Abstract Video surveillance has benefited greatly from advances in artificial intelligence, particularly in computer vision, and the number of monitored environments has consequently increased, creating new challenges in extracting relevant information. When multiple cameras cover adjacent areas, overlapping fields of view can cause the same subject to be detected across cameras, introducing duplicate counts. Although the literature offers established solutions, many rely on high-quality recordings and complex, computationally intensive methods, limiting their use on resource-constrained devices. In this work, we address these challenges with a solution that leverages minimal information about target subjects and uses a transductive strategy to detect duplicates based on interactions within overlapping fields of view. Experiments in real-world settings show that our approach suppresses duplicates effectively, making it suitable for deployment on resource-limited hardware. We evaluate state-of-the-art lightweight models with high inference speed, as well as classical re-identification methods, in scenarios with low-quality video and constrained devices. The results underscore the effectiveness of the proposed approach and motivate exploration of re-ID with domain adaptation, as well as anchor-free methods with weak or semi-supervised learning.

Due to limited hardware, consumer-grade surveillance cameras usually rely on cloud-based computer vision models to detect people in video footage. However, this approach introduces a recurring cost, as users must continuously pay for cloud processing. One possible solution is to use mobile devices for person detection, as they can be found in most households. Yet, experimental evaluations on the impact of different computer vision models on mobile device resource usage are limited. This study examines the efficiency of YOLOv8 models on Android devices, assessing detection performance, inference time, memory consumption, and energy efficiency. The models were tested using two machine learning frameworks, LiteRT (formerly TensorFlow Lite) and ONNX Runtime, to determine the most suitable approach for mobile inference. Experimental results confirm that compact models, such as YOLOv8n and YOLOv8s, offer the best trade-off between computational efficiency and detection accuracy, while LiteRT outperforms ONNX in all evaluated metrics.

Smart homes are transforming domestic environments by integrating connected devices and sensors, enabling lighting, temperature, and security automation. While these systems enhance comfort and efficiency, they often rely on predefined settings or manual input due to the absence of adaptive recommendation systems. AI-driven recommendation systems personalize actions by learning from user behavior and environmental data, improving the smart home experience. However, they also introduce cybersecurity risks, particularly data poisoning attacks, where manipulated data disrupts system functionality. This paper exposes and examines vulnerabilities in smart home recommendation systems, categorizing data poisoning attacks and analyzing their impact. Through a literature review and attack vector analysis, we identify key weaknesses and propose mitigation strategies to enhance security. Our goal is to contribute to developing robust smart home technologies that protect user privacy, ensure reliability, and withstand adversarial threats.

ABSTRACT Context Smart home devices have become increasingly popular in modern households, powered by the Internet of Things (IoT) advances. The data generated by smart devices can provide valuable insights into users' behavior and preferences. By analyzing the data, one can understand how people interact with their homes, thus creating a “smart home profile”. To comprehend the complete IoT ecosystem dynamics of an intelligent environment, it is necessary to learn from each IoT device to predict its status in the future time. Nevertheless, dealing with real‐world IoT data structure requires considerable preprocessing tasks and the employment of classifiers that can learn multiple IoT inputs from a single IoT message. Objective Aware of these challenges, this paper proposes a novel methodology to process multi‐label IoT data and provide a comprehensive comparison of multi‐label classifiers for forecasting the status of smart devices, considering their efficiency and accuracy. Method We propose a data transformation method to preprocess the IoT data to be used by multi‐label classifiers. This method is based on real data structure. Results We evaluate our proposal in two real‐world scenarios and various multi‐label classifiers. The promising findings indicate that efficient classifiers can generate many correct predictions for a comprehensive IoT ecosystem in a small fraction of a second. Conclusions Our proposed data transformation can fit the context of prediction to smart homes and work with multi‐label classifiers to understand user behavior.

Network operators must deal with Classes of Service (CoS), which have several quality requirements, such as latency, bandwidth/capacity, priority, etc. Besides, it is observed an increase in the volume of traffic that is offered to the transport network. This traffic can be affected by network natural disasters or human-made attacks. In this case, network operators must decide which services to restore, considering their different requirements. In this work, we present a Lightpath Selection Algorithm (LSA) that aims to select lightpaths to be restored after a resource crunch. This algorithm has a multicriteria decision approach considering CoS, Bandwidth, number of Hops, and Holding time. Moreover, service degradation is also considered for those lightpaths that can not be restored with full bandwidth. Results show that our proposed algorithm can improve network restorability and availability without penalizing low-level CoSs.

The baseband processing centralization enabled by the Cloud Radio Access Network generates stringent latency and high bandwidth requirements. Therefore, some studies proposed using hybrid architectures and baseband functional splitting to ease such requirements. In the literature, the ideal functional splitting in hybrid RAN architectures has been tackled using integer linear programming. Such approaches guarantee optimality, but they have low scalability, making them infeasible for real deployments. On the other hand, Meta Heuristics can provide practical solutions to large combinatorial problems with a good level of accuracy (not rarely achieving optimality). This paper proposes new modeling of the functional splitting problem in Cloud Fog RANs using the meta-heuristic optimization named Soccer Game Optimization. We compared our solution to an integer linear programming formulation evaluating the correctness, energy efficiency, and network coverage. Results show that meta-heuristic achieve statistically optimality equal to the ILP in coverage and energy-efficiency.

Optical networks can support service degradation by providing bandwidth lower than that required to adapt the network provisioning when optical resources are insufficient. This paper proposes a service-degradation algorithm that is aware of application characteristics in Elastic Optical Networks (EONs). The algorithm considers a proportional Quality-of-Service (QoS) model and cross-layer information to decide which lightpath to be degraded, and it aims to reduce the impact of resource unavailability on delay and bandwidth sensitive applications. Results show that the proposed strategy can decrease blocking probability by up to 93% and reduce the number of applications penalized by service degradation by 100% compared to other approaches unaware of application characteristics.

Cloud-Fog Radio Access Networks (CF-RAN) were proposed as an alternative network architecture to alleviate the high fronthaul capacity requested in traditional Cloud RAN (CRAN) by moving some BaseBand Units (BBUs) from the cloud nodes to fog nodes closer to users. However, when BBU processing is moved into fog nodes, OPEX and CAPEX will increase, and the cost and energy savings introduced by CRAN will also reduce. Moreover, mobile traffic fluctuations may lead to an unbalanced resource utilization and energy- inefficient operation in fog nodes. To address this problem, processing functions in fog nodes could be activated and deactivated in function of network traffic and BBUs placed on fog nodes could be migrated to cloud nodes when network traffic is low. In this paper, we propose an Integer Linear Programming (ILP) formulation to address this dynamic resource allocation problem. By means of Network Functions Virtualization (NFV), virtualized BBUs (vBBUs) can be dynamically allocated and deallocated in fog nodes. Furthermore, considering the availability of cloud nodes and the optical fronthaul, vBBUs can be migrated from fog nodes to cloud nodes in order to balance processing loads and save energy. Compared to a baseline incremental algorithm without vBBU migration, our proposal reduces blocking probability in 89% and achieves power savings of 38%, while providing a very small rate of service interruption due to vBBUs migration.

Mobile fronthaul networks can benefit from TWDM-PON systems, such as NG-EPON, by employing an efficient DWBA scheme to fulfill CPRI requirements. This paper introduces a prediction-based DWBA algorithm to meet the bandwidth and delay requirements of the CPRI traffic regardless of the used line bit rate. The central idea is to provide for each ONU multiple time slots in one or more wavelengths. Moreover, a prediction method is employed to determine the number of time slots required for each ONU in future cycles. To the best of our knowledge, this is the first NG-EPON's DWBA algorithm focused in C-RAN architectures. Numerical results show that the proposed strategy decreases the number of active wavelengths, while improves time slot utilization and throughput rate without violating the fronthaul delay requirements, when compared with the state-of-the-art algorithms.

Hybrid Cloud-Fog Radio Access Network (CF-RAN) is a recent network architecture proposed to increase network coverage from CRAN while leveraging power consumption in future 5G networks. In CF-RAN, the processing of baseband signals from Remote Radio-Heads (RRHs) can be performed in virtualized BaseBand Units (vBBUs) located in the cloud or in fog nodes that are instantiated in function of the network demand. Through a Time-and-Wavelength Division Multiplexing Passive Optical Network (TWDM-PON), virtualized PONs (VPONs) can be dynamically created to support transmissions from RRHs to vBBUs. However, due to traffic fluctuations, the amount of necessary vBBUs and VPONs may change along a day. In this paper, we propose a batch scheduling algorithm based on Integer Linear Programming (ILP) to perform reconfiguration of VPONs and migration of vBBUs among processing nodes in function of fluctuation on traffic demands. Our results show that, in comparison to an incremental algorithm without reconfiguration of VPONs and vBBUs migration capacities, our algorithm reduces power consumption and bandwidth wastage by up to 28% and 57%, respectively, and also eliminates blocking probability.

In this paper, we propose a new restoration algorithm for disaster recovery in Elastic Optical Networks (EONs). The algorithm considers the different requirements on restoration delay and bandwidth degradation while restoring over multiple routes. Furthermore, the algorithm is able to adapt the connection recover process to the post-disaster network conditions and, consequently, increase the number of restored connections. Results indicate that the proposed strategy is able to improve connections restorability and decrease bandwidth blocking probability when compared to restoration approach based on bandwidth degradation and multipath recovery.

Elastic Optical Networks (EONs) represent a new approach for dealing with the enormous traffic demand in core networks as they can offer bandwidth granularities closer to those requested by the user and hence improve spectral utilization. In current literature there is a lack of dynamic strategies for service degradation which is a possible measure to address problems related to network congestion and consists in reducing the amount of resources provided. Since services of different classes can be requested, we propose in this paper an online strategy for service degradation using proportional Quality of Service (QoS). Our proposed strategy aims at minimizing the number of blocked requests due to lack of resources while provides throughput and delay guarantees for provisioned lightpaths. Thus, in order to quantify the impact of the degradation on the lightpaths we modeled source-destination pairs in an EON as a queuing system working under the Generalized Processor Sharing (GPS) service discipline with admission control of Leaky Bucket policy. The obtained results show that the proposed algorithm can reduce the blocking probability and give network operators more control between different degraded service classes.

Optical networks are commonly adopted to implement the fronthaul of Cloud-RAN (CRAN) and Hybrid-CRAN (H-CRAN) architectures to support the 5G traffic. The Cloud-Fog RAN (CF-RAN) architecture promotes cloud and local fog processing of baseband signals with the novelty of activating on demand processing capabilities under the Fog Computing and Network Functions Virtualization (NFV) paradigms. Timeand-Wavelength-Division-Multiplexed Passive Optical Networks (TWDM-PON) are used to implement the fronthaul due to the capacity of Virtual-PON (VPON) formation, which can reduce power consumption by allocating as many Remote Radio Heads (RRH) as possible into a single optical channel. However, the TWDM-PON wavelengths must be assigned to the VPONs in an exclusive manner so different VPONs do not share a common wavelength and thus VPONs do not collide on the optical links. In this paper we propose an Energy-Efficient VPON Formation and Wavelength Dimensioning formulation through Integer Linear Programming (ILP) to efficiently form VPONs and dimension the wavelengths so collisions never occurs. Results from the experiments show that the ILP formulation was able to form VPONs and dimension the wavelengths to the network nodes on the most energy-efficient way, achieving gains in the order of 80.27% in comparison to a random VPON formation approach.

The Internet of Things (IoT) allows for the communication of a large number of physical or virtual objects through different technologies, protocols and patterns. Several organizations have tried to predict the number of IoT devices that will be connected to the Internet by 2020. Although they have not yet to agree on an exact `magical' number of such devices (in billions), we can expect this huge number of devices to be active which may pose several challenges regarding connectivity and information exchange over network infrastructures. And despite the fact that Cloud and Fog computing addresses some of these challenges, it can also create other problems due to the high demand for storage as well as high network traffic at the edge level. In this paper, we model the IoT data traffic in order to execute several experiments related to the demand for storage and network traffic. As a consequence, the results of our experiments, which have been validated in a modeling based on the Fog of Things paradigm, can be used to evaluate the impact that variation in IoT data traffic patterns has on IoT/Fog environments.

The use of an optical fiber infrastructure to accommodate cloud services is gaining a lot of momentum. This is mainly driven by the bandwidth and latency performance that optical transmission can guarantee. On the other hand failures in the optical infrastructure may result in the concurrent loss of a possibly high number of cloud services. For this reason, being able to offer cloud service resiliency at a contained cost is of the utmost importance for operators. This paper proposes a heuristic for the restoration of optical cloud services in the presence of a single fiber link failure. The heuristic leverages on two parameters specified in the service class (i.e., restoration delay and bandwidth degradation) in order to make the best use of the available optical resources during the recovery process. The numerical results presented in the paper show that the proposed restoration algorithm is able to improve cloud service restorability without a negative impact on the cloud service blocking probability.

In the context of future Cloud Radio Access Networks (CRAN), optical networks will play an important role to provide the required transport capacity between cell-sites and processing pools, especially for future 5G scenarios. For instance, using CPRI fronthaul technologies a single antenna element can generate data up to 24.3Gbps even with current configurations of radio transmissions, and it is expected to generate up to Tbps with the advance of technology. So, the transport segment of a 5G network needs to be accurately planned to accommodate all the generated traffic. In this work, we propose the use of a Passive Optical Network (PON) jointly with the emergent paradigms of Fog Computing and Network Function Virtualization (NFV) to energy-efficiently support the high traffic transported in emergent mobile networks in an hybrid architecture called Cloud/Fog RAN (CF-RAN) that allows local and remote baseband processing. We introduce an Integer Linear Programming (ILP) model to schedule the processing of CPRI demands among the processing nodes of the network and turn on or off processing functions on demand. Our approach is able to accommodate demands on the nodes of the network in the most energy efficient way. We compare our results with CRAN and distributed architectures (DRAN) and show that an energy efficient planning can achieve considerable gains in power consumption.

Today's communication networks require special redundancy to overcome severe multiple element failures. These severe failures can isolate entire sub-components of terrestrial networks - e.g., possible aftermath of natural disasters and threats such as a High-Altitude Electromagnetic Pulse (HEMP). An integrated use of all communication systems available is important to help lessen the impact over distressed areas. This work investigates the use of aerial platforms with well-defined trajectories (such as LEO satellites) to evacuate data from systems within the affected regions. We propose an algorithm capable of generating an evacuation plan for data located in terrestrial isolated systems, such as Data Centers, through the satellite network, towards final destinations in the main network. Our method works whether the satellite network is damaged or not. The evacuation plan is a node-to-node transmission schedule that maximizes the amount of evacuated data. Post-disaster scenarios are used to analyze how our method performs under different impact sizes and satellite network configurations. The results show that the proposed algorithm produces node-to-node transmission schedules that maximize evacuated data while maintaining fairness among disconnected components.

Degraded provisioning provides an effective solution to flexibly allocate resources in various dimensions to reduce blocking for differentiated demands when network congestion occurs. In this work, we investigate the novel problem of online degraded provisioning in service-differentiated multi-layer networks with optical elasticity. Quality of Service (QoS) is assured by service-holding-time prolongation and immediate access as soon as the service arrives without set-up delay. We decompose the problem into degraded routing and degraded resource allocation stages, and design polynomial-time algorithms with the enhanced multi-layer architecture to exploit network flexibility in temporal and spectral dimensions. Numerical results verify that we can achieve significant blocking reduction, especially for requests with higher priorities. They also indicate that degradation in optical layer can increase the network capacity, while degradation in electric layer provides flexible time-bandwidth exchange.

In emerging cellular systems, optical fronthaul is expected to play a major role to support many control operations, e.g., Coordinated Multipoint (CoMP). CoMP is a promising technique for interference mitigation as it can transform interfing signals into joint transmission (reception) in which signals from adjacent cell sites are simultaneously transmitted (received) to (from) mobile terminals. But the exchange of information required by CoMP demands high flexibility and capacity. This paper proposes a new architecture for supporting CoMP operations in emerging cellular systems. It is based on a time-and-wavelength-division-multiplexed passive optical network (TWDM-PON) fronthaul, using virtualized base stations and a cloud radio access network (C-RAN) architecture. We also propose techniques to distribute the load on controllers to minimize the coordination delay. Results show that, for a typical setting, our methods can save up to 37% on the time required to distribute channel state information among multiple base stations.

This paper introduces a novel traffic grooming algorithm for the connection establishment of deadline-driven requests. The algorithm grooms batches of requests rather than individual requests. Results indicate that the algorithm is able to significantly reduce the blocking ratio and promote a fair distribution of blocking among source destination pairs of nodes.

The combination of service-orientation and software product line engineering, called Service-Oriented Product Line Engineering (SOPLE) have received attention by researchers and practitioners in the last years, and these areas can address issues of each other. One service-orientation issue is service identification. It consists of determining candidate services to a service-oriented environment based on pre-existing software artifacts, e.g., business process, source code, and so on. In order to provide a systematic identification of services, there are many available service identification methods in the literature, regarding different understanding of services, goals, and techniques. Due to this heterogeneity, this paper presents an in-depth comparison of service identification methods as well as a recommendation of the most suitable ones in the SOPLE context. This work can help the decision making of the most suitable method according to stakeholders' needs.

This paper introduces a batch scheduling algorithm for OBS networks which has linear computational complexity. A transformation of the problem formulation is applied to allow the modeling of the problem as a job scheduling with identical machine type of problem. The proposed algorithm was compared to several other heuristic. Results show that it produces a lower bound to the blocking probability of those given by heuristics and it is only 3% slower than the fastest one.

This paper introduces an optimal batch scheduling algorithm for the scheduling of batches of bursts in optical burst switching networks. The algorithm, called BATCHOPT, considers both the requests being processed in the current batch and the requests previously scheduled in the search for an optimal solution. Moreover, an extended version of the JET reservation protocol is proposed for efficiently handling batches of bursts. Results obtained via simulation show that the BATCHOPT algorithm produces good performance when compared to other proposed existing algorithms.

This paper presents a novel channel scheduling discipline for optical burst switching networks called Least Reusable Channel (LRC). LRC decides on which void to allocate an incoming burst based on the potentiality of reusing the voids created. These decisions consider the routes as well as the network load. It is shown via numerical examples that the LRC produces lower blocking probability and distributes more uniformly the losses among routes than existing scheduling disciplines.

Multiscaling traffic in optical burst switching networks yields to considerably higher loss rates than monoscale traffic. The assembly policy at the edge of an OBS network can change the multiscaling characteristic of ingress traffic to monoscaling. The difficulty, however, is the identification of the cut-off scale of multiscaling traffic so that ingress traffic can be aggregated on scales larger than the cut-off one. This paper introduces a method for identifying the cut-off scales as well as proposes and evaluates different assembling policies.

Ingress nodes of Optical Burst Switching (OBS) networks are responsible for assembling burst out of incoming packets and to forward these bursts into the OBS network. The most common criteria for burst assembly are based either on time or on traffic volume. The change of the statistical characteristics of a traffic stream at an ingress switch can affect the quality of service provisioning. Therefore, the statistical characteristics of the output flow of an ingress node need to be known for appropriate network dimensioning and service provisioning. This paper evaluates the impact of burst assembly mechanisms on the scaling properties of multifractal traffic flows. It is showed that the most relevant factor that determines the nature of the output traffic flow is the relationship between the cut-off time scale of the input traffic and the time scale of assembly threshold.

Minimum interference routing is instrumental to MPLS traffic engineering under realistic assumptions of unknown traffic demand. This work presents a new algorithm for minimum interference routing, called light minimum interference routing (LMIR). This algorithm introduces a new approach for critical link identification that reduces the computational complexity. Results, derived via simulation, show that LMIR is precise and has indeed a low computational complexity.