independent techniques were applied for model validation and robustness testing with three supervised machine learning classifiers, namely K-nearest neighbours (KNN), support vector machines (SVM) and random forest (RF), and two explainable methods, namely SHapley Additive exPlanation (SHAP) analysis and partial dependency analysis (PDA) were leveraged for model interpretation. ResultsThe study identified the leave one participant out, a subject-independent validation technique to be most useful, with the best sensitivity of 70.3 %, specificity of 82.2 %, and an accuracy of 80.1 % using the random forest classifier in addressing the autocorrelation issue due to inter-individual differences in physiological signals. Moreover, the explainable results suggest most important physiological features for drowsiness detection, with a clear cut-off in the decision boundary. ConclusionsThe implication of the study will ensure a rigorous validation for robustness testing and an explainable machine learning approach to developing a trustworthy drowsiness detection system and enhancing road safety. The explainable machine learning-based results show promise in real-life deployment of the physiological-signal based in-vehicle trustworthy drowsiness detection system, with higher reliability and explainability, along with a lower system cost.

Novel Concepts on Domination in Neutrosophic Incidence Graphs with Some Applications

In graph theory, the concept of domination is essential in a variety of domains. It has broad applications in diverse fields such as coding theory, computer network models, and school bus routing and facility location problems. If a fuzzy graph fails to obtain acceptable results, neutrosophic sets and neutrosophic graphs can be used to model uncertainty correlated with indeterminate and inconsistent information in arbitrary real-world scenario. In this study, we consider the concept of domination as it relates to single-valued neutrosophic incidence graphs (SVNIGs). Given the importance of domination and its utilization in numerous fields, we propose the application of dominating sets in SVNIG with valid edges. We present some relevant definitions such as those of valid edges, cardinality, and isolated vertices in SVNIG along with some examples. Furthermore, we also show a few significant sets connected to the dominating set in an SVNIG such as independent and irredundant sets. We also investigate a relationship between the concepts of dominating sets and domination numbers as well as irredundant and independence sets in SVNIGs. Finally, a real-life deployment of domination in SVNIGs is investigated in relation to COVID-19 vaccination locations as a practical application.

Protecting critical infrastructure with game theory, optimization techniques, and AI algorithms

In light of recent geopolitical developments, Europe and Poland are acutely aware of the urgent importance of infrastructure security. Despite heightened interest and increased investments, our security resources remain severely limited, rendering continuous protection for every potential target unattainable. Consequently, the strategic allocation of security resources becomes an ongoing imperative. This paper presents a short introduction to the core principles behind advanced methods that facilitate automated decision-making in security resource allocation. These methods leverage artificial intelligence (AI), game theory, and optimization techniques, and have demonstrated their effectiveness through multiple real-life deployments in the USA. We also provide a concise overview of this exciting body of research and discuss the solutions and software developed by our team, “AI for Security” at the IDEAS NCBR research institute to protect critical infrastructure in Poland and in Europe.

Conflicting Interactions among Protection Mechanisms for Machine Learning Models

Nowadays, systems based on machine learning (ML) are widely used in different domains. Given their popularity, ML models have become targets for various attacks. As a result, research at the intersection of security/privacy and ML has flourished. Typically such work has focused on individual types of security/privacy concerns and mitigations thereof. However, in real-life deployments, an ML model will need to be protected against several concerns simultaneously. A protection mechanism optimal for a specific security or privacy concern may interact negatively with mechanisms intended to address other concerns. Despite its practical relevance, the potential for such conflicts has not been studied adequately. In this work, we first provide a framework for analyzing such conflicting interactions. We then focus on systematically analyzing pairwise interactions between protection mechanisms for one concern, model and data ownership verification, with two other classes of ML protection mechanisms: differentially private training, and robustness against model evasion. We find that several pairwise interactions result in conflicts. We also explore potential approaches for avoiding such conflicts. First, we study the effect of hyperparameter relaxations, finding that there is no sweet spot balancing the performance of both protection mechanisms. Second, we explore whether modifying one type of protection mechanism (ownership verification) so as to decouple it from factors that may be impacted by a conflicting mechanism (differentially private training or robustness to model evasion) can avoid conflict. We show that this approach can indeed avoid the conflict between ownership verification mechanisms when combined with differentially private training, but has no effect on robustness to model evasion. We conclude by identifying the gaps in the landscape of studying interactions between other types of ML protection mechanisms.

Analysis of a decentralized Bayesian optimization algorithm for improving spatial reuse in dense WLANs

Despite representing the prominent means of accessing the Internet, WLANs remain subject to performance issues, which may be mitigated through more efficient spatial reuse of radio channels. In this perspective, the IEEE 802.11ax amendment enables the dynamic update of two key parameters in wireless transmission: the transmission power (TX_POWER) and the sensitivity threshold (OBSS_PD). In this paper, we present INSPIRE, a distributed online learning solution performing local Bayesian optimizations based on Gaussian processes to improve spatial reuse in WLANs. INSPIRE makes no explicit assumptions on the WLANs’ topology and favors altruistic behaviors of the access points in their search for adequate configurations of their TX_POWER and OBSS_PD parameters. INSPIRE can easily be extended to work with a limited number of observations to throttle its computational complexity. We demonstrate the superiority of INSPIRE over other state-of-the-art strategies using the ns-3 simulator and two examples inspired by real-life deployments of dense WLANs. Our results show that, in only a few seconds, INSPIRE is able to drastically increase the quality of service of operational WLANs by improving their fairness and throughput. Finally, we discuss the configurations recommended by INSPIRE. We show that they comply with an 802.11ax empirical recommendation, and we correlate their values with some graph-based metrics of the WLAN topologies.

Impacts of Chronic Obstructive Pulmonary Disease Care Pathway on Healthcare Utilization and Costs: A Matched Multiple Control Cohort Study in Saskatchewan, Canada

ObjectivesThis study aimed to evaluate the real-world impacts of a chronic obstructive pulmonary disease (COPD) care pathway program on healthcare utilization and costs in Saskatchewan, Canada. MethodsA difference-in-differences evaluation of a real-life deployment of a COPD care pathway, using patient-level administrative health data in Saskatchewan, was conducted. The intervention group (n = 759) included adults (35+ years) with spirometry-confirmed COPD diagnosis recruited into the care pathway program in Regina between April 1, 2018 and March 31, 2019. The 2 control groups comprised adults (35+ years) with COPD who lived in Saskatoon during the same period (n = 759) or Regina between April 1, 2015 and March 31, 2016 (n = 759) who did not participate in the care pathway. ResultsCompared with the individuals in the Saskatoon control groups, individuals in the COPD care pathway group had shorter inpatient hospital length of stay (average treatment effect on the treated [ATT] −0.46, 95% CI −0.88 to −0.04) but a higher number of general practitioner visits (ATT 1.46, 95% CI 1.14 to 1.79) and specialist physician visits (ATT 0.84, 95% CI 0.61 to 1.07). Regarding healthcare costs, individuals in the care pathway group had higher COPD-related specialist visit costs (ATT $81.70, 95% CI $59.45 to $103.96) but lower COPD-related outpatient drug dispensation costs (ATT −$4.81, 95% CI −$9.34 to −$0.27). ConclusionsThe care pathway reduced inpatient hospital length of stay, but increased general practitioner and specialist physician visits for COPD-related services within the first year of implementation.

A framework for modeling and analyzing cyber-physical systems using statistical model checking

The trustworthiness of a cyber–physical system is essential for it to be qualified for utilization in most real-life deployments. This is especially critical for systems that deal with precious human lives. Although these safety-critical systems can be investigated using both experimental testing and model-based verification, accurate models have the potential to permit risk-free mimicking of the system behavior even in the most extreme scenarios. To overcome the CPS modeling and design challenges, the INCOSE/OMG standard System Modeling Language (SysML) is utilized in this work to accurately specify cyber–physical systems. For that, a bounded set of SysML constructs are defined to precisely capture the semantics of continuous-time and discrete-time system behaviors. Then, the SysML constructs are substituted by developing a new algebra, called Enhanced Activity Calculus (EAC). So, EAC helps construct equivalent priced timed automata models by developing a new systematic procedure to correctly translate the SysML models into the statistical model checking tool UPPAAL-SMC inputs. The latter checks whether the system is correct and safe or not. Moreover, the soundness of the developed translation mechanism has been proved and its effectiveness has been shown on a real use case, namely the artificial pancreas.

An Overview of Machine Learning Techniques in Local Path Planning for Autonomous Underwater Vehicles

Autonomous underwater vehicles (AUVs) have become attractive and essential for underwater search and exploration because of the advantages they offer over manned underwater vehicles. Hence the need to improve AUV technologies. One crucial area of AUV technology involves efficiently solving the path planning problem. Several approaches have been identified from the literature for AUV global and local path planning. The use of machine learning (ML) techniques in overcoming some of the challenges associated with AUV path planning problems such as safety and obstacle avoidance, energy consumption, and optimal time and distance travelled remains an active research area. While there is literature on global and local path planning that explores different techniques, there is still a lack of paper that provides an overview of the application of ML for local path planning. Hence the main objective of this paper is to present an overview of the state-of-the-art application of ML techniques on local path planning for AUVs. The ML algorithms are discussed under supervised, unsupervised, and reinforcement learning. The challenges faced in real-life deployment, simulated scenarios, computational issues, and application of ML algorithms are discussed, with future research directions presented.

Ammonium-ion energy storage devices for real-life deployment: storage mechanism, electrode design and system integration

This review will pave the way to next-generation ammonium ion storage systems from fundamental science breakthroughs, practical prototypes and eventually to real-life deployment and commercialization.

LoCKAmp: lab-on-PCB technology for

The recent COVID-19 outbreak highlighted the need for lab-on-chip diagnostic technology fit for real-life deployment in the field. Existing bottlenecks in multistep analytical microsystem integration and upscalable, standardized fabrication techniques delayed the large-scale deployment of lab-on-chip solutions during the outbreak, throughout a global diagnostic test shortage. This study presents a technology that has the potential to address these issues by redeploying and repurposing the ubiquitous printed circuit board (PCB) technology and manufacturing infrastructure. We demonstrate the first commercially manufactured, miniaturised lab-on-PCB device for loop-mediated isothermal amplification (LAMP) genetic detection of SARS-CoV-2. The system incorporates a mass-manufactured, continuous-flow PCB chip with ultra-low cost fluorescent detection circuitry, rendering it the only continuous-flow μLAMP platform with off-the-shelf optical detection components. Ultrafast, SARS-CoV-2 RNA amplification in wastewater samples was demonstrated within 2 min analysis, at concentrations as low as 17 gc μL-1. We further demonstrate our device operation by detecting SARS-CoV-2 in 20 human nasopharyngeal swab samples, without the need for any RNA extraction or purification. This renders the presented miniaturised nucleic-acid amplification-based diagnostic test the fastest reported SARS-CoV-2 genetic detection platform, in a practical implementation suitable for deployment in the field. This technology can be readily extended to the detection of alternative pathogens or genetic targets for a very broad range of applications and matrices. LoCKAmp lab-on-PCB chips are currently mass-manufactured in a commercial, ISO-compliant PCB factory, at a small-scale production cost of £2.50 per chip. Thus, with this work, we demonstrate a high technology-readiness-level lab-on-chip-based genetic detection system, successfully benchmarked against standard analytical techniques both for wastewater and nasopharyngeal swab SARS-CoV-2 detection.

Contrast phase recognition in liver computer tomography using deep learning

Hepatocellular carcinoma (HCC) has become the 4th leading cause of cancer-related deaths, with high social, economical and health implications. Imaging techniques such as multiphase computed tomography (CT) have been successfully used for diagnosis of liver tumors such as HCC in a feasible and accurate way and its interpretation relies mainly on comparing the appearance of the lesions in the different contrast phases of the exam. Recently, some researchers have been dedicated to the development of tools based on machine learning (ML) algorithms, especially by deep learning techniques, to improve the diagnosis of liver lesions in imaging exams. However, the lack of standardization in the naming of the CT contrast phases in the DICOM metadata is a problem for real-life deployment of machine learning tools. Therefore, it is important to correctly identify the exam phase based only on the image and not on the exam metadata, which is unreliable. Motivated by this problem, we successfully created an annotation platform and implemented a convolutional neural network (CNN) to automatically identify the CT scan phases in the HCFMUSP database in the city of São Paulo, Brazil. We improved this algorithm with hyperparameter tuning and evaluated it with cross validation methods. Comparing its predictions with the radiologists annotation, it achieved an accuracy of 94.6%, 98% and 100% in the testing dataset for the slice, volume and exam evaluation, respectively.

Accurate Real-Life Chinese Dish Recognition

Deep learning is a new research direction in the field of machine learning, which was introduced into machine learning to bring it closer to its original goal. Accurate dish recognition becomes increasingly important in the multimedia community since it can help cuisine recommendation, calorie management, service improvement, and other food computing tasks. Many novel approaches have been developed on web recipes and menu pictures, while few are concerned real-life dish image analysis. In this study, a deep learning-based prototype system is deployed in a Chinese canteen, and 28 dish types, 16,904 images, and 45,061 instances have been collected. Specifically, in the prototype system, three practical issues are explored, including the backbone network selection, the training strategy determination, and the minimum number of samples for model upgrading. Experimental results suggest that fine-tuned Faster-RCNN can serve as the backbone network of the prototype system since it outperforms the other four fine-tuned networks on dish recognition (accuracy, 98.10%; recall, 97.20%; MAP (mean average precession), 98.30%) and satisfies real-time requirement (0.15 second per image). Meanwhile, the transferred backbone network achieves superior results (MAP, 96.48%) over the same architecture trained from image scratches (MAP, 87.84%). On model upgrading, a good (MAP, 91.34%) to better (MAP, 96.48%) outcome is obtained when the training size is increased from 50 to 200 samples per dish type, and 150 and more instances should be annotated if a new dish type is added to the system’s recognition list. Conclusively, the real-life deployment and evaluation of the prototype system indicate that deep learning is full of potential to enhance customer experience through accurate daily dish recognition.

A Sustainable Business Model for a Neutral Host Supporting 5G and beyond (5GB) Ultra-Dense Networks: Challenges, Directions, and Architecture.

With the deployment of the fifth generation (5G) mobile network systems and the envisioned heterogeneous ultra-dense networks (UDNs), both small cell (SmC) and distributed antenna system (DAS) technologies are required by mobile network operators (MNOs) and venue owners to support multiple spectrum bands, multiple radio access technologies (RATs), multiple optical central offices (COs), and multiple MNOs. As a result, the neutral host business model representing a third party responsible for managing the network enterprise on behalf of multiple MNOs has emerged as a potential solution, mainly influenced by the desire to provide a high user experience without significantly increasing the total cost of ownership (TCO). However, designing a sustainable business model for a neutral host is a nontrivial task, especially when considered in the context of 5G and beyond (5GB) UDNs. In this paper, under an integrated optical wireless network infrastructure, we review how SmC and DAS technologies are evolving towards the adoption of the neutral host business model and identify key challenges and requirements for 5GB support. Thus, we explore recent candidate advancements in heterogeneous network integration technologies for the realization of an efficient 5GB neutral host business model design capable of accommodating both SmC and DAS. Furthermore, we propose a novel design architecture that relies on virtual radio access network (vRAN) to enable real-time dynamic resource allocation and radio over Ethernet (RoE) for flexible and reconfigurable fronthaul. The results from our simulations using MATLAB over two real-life deployment scenarios validate the feasibility of utilizing switched RoE considering end-to-end delay requirements of 5GB under different switching schemes, as long as the queuing delay is kept to a minimum. Finally, the results show that incorporating RoE and vRAN technologies into the neutral host design results in substantial TCO reduction by about 81% in an indoor scenario and 73% in an outdoor scenario.