Minimum Latency Research Articles

A systematic review on WebRTC for potential applications and challenges beyond audio video streaming

AbstractVideo conferencing and live streaming are being used in various industries, such as healthcare, gaming, telecommunication, manufacturing and others. As technology progresses, the need for real-time data transmission with minimal latency has increased. Web Real-Time Communication () addresses this need effectively. WebRTC is a technology designed to provide real-time communication through web and mobile browsers. Its low latency and P2P communication capabilities make it a convenient technology for secure, efficient communication in real-time applications. This paper reviews the key features of WebRTC, discusses its strengths and weaknesses and investigates a detailed analysis of 83 existing studies. Moreover, It evaluates all use cases that can be adopted by WebRTC by examining their descriptions, problem statements, and research gaps based on literature to date. Finally, It highlights the open research directions for the emerging technologies and enhancements of WebRTC. to identify their potential applications.

Research on the Application of Terahertz Communication Technology in 5G and Future Networks

Abstract: As the limitations of 5G in spectrum scarcity and massive data transmission capability become evident, the vision of 6G technology emerges, aiming for global coverage, abundant spectrum resources, ultra-high-speed data transmission, minimal latency, and high reliability. Terahertz (THz) communication technology, with its rich spectrum resources, is seen as a key to achieving 6G objectives. This paper explores the application and challenges of THz communication in 5G and future networks, analyzes its basic theory, key technologies, channel characteristics, and modeling, and discusses its prospects and challenges in 6G and beyond. The paper provides an overview of THz communication, including its position in the electromagnetic spectrum, bandwidth potential, and advantages of short wavelengths. It discusses THz-related technologies, particularly the roles of THz electronics, photonics, and plasmonics in device development, and their contributions to 6G advancements. Future research directions are outlined, including optimizing transmission windows, applying AI in wireless communication, and the potential of 6G as a new generation of intelligent digital infrastructure. While THz communication shows great potential in 6G networks, it faces technical and economic challenges. This paper provides a solid theoretical foundation for understanding THz technology, but future work requires deeper exploration in experimental validation, technical detail research, and cost-benefit analysis.

FPGA-based low-light image enhancement using Retinex algorithm and coarse-grained reconfigurable architecture

Advancements in digital imaging and video processing are often challenged by low-light environments, leading to degraded visual quality. This affects critical sectors such as medical imaging, aerospace, and underwater exploration, where uneven lighting can compromise safety and clarity. To enhance image quality in low-light conditions using a computationally efficient system. This paper introduces an FPGA-based system utilizing the Retinex algorithm for low-light image enhancement, implemented on a Coarse-Grained Reconfigurable Architecture (CGRA). The system is designed using Verilog HDL on a Xilinx FPGA, prioritizing hardware optimization to achieve high-quality outputs with minimal latency. The system achieves a processing rate of 60 frames per second (fps) for images with a resolution of 720 × 576. Quantitative evaluations show a Peak Signal-to-Noise Ratio (PSNR) improvement to 43.18 dB, a Structural Similarity Index (SSIM) of 0.92, and a Mean Squared Error (MSE) reduction, demonstrating significant enhancements in image quality. The design also achieves a low power consumption of 0.186 W and efficient resource utilization, with only 2.2% of Slice LUTs and Slice Registers used. The FPGA-based system demonstrates significant improvements in image quality with high computational efficiency, proving beneficial for critical applications in various sectors.

Robust Routing Optimization in VANET Communication Based on Bellman–Ford Multitier Metaheuristic Algorithms

ABSTRACTIn modern cities, vehicle ad hoc networks (VANETs) hold great promise for improving safety on the roads, traffic control, and communicating. The constantly changing and intricate nature of VANETs hampers adequate routing due to a variety of factors. The work examines the use of a multitiered metaheuristic algorithm that combines Bellman–Ford routing with the Fire Hawk, Gannet Optimization, and Mayfly Optimization algorithms to find optimal routing solutions for VANETs. The suggested work addresses network topological dynamics, actual time limitations, flexibility, connectivity problems, and security issues, in addition to the different quality of service (QoS) requirements in VANETs. Bellman–Ford serves as the foundational routing layer and maintains communication, whereas Fire Hawk, Gannet Optimization, and Mayfly Optimization improve and optimize routing patterns at subsequent levels. The multitier approach aims to strike a balance between the rational and probabilistic aspects of routing so as to satisfy the minimal latency needs of applications that are safety‐critical and respond quickly to the rapid changes in VANETs. The suggested work is evaluated with a traffic density ranging from 500 to 5000 with a step size of 500, and it has been observed that for the highest traffic density of 5000, the proposed work obtains a latency of 3.88 ms, data delivery rate of 0.944, a throughput of 2.94 Mbps, and energy efficiency of 97.14%. The proposed work remains robust compared to integrating state‐of‐the‐art metaheuristic algorithms such as ECRA, HOPRP, and CM. Overall, the proposed routing techniques increase routing efficiency in VANETs and make it possible to create vehicle communication networks that are more trustworthy and safe.

Automated Virtual Machine Migration with Load Shifting Technique Avoiding Latency in Cloud Environment

Efficient management of Virtual Machines (VMs) in cloud computing is one key aspect to achieve the best utilization of resources and performance. The study is about improving Automated Virtual Machine Migration with a Load Shifting Technique in order to reduce latency through the cloud domain. This methodology dynamically distribute the load across multiple VMs by migrating their workloads to other physical hosts determined according certain resource utilization metrics available in real-time. The dynamic and unpredictable nature of cloud workloads requires a solution that is more responsive in the way it manages VMs. Live migration and pre-copy migration offer suboptimal performance in terms of time to migrate, downtime due to the prolonged copying period as heaps swell/collapse during live migrations. Utilizing sophisticated algorithms and real-time monitoring to avoid these pitfalls, our methodology allows us for proper load balancing with minimal latency. The primary elements of the proposed framework are a detailed monitoring module, an intelligent load shifting algorithm, high-performance migration method and feedback loop for continuous optimization. The monitoring module is responsible for collecting on-the-fly information about CPU, memory, network traffic and disk I/O such that can be exploited by the load shifting algorithm. The algorithm processes this data to decide which VMs should be migrated with the purpose of ensuring load balancing and decreasing latency achieved using these migrations. In order to test the proposed approach, comprehensive simulations and comparatives analyses utilizing other models were performed. This results in better latency reduction, increased resource utilization and improved overall system performance. In fact, the proposed method can handle not only current challenges in cloud computing such as load balancing and latency issued but also a scalable and effective approach for general cloud infrastructure usage. Finally, the Automated Virtual Machine Migration with Load Shifting Technique proves to be a reliable and successful solution for assisting in managing dynamic cloud workloads. Through real-time monitoring, advanced algorithms and continuous optimization this method has now been able to improve cloud services performance and reliability thereby allowing it to better serve service level agreements as well meet up with user satisfaction. We hope to adapt the algorithm and generalize it for even more advanced predictive models in future iterations, as well as evaluation of practical implementation on cloud-based deployment environments.

AI and Machine Learning Solutions for Real-Time Fault-Tolerant System Design in Critical Applications

This study presents an innovative framework for implementing real-time fault-tolerant systems using artificial intelligence (AI) and machine learning (ML) to enhance reliability and resilience in critical applications. Addressing the needs of sectors such as aerospace, healthcare, automotive, and industrial automation, the proposed system integrates fault detection, isolation, and recovery mechanisms into a multi-layered architecture. Through the use of deep learning for accurate anomaly detection and reinforcement learning for rapid fault isolation, the system achieves high fault tolerance with minimal latency. The framework leverages edge computing for real-time data processing, ensuring timely responses to faults without excessive computational demands. Results from multiple case studies demonstrate significant improvements in fault detection accuracy, isolation speed, and recovery rates, affirming the framework’s adaptability and effectiveness in high-stakes environments. These findings highlight the potential of AI-driven fault-tolerant systems to elevate operational safety and reliability standards across diverse critical industries.

A Secure and Decentralized Using Blockchain and Ring‐Based Cryptosystem

ABSTRACTThe advancement of Intelligent Transport Systems (ITSs) and the Internet of Vehicles (IoVs) introduces significant security challenges, particularly in ensuring secure communication between vehicles and vehicles (V2V), vehicles and Road Side Units (V2RSU). These vulnerabilities could potentially compromise the integrity of vehicular networks, making robust security measures essential. This paper aims to develop a secure communication framework for IoV environments that addresses these security concerns by enhancing authentication and encryption processes. The proposed approach combines the Ring‐Based Degree Truncated Polynomial Cryptosystem (RNTPC) with a Blockchain framework incorporating an enhanced proof‐of‐authentication (BEPoAuth) consensus algorithm. The approach is initiated from the registration phase and managed by a Trusted Authority (TA), ensuring secure vehicle and RSU registration. Authentication is carried out in two ways: V2V authentication through vehicle signatures and V2RSU authentication via batch verification of cluster members. Group keys are generated using RNTPC for secure communication, and blockchain technology is integrated to secure transactions within the IoV environment. The proposed methodology is evaluated using various performance metrics, demonstrating a 15%–20% improvement in throughput compared to existing techniques such as Practical Byzantine Fault Tolerance (PBFT), Secure and Highly Efficient Blockchain PBFT Consensus Algorithm (SG‐PBFT), credit‐based PBFT consensus algorithm (CPBFT) and Geographic‐PBFT (G‐PBFT). The system also exhibited minimal computational and communication latency while enhancing V2V and V2RSU communication efficiency. The RNTPC‐BEPoAuth framework offers a robust and secure solution for IoV communications, significantly outperforming existing methods in terms of throughput and efficiency. This approach provides a reliable foundation for secure ITSs, addressing critical security concerns in vehicular networks.

Implementation of A Delay-Tolerant Routing Protocol in the Network Simulator NS-3

The paper explores the implementation of the Epidemic Routing Protocol using the NS-3 simulator, focusing on its application in intermittently connected networks like Mobile Adhoc Networks (MANETs). The Epidemic Protocol maximizes message delivery by flooding the network with multiple copies, ensuring reliability even in sparse or highly mobile environments. Through NS-3 simulations, the study examines the protocol’s performance, considering factors such as network size, node mobility, and message Time-To-Live (TTL). The simulation, which models an ad-hoc wireless network using IEEE 802.11b, demonstrates that the protocol achieves a 100-percentage packet delivery ratio with minimal latency. Tools like NetAnim provide detailed insights into node interactions and message dissemination, highlighting the protocol’s effectiveness in maintaining communication under challenging conditions. The paper concludes that while the Epidemic Protocol is resource- intensive, it is highly effective for scenarios where traditional routing methods fail, offering a reliable solution for dynamic and partitioned network.

Worker Presence Monitoring in Complex Workplaces Using BLE Beacon-Assisted Multi-Hop IoT Networks Powered by ESP-NOW

The increasing adoption of Internet of Things (IoT) technologies has facilitated the creation of advanced applications in various industries, notably in complex workplaces where safety and efficiency are paramount. This paper addresses the challenge of monitoring worker presence in vast workplaces such as shipyards, large factories, warehouses, and other construction sites due to a lack of traditional network infrastructure. In this context, we developed a novel system integrating Bluetooth Low Energy (BLE) beacons with multi-hop IoT networks by using the ESP-NOW communications protocol, first introduced by Espressif Systems in 2017 as part of its ESP8266 and ESP32 platforms. ESP-NOW is designed for peer-to-peer communication between devices without the need for a WiFi router, making it ideal for environments where traditional network infrastructure is limited or nonexistent. By leveraging the BLE beacons, the system provides real-time presence data of workers to enhance safety protocols. ESP-NOW, a low-power communications protocol, enables efficient, low-latency communication across extended ranges, making it suitable for complex environments. Utilizing ESP-NOW, the multi-hop IoT network architecture ensures extensive coverage by deploying multiple relay nodes to transmit data across large areas without Internet connectivity, effectively overcoming the spatial challenges of complex workplaces. In addition, the Message Queuing Telemetry Transport (MQTT) protocol is used for robust and efficient data transmission, connecting edge devices to a central Node-RED server for real-time remote monitoring. Moreover, experimental results demonstrate the system’s ability to maintain robust communication with minimal latency and zero packet loss, enhancing worker safety and operational efficiency in large, complex environments. Furthermore, the developed system enhances worker safety by enabling immediate identification during emergencies and by proactively identifying hazardous situations to prevent accidents.

Scalable watermarking for identifying large language model outputs

Large language models (LLMs) have enabled the generation of high-quality synthetic text, often indistinguishable from human-written content, at a scale that can markedly affect the nature of the information ecosystem1–3. Watermarking can help identify synthetic text and limit accidental or deliberate misuse4, but has not been adopted in production systems owing to stringent quality, detectability and computational efficiency requirements. Here we describe SynthID-Text, a production-ready text watermarking scheme that preserves text quality and enables high detection accuracy, with minimal latency overhead. SynthID-Text does not affect LLM training and modifies only the sampling procedure; watermark detection is computationally efficient, without using the underlying LLM. To enable watermarking at scale, we develop an algorithm integrating watermarking with speculative sampling, an efficiency technique frequently used in production systems5. Evaluations across multiple LLMs empirically show that SynthID-Text provides improved detectability over comparable methods, and standard benchmarks and human side-by-side ratings indicate no change in LLM capabilities. To demonstrate the feasibility of watermarking in large-scale-production systems, we conducted a live experiment that assessed feedback from nearly 20 million Gemini6 responses, again confirming the preservation of text quality. We hope that the availability of SynthID-Text7 will facilitate further development of watermarking and responsible use of LLM systems.

Performance Evaluation of mmWave Communication for Real-Time UAV-Base Station Links

Introduction: In the context of modern communication networks, the increasing demand for high data rates, minimal latency, and improved reliability—especially in scenarios involving Unmanned Aerial Vehicles (UAVs)—calls for innovative solutions. Millimeter-wave (mmWave) communication emerges as a fitting choice due to its capacity to offer ample bandwidth and high data rates. Methods: This research paper presents a comprehensive study that compares the performance of mmWave communication links across four different frequencies (6, 28, 60, and 120 GHz) for UAV-to-Base Station (BS) communication. The investigation, conducted through simulations, adheres to atmospheric conditions such as rain and fog, different environments (urban and rural) scenarios. It evaluates the suitability of mmWave frequencies using critical performance metrics: latency, received power, Bit Error Rate (BER) and Signal-to-Interference plus Noise Ratio (SINR). These metrics play a fundamental role in assessing communication link quality. Results: Our findings reveal that leveraging higher mmWave frequencies—specifically 60 and 120 GHz—significantly enhances UAV-to-BS communication performance. Compared to lower frequencies, these higher bands exhibit reduced latency, higher data rates, and improved throughput. Consequently, mmWave frequencies beyond 60 GHz prove well-suited for UAV communication, facilitating efficient data exchange with minimal delay. Conclusion: Furthermore, the potential for reduced interference and enhanced reliability at these elevated frequencies makes them particularly advantageous in applications requiring seamless connectivity. Examples include UAV-assisted emergency response, surveillance, and remote sensing. These insights hold substantial implications for the design and deployment of UAV communication systems, highlighting mmWave technology’s transformative potential in the emerging era of aerial communication.

Wearable Online Freezing of Gait Detection and Cueing System.

This paper presents a real-time wearable system designed to assist Parkinson's disease patients experiencing freezing of gait episodes. The system utilizes advanced machine learning models, including convolutional and recurrent neural networks, enhanced with past sample data preprocessing to achieve high accuracy, efficiency, and robustness. By continuously monitoring gait patterns, the system provides timely interventions, improving mobility and reducing the impact of freezing episodes. This paper explores the implementation of a CNN+RNN+PS machine learning model on a microcontroller-based device. The device operates at a real-time processing rate of 40 Hz and is deployed in practical settings to provide 'on demand' vibratory stimulation to patients. This paper examines the system's ability to operate with minimal latency, achieving an average detection delay of just 261 milliseconds and a freezing of gait detection accuracy of 95.1%. While patients received on-demand stimulation, the system's effectiveness was assessed by decreasing the average duration of freezing of gait episodes by 45%. These preliminarily results underscore the potential of personalized, real-time feedback systems in enhancing the quality of life and rehabilitation outcomes for patients with movement disorders.

Comparative exploration on EEG signal filtering using window control methods

Electroencephalography (EEG) is used to monitor brain activity. The brain signals consist of different frequency band signals delta, theta, alpha, beta, and gamma waves. The signals are affected by external noise which reduces the quality of the EEG signal due to which it becomes difficult to do further processing of EEG signals like feature extraction or extraction of meaningful features from EEG signal. Therefore, it becomes important to filter the noise from the EEG signal before feature extraction or classification of the EEG signal. The research article presents an overview of different types of windowing filter techniques like Rectangular, Bartlett, Hamming, Hanning, and Kaiser windows applied for finite impulse response (FIR) behavior which is used for EEG signal processing for different brain waves processed in different frequency bands. The comparative analysis is carried out in terms of the response time of brain frequency bands for different windowing filter techniques using the MATLAB 2023 signal processing simulation tool. The novelty of the work lies in estimating minimum latency and appropriate filter selection for various typical EEG waves, since the EEG signals are pre-supposed in the hardware chip design, noise elimination is the first step in high-performance computing applications. The Bartlett window band stop has an optimal response time of 12.666 s for delta waves, a highpass filter with a response time of 16.187 s for theta waves, a bandpass with a response time of 13.122 s for alpha waves, a highpass filter with a response time of 17.866 s for beta waves, and a highpass filter with a response time of 13.797 s for gamma waves. The Barlett window FIR filter is well-suited for EEG applications.

Secured Cyber-Internet Security in Intrusion Detection with Machine Learning Techniques

The rapid proliferation of Internet-connected devices has elevated the significance of cybersecurity, making intrusion detection a critical aspect of maintaining network integrity. Traditional security measures often fail to provide adequate protection against sophisticated attacks, necessitating advanced and robust solutions. This paper introduces a comprehensive cyber-internet security framework that leverages machine learning techniques for real-time intrusion detection and prevention. The proposed methodology employs a hybrid approach, integrating supervised and unsupervised learning models to detect anomalies and classify intrusions effectively. Specifically, a combination of Support Vector Machine (SVM), Decision Trees (DT), and K-means clustering is used to enhance detection accuracy and reduce false-positive rates. The experimental results demonstrate that the proposed model achieved a detection accuracy of 97.8%, a precision of 96.5%, and a recall of 95.2% on the NSL-KDD dataset. The implementation also reduced the false-positive rate to 1.2% and the computational overhead by 15% compared to traditional detection systems. Additionally, the proposed system was tested on real-time traffic data, where it successfully identified and mitigated various cyber threats, including Distributed Denial of Service (DDoS) attacks and network infiltrations, with minimal latency and high reliability. In conclusion, the study presents an efficient and secured cyber-internet security framework that significantly enhances intrusion detection capabilities using machine learning techniques. The proposed system provides a scalable and adaptive solution for securing critical infrastructure and networks against evolving cyber threats, making it an ideal candidate for deployment in real-world cybersecurity applications.

Fish feeding behavior quantification and automatic acoustic event detection: algorithmic optimization with deep learning

In recent years, the integration of sensor processing technologies with intelligent feeding control systems has become crucial for enhancing recirculating aquaculture systems (RAS). Intelligent feeding control significantly improves profitability and animal welfare in aquaculture by accurately monitoring feed intake and reducing waste. Acoustic methods have proven effective in automating feed monitoring, particularly in assessing feeding intensity. Automatic Audio Event Detection (AED) holds promise for identifying sound events associated with feeding activities, enabling precise monitoring of feeding behavior and estimation of feed consumption through frequency-based feeding signatures. Advancements in sensor technologies and computational capabilities enable real-time collection, processing, and analysis of acoustic data. However, handling high volumes of data with minimal latency and high accuracy poses challenges. To address these challenges, we propose a systematic approach: (1) Data preprocessing using wavelet transforms enhances efficiency and reduces computational overhead. (2) Data pruning focuses on informative features and reduces input data dimensionality, improving classification accuracy. (3) We adopt a weakly-labeled semi-supervised learning model with Convolutional Neural Networks (CNNs) to detect feeding events, improving classification performance. (4) Ensemble learning combines multiple classifiers to mitigate overfitting and enhance model generalization. Our methodology shows promise in accurately quantifying fish feeding behavior using acoustic signals.

Gauss Markov and Flow Balanced Vector Radial Learning network traffic classification on IoT with SDN.

Recent evolution in connected devices modelled a massive stipulation for network traffic resources and classification. Software-defined networking (SDN) enables ML techniques with the Internet of Things (IoT) to automate network traffic. This helps to reduce accuracy and improves latency. Problems by conventional techniques to categorize network traffic acquired from IoT and assign resources can be resolved through SDN solutions. This manuscript proposes a proposed network traffic classification technique on IoT with SDN called Gauss Markov and Flow-balanced Vector Radial Learning (GM-FVRL). With the network traffic features acquired from the IoT devices, SDN-enabled Gauss Markov Correlation-based IoT Network Traffic Feature Extraction is applied to extort relevant network aspects. Next, the flow-balanced radial-based ML model for network traffic categorization uses the relevant extracted network traffic features. With the aid of flow, the balanced radial basis function reduces the influence of noise due to distinct network flow. This helps to improve accuracy and minimize latency. Due to this, better precision and recall is ensured. Performance of our method has been evaluated utilizing a scheme using an SDN traffic dataset. The results show that our method classifies the network traffic with high classification accuracy and minimum latency, ensuring better precision and recall.

Neural decoding and feature selection methods for closed-loop control of avoidance behavior

Objective.Many psychiatric disorders involve excessive avoidant or defensive behavior, such as avoidance in anxiety and trauma disorders or defensive rituals in obsessive-compulsive disorders. Developing algorithms to predict these behaviors from local field potentials (LFPs) could serve as the foundational technology for closed-loop control of such disorders. A significant challenge is identifying the LFP features that encode these defensive behaviors.Approach.We analyzed LFP signals from the infralimbic cortex and basolateral amygdala of rats undergoing tone-shock conditioning and extinction, standard for investigating defensive behaviors. We utilized a comprehensive set of neuro-markers across spectral, temporal, and connectivity domains, employing SHapley Additive exPlanations for feature importance evaluation within Light Gradient-Boosting Machine models. Our goal was to decode three commonly studied avoidance/defensive behaviors: freezing, bar-press suppression, and motion (accelerometry), examining the impact of different features on decoding performance.Main results.Band power and band power ratio between channels emerged as optimal features across sessions. High-gamma (80-150 Hz) power, power ratios, and inter-regional correlations were more informative than other bands that are more classically linked to defensive behaviors. Focusing on highly informative features enhanced performance. Across 4 recording sessions with 16 subjects, we achieved an average coefficient of determination of 0.5357 and 0.3476, and Pearson correlation coefficients of 0.7579 and 0.6092 for accelerometry jerk and bar press rate, respectively. Utilizing only the most informative features revealed differential encoding between accelerometry and bar press rate, with the former primarily through local spectral power and the latter via inter-regional connectivity. Our methodology demonstrated remarkably low training/inference time and memory usage, requiring<310 ms for training,<0.051 ms for inference, and 16.6 kB of memory, using a single core of AMD Ryzen Threadripper PRO 5995WX CPU.Significance.Our results demonstrate the feasibility of accurately decoding defensive behaviors with minimal latency, using LFP features from neural circuits strongly linked to these behaviors. This methodology holds promise for real-time decoding to identify physiological targets in closed-loop psychiatric neuromodulation.

End-to-End Resource Allocation Management Model in Next-Generation Network: Survey

Network communication has grown rapidly with massive demands of services. Moreover, resource allocation in networking is a fundamental and crucial issue that cannot ensure the network's stability and efficiency with the myriad requirements of different services. Various vertical businesses may seek varied network services, particularly in the Fifth-Generation networks and Beyond (5G+). The pros of Fifth-Generation communication networks are to outperform 4G in performance by having higher bandwidth, minimum latency, more capacity, and QoS (Quality of Service). Software-Defined Network (SDN) and Network Function Virtualization (NFV) are two technologies that are combined in the next generation cellular network to provide improved network management. The primary idea behind resource allocation (RA) in the next generation network is the concept of network slicing where the network resources are virtually partitioning into many separate networks. Each separated network must satisfy the unique needs of the required service to achieve the required QoS. In this survey, we focus on resource management issues related to network slicing and tackling the biggest obstacles in this field while offering a thorough and up-to-date overview of this field. Thus, thorough analysis of the allocation of resources on the access side and core side of the network communication was sought. Also, demonstrates how revolutionary techniques that are used to support the management of sliced networks which are based on Machine Learning (ML) and Artificial Intelligence (AI). Importantly, use appropriate ML techniques such as deep learning for predicting the network condition and Reinforcement learning to learn optimal allocation policy without depending on prior knowledge and other techniques such as classification and clustering to aggregate the similar needs of users into separate slices. This could help to enhance resource utilization by allocating a sufficient amount of resource as needed based on ML algorithms and optimal utilization of resources and reducing operational costs by real-time adjustment of it based on user demands and network conditions

Night time Surveillance in Communication Using YOLOv3

Nighttime surveillance presents significant challenges due to low visibility, varying lighting conditions, and interference from artificial light sources. Hence, this study proposed an effective object detection framework using the YOLOv3 model to enhance real-time monitoring in night surveillance applications, specifically within communication and security systems. YOLOv3's architecture, with its multi-scale detection and use of predefined anchor boxes, enables robust detection of objects under low-light environments and amidst light interference from vehicles and streetlights. The proposed system is tested on a night surveillance dataset, where it demonstrates high precision and speed in identifying objects, making it suitable for real-time applications. With Mean Average Precision (mAP) 87.9%, YOLOv3 effectively balances detection accuracy with inference time, ensuring minimal latency in live surveillance feeds. The results indicate that YOLOv3 outperforms traditional models such as Faster RCNN, particularly in detecting small objects under poor illumination. This approach offers a reliable solution for enhancing communication and security systems in nighttime surveillance scenarios.

Self-aware collaborative edge inference with embedded devices for IIoT

Edge inference and other compute-intensive industrial Internet of Things (IIoT) applications suffer from a bad quality of experience due to the limited and heterogeneous computing and communication resources of embedded devices. To tackle these issues, we propose a model partitioning-based self-aware collaborative edge inference framework. Specifically, the device can adaptively adjust the local model inference scheme by sensing the available computing and communication resources of surrounding devices. When the inference latency requirement cannot be met by local computation, the model should be partitioned for collaborative computation on other devices to improve the inference efficiency. Furthermore, for two typical IIoT scenarios, i.e., bursting and stacking tasks, the latency-aware and throughput-aware collaborative inference algorithms are designed, respectively. Via jointly optimizing the partition layer and collaborative device selection, the optimal inference efficiency, characterized by minimum inference latency and maximum inference throughput, can be obtained. Finally, the performance of our proposal is validated through extensive simulations and tests conducted on 10 Raspberry Pi 4Bs using popular models. Specifically, in the case of two collaborative devices, our platform reaches up to 92.59% latency reduction for bursting tasks and 16.19× throughput growth for stacking tasks. In addition, the divergence between simulations and tests ranges from 1.64% to 9.56% for bursting tasks and from 3.24% to 11.24% for stacking tasks, which indicates that the theoretical performance analyses are solid. For the general case where the data privacy is not considered and the number of collaborative devices is optimally determined, up to 14.76× throughput speed up and 84.04% latency reduction can be obtained.