Searcharxiv⌕ Search

arXiv subjects

Milena Radenkovic

Publications and source records attributed to Milena Radenkovic.

At least 19 recordsLinked to original sources

Distributed Intelligent Ad Hoc Routing Agents for Communication at Protests During Internet Shutdowns

Governments around the world are increasingly using internet shutdowns to block digital communication. Internet shutdowns are particularly dangerous during political protests, as they inhibit protesters' ability to organize and communicate critical, potentially life-saving information and may be used to prevent awareness of state-based violence. Recent prior works have explored secure, private messaging systems for use during political protests and internet shutdowns but have encountered long delivery delays or low delivery rates under challenging conditions. In this work, we explore the use of distributed, adaptive, social- and congestion-aware heuristics and efficient communication patterns to create an intelligent ad hoc routing protocol in which network agents learn and leverage the network social environment and monitor and respond to dynamic network congestion behaviour to route messages in a time- and resource-efficient manner. Through this exploration, we contribute an intelligent algorithm and protocol which overall achieves comparable or higher delivery rates relative to the leading dynamic clique routing protocol which it extends while also transmitting messages through notably fewer carriers and, when network devices have larger amounts of buffer storage, transmitting fewer message-related packets. We additionally contribute a further understanding of routing limitations and challenges and conceptualizations of extensions of intelligent dynamic clique routing to overcome them.

cs.NI↗

NLP-based Message Prioritisation for Opportunistic Routing: A Case Study of Nepal Earthquake Emergency Communication

This dissertation investigates whether Natural Language Processing (NLP) based semantic message prioritisation can improve the treatment of urgent disaster messages in opportunistic emergency communication, using the 2015 Nepal earthquake as a case study. TF IDF with Logistic Regression and BERT are used to classify social media messages into nine humanitarian categories, which are mapped to four routing priorities from P1 to P4. These priorities are integrated into the ONE simulator within a controlled Kathmandu earthquake scenario using a modified Binary Spray and Wait router. BERT achieves higher priority accuracy and P1 recall than Logistic Regression, while reducing under prioritisation and severe priority errors. However, improved NLP performance does not produce uniform network gains. P1 delivery probability remains high and similar across routing conditions, and no consistent latency reduction is observed. The clearest difference appears under buffer pressure, where the BERT condition shows fewer observed ground truth P1 buffer losses than Logistic Regression. The results indicate that semantic prioritisation is most valuable when limited transmission and storage resources are contested, although it cannot overcome mobility and contact constraints in opportunistic networks.

cs.NI↗

Lightweight Machine Learning for Classifying Disaster-Related Social Media Posts

Classifying disaster-related social-media posts into humanitarian information labels is challenging because individual messages are typically short and context-limited, humanitarian categories can overlap semantically, and the CrisisBench label distribution is strongly imbalanced. This study investigates whether lightweight lexical machine-learning methods can provide effective and computationally efficient classification under these conditions. Using the CrisisBench v1.0 English humanitarian task, comprising 87,455 records across predefined Train, Development and Test splits, we conduct a controlled comparison of TF-IDF representations and lightweight classifiers, with Macro-F1 as the primary model-selection metric. Beyond aggregate predictive performance, the study jointly examines per-label behaviour, directional error patterns, relative decision margins, computational requirements, and the effect of domain-specific training through a tropical-cyclone specialist model. The selected balanced LinearSVC with unigram-bigram TF-IDF and C=0.5 achieved an Accuracy of 0.8282, a Macro-F1 of 0.6578 and a Weighted-F1 of 0.8315 on the held-out Test split. Lower decision margins were substantially associated with prediction errors, while the tropical-cyclone specialist achieved modestly higher in-domain aggregate performance with substantially lower feature and fitting requirements. The study contributes a controlled, label-sensitive and resource-aware evaluation of lightweight humanitarian text classification that characterises not only aggregate performance, but also where these models fail and how their behaviour changes under domain-specific training.

cs.NI↗

Traffic Congestion Awareness and On-Demand Distribution in Vehicular Delay-Tolerant Networks in California I-210 Freeway

In vehicular networks under edge computing environments, vehicle-to-vehicle delay-tolerant networking (V-DTN) can disseminate congestion warnings to other vehicles via a store-carry-forward mechanism, helping them proactively choose suitable routes. However, most existing in-vehicle information dissemination methods rely on flooding or limited flooding strategies, broadcasting alerts across the entire network whenever congestion is detected. This leads to excessive redundant copies and consumes node cache space. To address this issue, this paper proposes a congestion-event- aware on-demand message dissemination mechanism. By considering the recurrence and duration of congestion, the mechanism suppresses broadcasts of short-lived, self-dissipating congestion events. Based on real-world PeMS data from California's I-210 corridor, we construct an I-210 Freeway Traffic Congestion Use Case. Experiments show that predicting congestion recurrence heavily relies on historical free-flow speed data of lanes. Without incorporating additional feature dimensions, the accuracy of modeling and comparing lane free-flow speeds across different day types outweighs the choice of machine learning models, and such patterns are difficult to reproduce in simulators. Meanwhile, congestion duration prediction serves effectively as the basis for dynamic-threshold on-demand dissemination in V-DTN, which demonstrate that machine learning-based congestion awareness combined with dynamic-threshold on-demand dissemination significantly reduces network resource consumption and network-layer congestion, enhances delivery reliability, and lowers delivery latency.

cs.NI↗

Investigating the Suitability of Delay Tolerant Networks for Broadcasting Tsunami Warnings in Palu, Indonesia

On the 28th of September, 2018, a tsunami hit the city of Palu in Indonesia, killing 4,340 people. The earthquake preceding the tsunami crippled communication lines and may have rendered the transmission of tsunami warning messages using traditional end-to-end approaches impossible. This paper proposes an alternative approach using Delay Tolerant Networks (DTNs) for tsunami warning message routing given their resilience to disruptions and sparse connections. Both Epidemic and Spray and Wait routing protocols were simulated in a pseudo-realistic environment to evaluate their effectiveness for transmitting tsunami warning messages in Palu. Results indicated that these protocols are not suitable for the tight time constraints of post-earthquake tsunami warnings with the currently available technology. However, they may have promising applications for the earthquakes that precede tsunamis.

cs.NI↗

Performance Evaluation of Delay Tolerant Network Protocols to Improve Nepal Earthquake Rescue Communications

In the fields of disaster rescue and communication in extreme environments, Delay Tolerant Network (DTN) has become an important technology due to its "store-carry-forward" mechanism. Selecting the appropriate routing strategy is of crucial significance for improving the success rate of distress message transmission and reducing delays in material dispatch. We design a pseudo realistic use case of Nepal Kathmandu earthquake rescue based on dynamically changing population distribution model and characteristics of rescue activities in the initial rescue efforts in Nepal Kathmandu earthquakes to conducted the multi criteria two benchmark routing protocols performance analysis in the face of different buffer sizes of the rescue team nodes. We identify multiple real world node groups, including affected residents, rescue teams, drones and ground vehicles and communication models are established according to the movement behaviors of these groups. We analyze the communication of distress messages between edge nodes to obtain performance metrics such as delivered probability, average delay, hop count, and buffer time. By analyzing the multi layer complex data and protocols differences, the research results show the effectiveness of distributed DTN communication methods in the Nepal earthquake rescue use case, reveal existence of trade-offs between transmission reliability and resource utilization of different routing protocols in disaster communication environment and provide a basis for the design of next-generation emergency communication services based on edge nodes.

cs.NI↗

Evaluating Performance Characteristic of Opportunistic Routing Protocols: A Case Study of the 2016 Italian League Match Earthquake in the Stadio Adriatico

Delay Tolerant Networks (DTNs) can provide emergency communication support when conventional infrastructure is disrupted during disasters. This paper evaluates the performance of opportunistic routing protocols in a realistic disaster scenario based on the 2016 Central Italy earthquake, modelled as an emergency occurring during a football match at Stadio Adriatico in Pescara. We identify multiple suitable groups of mobile and static nodes, such as audiences, a range of different emergency responders, stage sensors, and vehicles, to design and build evacuation and rescue activities in a partially connected environment. Two representative DTN routing protocols, Epidemic and Spray and Wait, are tested under identical simulation settings and compared using delivery probability, latency, overhead ratio, hop count and dropped messages. The results highlight that Spray and Wait provides a better balance between reliability and efficiency in this scenario, achieving higher delivery probability while reducing overhead and using network resources more efficiently. The study shows the usefulness of DTN simulation for analysing disaster communication performance in emergency response scenarios.

cs.NI↗

A-FC: An Activity-Based Delay Tolerant Routing Protocol for Improving Future School Campus Emergency Communications

School Campus emergency communication systems are vital for safeguarding student safety during sudden disasters such as typhoons, which frequently cause widespread paralysis of communication infrastructure. Traditional Delay-Tolerant Network (DTN) protocols, such as Direct Delivery and First Contact, struggle to maintain reliable connections in such scenarios due to high latency and low delivery rates. This paper proposes the Activity-based First Contact (A-FC) protocol, an innovative routing scheme that leverages real-world social roles to overcome network partitioning by mandatorily uploading messages to highly active "staff nodes". We constructed a real-world evaluation scenario based on the topology of Fuzhou No. 1 Middle School. Simulation results demonstrate that the A-FC protocol significantly outperforms baseline protocols, achieving approximately 68% message delivery probability and reducing average delay to 4311 seconds. With an average hop count of merely 1.68, this protocol establishes a low-cost, highly reliable backup communication model for school campus disaster response.

cs.NI↗

Delay-Tolerant Networking for Tsunami Evacuation on the Small Island of Hachijojima: A Study of Epidemic and Prophet Routing

Tsunami disasters pose a serious and recurring threat to coastal and island communities. When a large earthquake occurs, people are forced to make evacuation decisions under extreme time pressure, often at the same time as the communication infrastructure is damaged or completely lost. In such circumstances, the familiar channels for sharing information - cellular networks, the internet, and even landlines - can no longer be relied upon. What typically remains are the mobile devices that evacuees carry with them. These devices can form Delay Tolerant Networks (DTNs), in which messages are forwarded opportunistically whenever people come into contact. To explore this, we evaluate multi-criteria performance characteristics of two DTN routing schemes in a pre-tsunami evacuation scenario for the island of Hachijojima, Japan use case.

cs.NI↗

Distributed Accountability in Democracy: Using MANETs and DTNs in the Face of Acts of Questionable Legality

In this paper, we explore the behavior of the Epidemic and Wave DTN routing protocols in a realistic setting where individuals may wish to communicate with others for support regarding an act of questionable legality. We identify situations where using the Epidemic routing protocol may be more advantageous in such a scenario, and situations where using the Wave routing protocol may be more advantageous instead. We discuss other aspects of our findings in detail and suggest multiple approaches to future works.

cs.NI↗

Improving Reliability of Human Trafficking Alerts in Airports

This paper investigates the latter scenario of individual emergency alerts in airports by applying two existing benchmark delay tolerant network protocols and evaluating their performance of delivery ratio and latency. First, the paper provides a background on Mobile Ad Hoc Networks (MANETs) and Delay Tolerant Networks (DTNs), as well as Vehicular Ad Hoc Networks (VANETs) as a subset of MANETs. Next, the scenario is simulated using the Opportunistic Network Environment (ONE) simulator and runs the DTN protocols applying Spray and Wait and Epidemic. The study discusses the results, highlighting the advantages and limitations of each protocol within the scenario and addressing constraints of the simulation or experimental setup. A wider discussion then considers related research on technologies that combat human trafficking and the potential role of DTN networks in improving this global issue for the better.

cs.NI↗

Multi-Layer Perceptron-Based Relay Node Selection for Next-Generation Intelligent Delay-Tolerant Networks

Delay Tolerant Networks (DTNs) are critical for emergency communication in highly dynamic and challenging scenarios characterized by intermittent connectivity, frequent disruptions, and unpredictable node mobility. While some protocols are widely adopted for simplicity and low overhead, their static replication strategy lacks the ability to adaptively distinguish high-quality relay nodes, often leading to inefficient and suboptimal message dissemination. To address this challenge, we propose a novel intelligent routing enhancement that integrates machine learning-based node evaluation into the Spray and Wait framework. Several dynamic, core features are extracted from simulation logs and are used to train multiple classifiers - Multi-Layer Perceptron (MLP), Support Vector Machine (SVM), and Random Forest (RF) - to predict whether a node is suitable as a relay under dynamic conditions. The trained models are deployed via a lightweight Flask-based RESTful API, enabling real-time, adaptive predictions. We implement the enhanced router MLPBasedSprayRouter, which selectively forwards messages based on the predicted relay quality. A caching mechanism is incorporated to reduce computational overhead and ensure stable, low-latency inference. Extensive experiments under realistic emergency mobility scenarios demonstrate that the proposed framework significantly improves delivery ratio while reducing average latency compared to the baseline protocols. Among all evaluated classifiers, MLP achieved the most robust performance, consistently outperforming both SVM and RF in terms of accuracy, adaptability, and inference speed. These results confirm the novelty and practicality of integrating machine learning into DTN routing, paving the way for resilient and intelligent communication systems in smart cities, disaster recovery, and other dynamic environments.

cs.NI↗

Domain Adaptive SAR Wake Detection: Leveraging Similarity Filtering and Memory Guidance

Synthetic Aperture Radar (SAR), with its all-weather and wide-area observation capabilities, serves as a crucial tool for wake detection. However, due to its complex imaging mechanism, wake features in SAR images often appear abstract and noisy, posing challenges for accurate annotation. In contrast, optical images provide more distinct visual cues, but models trained on optical data suffer from performance degradation when applied to SAR images due to domain shift. To address this cross-modal domain adaptation challenge, we propose a Similarity-Guided and Memory-Guided Domain Adaptation (termed SimMemDA) framework for unsupervised domain adaptive ship wake detection via instance-level feature similarity filtering and feature memory guidance. Specifically, to alleviate the visual discrepancy between optical and SAR images, we first utilize WakeGAN to perform style transfer on optical images, generating pseudo-images close to the SAR style. Then, instance-level feature similarity filtering mechanism is designed to identify and prioritize source samples with target-like distributions, minimizing negative transfer. Meanwhile, a Feature-Confidence Memory Bank combined with a K-nearest neighbor confidence-weighted fusion strategy is introduced to dynamically calibrate pseudo-labels in the target domain, improving the reliability and stability of pseudo-labels. Finally, the framework further enhances generalization through region-mixed training, strategically combining source annotations with calibrated target pseudo-labels. Experimental results demonstrate that the proposed SimMemDA method can improve the accuracy and robustness of cross-modal ship wake detection tasks, validating the effectiveness and feasibility of the proposed method.

cs.CV↗

EfficientNet-Based Multi-Class Detection of Real, Deepfake, and Plastic Surgery Faces

Currently, deep learning has been utilised to tackle several difficulties in our everyday lives. It not only exhibits progress in computer vision but also constitutes the foundation for several revolutionary technologies. Nonetheless, similar to all phenomena, the use of deep learning in diverse domains has produced a multifaceted interaction of advantages and disadvantages for human society. Deepfake technology has advanced, significantly impacting social life. However, developments in this technology can affect privacy, the reputations of prominent personalities, and national security via software development. It can produce indistinguishable counterfeit photographs and films, potentially impairing the functionality of facial recognition systems, so presenting a significant risk. The improper application of deepfake technology produces several detrimental effects on society. Face-swapping programs mislead users by altering persons' appearances or expressions to fulfil particular aims or to appropriate personal information. Deepfake technology permeates daily life through such techniques. Certain individuals endeavour to sabotage election campaigns or subvert prominent political figures by creating deceptive pictures to influence public perception, causing significant harm to a nation's political and economic structure.

cs.CV↗

ML-MaxProp: Bridging Machine Learning and Delay-Tolerant Routing for Resilient Post-Disaster Communication

In disaster-stricken and large-scale urban emergency scenarios, ensuring reliable communication remains a formidable challenge, as collapsed infrastructure, unpredictable mobility, and severely constrained resources disrupt conventional networks. Delay-Tolerant Networks (DTNs), though resilient through their store-carry-forward paradigm, reveal the fundamental weaknesses of classical protocols - Epidemic, Spray-and-Wait, and MaxProp - when confronted with sparse encounters, buffer shortages, and volatile connectivity. To address these obstacles, this study proposes ML-MaxProp, a hybrid routing protocol that strengthens MaxProp with supervised machine learning. By leveraging contextual features such as encounter frequency, hop count, buffer occupancy, message age, and time-to-live (TTL), ML-MaxProp predicts relay suitability in real time, transforming rigid heuristics into adaptive intelligence. Extensive simulations in the ONE environment using the Helsinki SPMBM mobility model show that ML-MaxProp consistently surpasses baseline protocols, achieving higher delivery probability, lower latency, and reduced overhead. Statistical validation further shows that these improvements are both significant and robust, even under highly resource-constrained and unstable conditions. Overall, this work shows that ML-MaxProp is not just an incremental refinement but a lightweight, adaptive, and practical solution to one of the hardest challenges in DTNs: sustaining mission-critical communication when infrastructure collapses and every forwarding decision becomes critical.

cs.NI↗

Resilient Communication For Avalanche Response in Infrastructure-Limited Environments

Delay Tolerant Networks (DTNs) offer a promising paradigm for maintaining communication in infrastructure limited environments, such as those encountered during natural disasters. This paper investigates the viability of leveraging an existing national transport system - the Swiss rail network - as a data mule backbone for disseminating critical avalanche alerts. Using The Opportunistic Network Environment (ONE) simulator, we model the entire Swiss rail network and conduct a rigorous comparative analysis of two seminal DTN routing protocols: Epidemic and PROPHET. Experiments are performed in two distinct scenarios: alerts originating from dense urban centres and from sparse, remote mountainous regions. Our results demonstrate that the rail network provides robust connectivity for opportunistic communication in both environments thus validating the integration of DTN principles in remote scenarios.

cs.NI↗

Improving Resiliency of Vital Services in Flood-Affected Regions of Bangladesh Using Next-Generation Opportunistic DTN Edge Ad Hoc Networks

Opportunistic routing architectures offer a resilient communication paradigm in environments where conventional networks fail due to disrupted infrastructure, dynamic node mobility, and intermittent connectivity conditions that commonly arise during large-scale disasters. In Bangladesh, recurring floods severely hinder communication systems, isolating affected populations and obstructing emergency response efforts. To address these challenges, there is a growing demand for intelligent and adaptive routing solutions capable of sustaining critical communication and services without relying on fixed infrastructure. This research presents AZIZA (Adaptive Zone-based Intelligent Fully Distributed Trust-Aware Routing Protocol), a next-generation opportunistic protocol designed to improve the resiliency of critical communication and services in disaster-prone and flood-affected regions. AZIZA supports adaptive data delivery for emergency alerts, sensor readings, and inter-zone coordination by integrating (1) zone-based forwarding to optimize localized transmission, (2) trust-aware logic to bypass uncooperative or malicious nodes, and (3) context-driven decision-making based on trust metrics, residual energy, and historical delivery patterns. AZIZA operates over lightweight, infrastructure-less edge ad hoc networks comprising mobile phones, UAVs, and ground vehicles acting as decentralized service relays. Simulation results using The Opportunistic Network Environment (ONE) Simulator configured with real-world mobility traces and flood data from Bangladesh demonstrate that AZIZA significantly outperforms benchmark approaches in delivery reliability, energy efficiency, and routing resilience. As a scalable and deployable framework, AZIZA advances the use of next-generation opportunistic routing in environments where traditional systems routinely collapse.

cs.NI↗

Enhancing Evacuation Safety: Detecting Post-Nuclear Event Radiation Levels in an Urban Area

The detonation of an improvised nuclear device (IND) in an urban area would cause catastrophic damage, followed by hazardous radioactive fallout. Timely dissemination of radiation data is crucial for evacuation and casualty reduction. However, conventional communication infrastructure is likely to be severely disrupted. This study designs and builds a pseudorealistic, geospatially and temporally dynamic post-nuclear event (PNE) scenario using the Opportunistic Network Environment (ONE) simulator. It integrates radiation sensing by emergency responders, unmanned aerial vehicles (UAVs), and civilian devices as dynamic nodes within Delay-Tolerant Networks (DTNs). The performance of two DTN routing protocols, Epidemic and PRoPHET, was evaluated across multiple PNE phases. Both protocols achieve high message delivery rates, with PRoPHET exhibiting lower network overhead but higher latency. Findings demonstrate the potential of DTN-based solutions to support emergency response and evacuation safety by ensuring critical radiation data propagation despite severe infrastructure damage.

cs.NI↗