Malware detection is increasingly challenged by evolving techniques like obfuscation and polymorphism, limiting the effectiveness of traditional methods. Meanwhile, the widespread adoption of software containers has introduced new security challenges, including the growing threat of malicious software injection, where a container, once compromised, can serve as entry point for further cyberattacks. In this work, we address these security issues by introducing a method to identify compromised containers through machine learning analysis of their file systems. We cast the entire software containers into large RGB images via their tarball representations, and propose to use established Convolutional Neural Network architectures on a streaming, patchbased manner. To support our experiments, we release the COSOCO dataset-the first of its kind-containing 3364 largescale RGB images of benign and compromised software containers at https://huggingface.co/datasets/k3ylabs/cosoco-imagedataset. Our method detects more malware and achieves higher F1 and Recall scores than all individual and ensembles of VirusTotal engines, demonstrating its effectiveness and setting a new standard for identifying malware-compromised software containers.

Autonomous Vehicles (AVs) rely on real-time processing of natural images and videos for scene understanding and safety assurance through proactive object detection. Traditional methods have primarily focused on 2D object detection, limiting their spatial understanding. This study introduces a novel approach by leveraging 3D object detection in conjunction with augmented reality (AR) ecosystems for enhanced real-time scene analysis. Our approach pioneers the integration of a synthetic dataset, designed to simulate various environmental, lighting, and spatiotemporal conditions, to train and evaluate an AI model capable of deducing 3D bounding boxes. This dataset, with its diverse weather conditions and varying camera settings, allows us to explore detection performance in highly challenging scenarios. The proposed method also significantly improves processing times while maintaining accuracy, offering competitive results in conditions previously considered difficult for object recognition. The combination of 3D detection within the AR framework and the use of synthetic data to tackle environmental complexity marks a notable contribution to the field of AV scene analysis. • A multimodal architecture for real-time 3D object detection in AV systems. • Efficient 3D bounding box prediction extrapolated from 2D images. • Novel synthetic dataset simulates diverse environmental conditions for AVs. • Comparative evaluation against state-of-the-art techniques for object detection.

With the ongoing integration of machine learning models into critical infrastructure, the resilience of these systems against adversarial attacks is important for all domains. This paper introduces an adversarial attack generator framework against a network dataset that is part of OCPP Dataset using CI-CFlowMeter parser. We conduct a comprehensive evaluation of various prominent adversarial attacks, including FGSMA, JSMA, PGD, C&W, and more to assess their efficacy on the OCCP dataset. The Adversarial Generator is meticulously evaluated, demonstrating a significant impact in the models performance to detect potential perturbations. The results showcased the impact of the different type of adversarial attacks, contributing to a critical advancement in future defense strategies that need to be utilised in order to protect industrial control systems.

The revolution of Artificial Intelligence (AI) has brought about a significant evolution in the landscape of cyberattacks. In particular, with the increasing power and capabilities of AI, cyberattackers can automate tasks, analyze vast amounts of data, and identify vulnerabilities with greater precision. On the other hand, despite the multiple benefits of the Internet of Things (IoT), it raises severe security issues. Therefore, it is evident that the presence of efficient intrusion detection mechanisms is critical. Although Machine Learning (ML) and Deep Learning (DL)-based IDS have already demonstrated their detection efficiency, they still suffer from false alarms and explainability issues that do not allow security administrators to trust them completely compared to conventional signature/specification-based IDS. In light of the aforementioned remarks, in this paper, we introduce an AI-powered IDS with explainability functions for the IoT. The proposed IDS relies on ML and DL methods, while the SHapley Additive exPlanations (SHAP) method is used to explain decision-making. The evaluation results demonstrate the efficiency of the proposed IDS in terms of detection performance and explainable AI (XAI).

The evolution of the telecommunication networks towards their 6th generation has emerged new research directions in order to deal with the challenges that are posed by the high-complexity of the new infrastructure. In this new era, the structuring of the network management infrastructure is a complex endeavor that should efficiently control both communication and computational resources. In this paper, we present a set of strategies for managing the computational load in a cell-free-based, converged optical-wireless 6G network. The proposed approaches target to control the computational load of a cell-free network, by either compressing the load by considering load-thresholds, or offloading the load to the SDN controller of the fixed network. Both strategies are mathematically formulated by considering traffic-engineering formulas, while their performance is evaluated by comparing analytical results with corresponding results from a baseline scenario, where no load control strategies are applied. Moreover, the proposed analysis can be applied in order to determine the capacity of the network controllers that is required in order to guarantee pre-determined Quality of Service requirements.

The transformation of the conventional electrical grid into a digital ecosystem brings significant benefits, such as two-way communication between energy consumers and utilities, self-monitoring and pervasive controls. However, the advent of the smart electrical grid raises severe cybersecurity and privacy concerns, given the presence of legacy systems and communications protocols. This paper focuses on False Data Injection (FDI) cyberattacks against a low-voltage distribution system, taking full advantage of Man In The Middle (MITM) actions. The first cyberattack targets the communication between a smart meter and an Active Distribution Management System (ADMS), while the second FDI cyberattack targets the communication between a smart inverter and ADMS. In both cases, the cyberattacks affect the operation of the distribution transformer, thus resulting in devastating consequences. Moreover, this paper provides an Artificial Intelligence (AI)-based Intrusion Detection System (IDS), detecting and mitigating the above cyberattacks in a timely manner. The evaluation results demonstrate the efficiency of the proposed IDS.

Motivated by the emergence of next-generation internet-of-things applications, this paper introduces a novel non-orthogonal semi-grant-free multiple access protocol that enables reliable massive connectivity of scheduled and random access devices with varying data rate requirements. Specifically, the protocol divides devices into two classes, namely primary and secondary devices. Primary devices (PDs) have high data rate requirements, while secondary devices (SDs) transmit data at a lower frequency compared to primary devices, and have a lower data rate requirement. The base-station (BS) enables a single PD and two SDs to access the same radio resource block (RRB). The PD uses grant-based (GB) access to the RRB, while the secondary devices use grant-free (GF) access. The BS first decodes the PD signal by treating the SDs signals as interference. Then, it decodes the signal of the SD with the higher channel gain. To assess the performance of the proposed protocol, we carried out simulations and measured the achieved throughput of each device under various parameters, such as power-to-noise gain, signal-to-noise threshold, and transmission probability.

Ultra-low latency connections for a massive number of devices are one of the main requirements of the next-generation tactile Internet-of-Things (TIoT). Grant-free non-orthogonal multiple access (GF-NOMA) is a novel paradigm that leverages the advantages of grant-free access and non-orthogonal transmissions, to deliver ultra-low latency connectivity. In this work, we present a joint channel assignment and power allocation solution for semi-GF-NOMA systems, which provides access to both grant-based (GB) and grant-free (GF) devices, maximizes the network throughput, and is capable of ensuring each device’s throughput requirements. In this direction, we provide the mathematical formulation of the aforementioned problem. After explaining that it is not convex, we propose a solution strategy based on the Lagrange multipliers and subgradient method. To evaluate the performance of our solution, we carry out system-level Monte Carlo simulations. The simulation results indicate that the proposed solution can optimize the total system throughput and achieve a high association rate, while taking into account the minimum throughput requirements of both GB and GF devices.

Real-time event detection and targeted decision making for emerging mission-critical applications require systems that extract and process relevant data from IoT sources in smart spaces. Oftentimes, this data is heterogeneous in size, relevance, and urgency, which creates a challenge when considering that different groups of stakeholders (e.g., first responders, medical staff, government officials, etc.) require such data to be delivered in a reliable and timely manner. Furthermore, in mission-critical settings, networks can become constrained due to lossy channels and failed components, which ultimately add to the complexity of the problem. In this article, we propose PrioDeX, a cross-layer middleware system that enables timely and reliable delivery of mission-critical data from IoT sources to relevant consumers through the prioritization of messages. It integrates parameters at the application, network, and middleware layers into a data exchange service that accurately estimates end-to-end performance metrics through a queueing analytical model. PrioDeX proposes novel algorithms that utilize the results of this analysis to tune data exchange configurations (event priorities and dropping policies), which is necessary for satisfying situational awareness requirements and resource constraints. PrioDeX leverages Software-Defined Networking (SDN) methodologies to enforce these configurations in the IoT network infrastructure. We evaluate our approach using both simulated and prototype-based experiments in a smart building fire response scenario. Our application-aware prioritization algorithm improves the value of exchanged information by 36% when compared with no prioritization; the addition of our network-aware drop rate policies improves this performance by 42% over priorities only and by 94% over no prioritization.

In this paper we consider a cloud radio access network (C-RAN) architecture where the baseband signal processing servers, named baseband units (BBUs) are separated from the remote radio heads (RRHs). The RRHs form a single cluster while the BBUs form a centralized pool of data center resources. Each RRH of the C-RAN accommodates bursty traffic which is expected to play a dominant role in 5G networks. We approximate bursty traffic via the compound Poisson process according to which batches of calls, with a generally distributed batch size, occur at time points that follow a negative exponential distribution. Each call of a new batch is treated separately from the other calls of the same batch. A new call requires a computational resource unit from the centralized pool of BBUs and a radio resource unit from the serving RRH. If both units are available, then the call is accepted in the RRH for an exponentially distributed service time. Otherwise, call blocking occurs. We model this C-RAN as a loss system and show that the steady state probabilities have a product form solution (PFS). Based on the PFS, we propose an efficient convolution algorithm for the accurate calculation of the main performance measures which are time and call congestion probabilities. The accuracy of this algorithm is verified via simulation.

Honeypots are powerful security tools, developed to shield commercial and industrial networks from malicious activity. Honeypots act as passive and interactive decoys in a network attracting malicious activity and securing the rest of the network entities. Since an increase in intrusions has been observed lately, more advanced security systems are necessary. In this paper a new method of adapting a honeypot system in a modern industrial network, employing the Modbus protocol, is introduced. In the presented NeuralPot honeypot, two distinct deep neural network implementations are utilized to adapt to network Modbus entities and clone them, actively confusing the intruders. The proposed deep neural networks and their generated data are then compared.

Climate change has introduced significant challenges that can affect multiple sectors, including the agricultural one. In particular, according to the Food and Agriculture Organization of the United Nations (FAO) and the International Telecommunication Union (ITU), the world population has to find new solutions to increase the food production by 70% by 2050. The answer to this crucial challenge is the suitable adoption and utilisation of the Information and Communications Technology (ICT) services, offering capabilities that can increase the productivity of the agrochemical products, such as pesticides and fertilisers and at the same time, they should minimise the functional cost. More detailed, the advent of the Internet of Things (IoT) and specifically, the rapid evolution of the Unmanned Aerial Vehicles (UAVs) and Wireless Sensor Networks (WSNs) can lead to valuable and at the same time economic Precision Agriculture (PA) applications, such as aerial crop monitoring and smart spraying tasks. In this paper, we provide a survey regarding the potential use of UAVs in PA, focusing on 20 relevant applications. More specifically, first, we provide a detailed overview of PA, by describing its various aspects and technologies, such as soil mapping and production mapping as well as the role of the Global Positioning Systems (GPS) and Geographical Information Systems (GIS). Then, we discriminate and analyse the various types of UAVs based on their technical characteristics and payload. Finally, we investigate in detail 20 UAV applications that are devoted to either aerial crop monitoring processes or spraying tasks. For each application, we examine the methodology adopted, the proposed UAV architecture, the UAV type, as well as the UAV technical characteristics and payload.

The authors study the downlink of an orthogonal frequency division multiplexing (OFDM) based cell that services calls from different service‐classes with various resource requirements. They assume that calls arrive in the cell as batches according to a compound Poisson process. They consider that the batch size is generally distributed while each call of a batch is treated separately from the other calls of the same batch, according to the complete sharing policy. To determine the most important performance metrics, i.e. congestion probabilities and resource utilisation in this OFDM‐based cell, they model it as a multirate loss model, show that the steady‐state probabilities can be determined via a product form solution (PFS) and propose recursive formulas which reduce the complexity of the calculations. In addition, they study the bandwidth reservation (BR) policy which can be used for the reservation of subcarriers in order to favour service‐classes whose calls have high subcarrier requirements. The existence of the BR policy destroys the PFS of the steady‐state probabilities. However, they show that there exist recursive formulas for the determination of the various performance measures. Simulation verifies the accuracy of the proposed formulas.

Internet of Things (IoT) aim to leverage data from multiple sensors, actuators and devices for improving peoples' daily life and safety. Multiple data sources must be integrated, analyzed from the corresponding application and notify interested stakeholders. To support the data exchange between data sources and stakeholders, the publish/subscribe (pub/sub) middleware is often employed. Pub/sub provides additional mechanisms such as reliable messaging, event dropping, prioritization, etc. The event dropping mechanism is often used to satisfy Quality of Service (Q0S) requirements and ensure system stability. To enable event dropping, basic approaches apply finite buffers or data validity periods and more sophisticated ones are informationaware. In this paper, we introduce a pub/sub mechanism for probabilistic event dropping by considering the stakeholders' intermittent connectivity and QoS requirements. We model the pub/sub middleware as a network of queues which includes a novel ON/OFF queueing model that enables the definition of join probabilities. We validate our analytical model via simulation and compare our mechanism with existing ones. Experimental results can be used as insights for developing hybrid dropping mechanisms.

Real-time event detection and targeted decision making for emerging mission-critical applications, e.g. smart fire fighting, requires systems that extract and process relevant data from connected IoT devices in the environment. In this paper, we propose FireDeX, a cross-layer middleware that facilitates timely and effective exchange of data for coordinating emergency response activities. FireDeX adopts a publish-subscribe data exchange paradigm with brokers at the network edge to manage prioritized delivery of mission-critical data from IoT sources to relevant subscribers. It incorporates parameters at the application, network, and middleware layers into a data exchange service that accurately estimates end-to-end performance metrics (e.g. delays, success rates). We design an extensible queueing theoretic model that abstracts these cross-layer interactions as a network of queues, thereby making it amenable for rapid analysis. We propose novel algorithms that utilize results of this analysis to tune data exchange configurations (event priorities and dropping policies) while meeting situational awareness requirements and resource constraints. FireDeX leverages Software-Defined Networking (SDN) methodologies to enforce these configurations in the IoT network infrastructure. We evaluate its performance through simulated experiments in a smart building fire response scenario. Our results demonstrate significant improvement to mission-critical data delivery under a variety of conditions. Our application-aware prioritization algorithm improves the value of exchanged information by 36% when compared with no prioritization; the addition of our network-aware drop rate policies improves this performance by 42% over priorities only and by 94% over no prioritization.

The authors consider a low earth orbit (LEO) mobile satellite system (MSS) that accepts new and handover calls of multirate service‐classes. New calls arrive in the system as batches, following the batched Poisson process. A batch has a generally distributed number of calls. Each call is treated separately from the others and its acceptance is decided according to the availability of the requested number of channels. Handover calls follow also a batched Poisson process. All calls compete for the available channels under the complete sharing policy. By considering the LEO‐MSS as a multirate loss system with ‘satellite‐fixed’ cells, it can be analysed via a multi‐dimensional Markov chain, which yields to a product form solution (PFS) for the steady‐state distribution. Based on the PFS, they propose a recursive and yet efficient formula for the determination of the channel occupancy distribution, and consequently, for the calculation of various performance measures including call blocking and handover failure probabilities. The latter are much higher compared to the corresponding probabilities in the case of the classical (and less bursty) Poisson process. Simulation results verify the accuracy of the proposed formulas. Furthermore, they discuss the applicability of the proposed model in software‐defined LEO‐MSS.

In the emerging area of the Internet of Things (IoT), the exponential growth of the number of smart devices leads to a growing need for efficient data storage mechanisms. Cloud Computing was an efficient solution so far to store and manipulate such huge amount of data. However, in the next years it is expected that Cloud Computing will be unable to handle the huge amount of the IoT devices efficiently due to bandwidth limitations. An arising technology which promises to overwhelm many drawbacks in large‐scale networks in IoT is Fog Computing. Fog Computing provides high‐quality Cloud services in the physical proximity of mobile users. Computational power and storage capacity could be offered from the Fog, with low latency and high bandwidth. This survey discusses the main features of Fog Computing, introduces representative simulators and tools, highlights the benefits of Fog Computing in line with the applications of large‐scale IoT networks, and identifies various aspects of issues we may encounter when designing and implementing social IoT systems in the context of the Fog Computing paradigm. The rationale behind this work lies in the data storage discussion which is performed by taking into account the importance of storage capabilities in modern Fog Computing systems. In addition, we provide a comprehensive comparison among previously developed distributed data storage systems which consist of a promising solution for data storage allocation in Fog Computing.

Internet of Things (IoT) applications consist of diverse Things (sensors and devices) in terms of hardware resources. Furthermore, such applications are characterized by the Things' mobility and multiple interaction types, such as synchronous, asynchronous, and streaming. Middleware IoT protocols consider the above limitations and support the development of effective applications by providing several Quality of Service (QoS) features. These features aim to enable application developers to tune an application by switching different levels of response times and delivery success rates. However, the profusion of the developed IoT protocols and the intermittent connectivity of mobile Things, result to a non-trivial application tuning. In this paper, we model the performance of the middleware overlay infrastructure using Queueing Network Models (QNMs). To represent the mobile Thing's connections/disconnections, we model and solve analytically an ON/OFF queueing center. We apply our approach to Streaming interactions with mobile peers. Finally, we validate our model using simulations. The deviations between the performance results foreseen by the analytical model and the ones provided by the simulator are shown to be less than 5%.

We consider a reference cell of fixed capacity in a wireless cellular network while concentrating on next-generation network architectures. The cell accommodates new and handover calls from different service-classes. Arriving calls follow a random or quasi-random process and compete for service in the cell under two bandwidth sharing policies: 1) a probabilistic threshold (PrTH) policy or 2) the multiple fractional channel reservation (MFCR) policy. In the PrTH policy, if the number of in-service calls (new or handover) of a service-class exceeds a threshold (difference between new and handover calls), then an arriving call of the same service-class is accepted in the cell with a predefined state-dependent probability. In the MFCR policy, a real number of channels is reserved to benefit calls of certain service-classes; thus, a service priority is introduced. The cell is modeled as a multirate loss system. Under the PrTH policy, call-level performance measures are determined via accurate convolution algorithms, while under the MFCR policy, via approximate but efficient models. Furthermore, we discuss the applicability of the proposed models in 4G/5G networks. The accuracy of the proposed models is verified through simulation. Comparison against other models reveals the necessity of the new models and policies.

In the recent years, mobile cellular networks are undergoing fundamental changes and many established concepts are being revisited. New emerging paradigms, such as Software-Defined Networking (SDN), Mobile Cloud Computing (MCC), Network Function Virtualization (NFV), Internet of Things (IoT), and Mobile Social Networking (MSN), bring challenges in the design of cellular networks architectures. Current Long-Term Evolution (LTE) networks are not able to accommodate these new trends in a scalable and efficient way. In this paper, first we discuss the limitations of the current LTE architecture. Second, driven by the new communication needs and by the advances in aforementioned areas, we propose a new architecture for next-generation cellular networks. Some of its characteristics include support for distributed content routing, Heterogeneous Networks (HetNets) and multiple Radio Access Technologies (RATs). Finally, we present simulation results which show that significant backhaul traffic savings can be achieved by implementing caching and routing functions at the network edge.

In this paper, we propose a multirate teletraffic loss model of a single link with certain bandwidth capacity that accommodates Poisson arriving calls, which can tolerate bandwidth compression (elastic traffic), under the threshold policy. When compression occurs, the service time of new and in-service calls increases. The threshold policy provides different QoS among service-classes by limiting the number of calls of a service-class up to a predefined threshold, which can be different for each service-class. Due to the bandwidth compression mechanism, the steady state probabilities in the proposed model do not have a product form solution. However, we approximate the model by a reversible Markov chain, and prove recursive formulas for the calculation of call blocking probabilities and link utilization. The accuracy of the proposed formulas is verified through simulation and found to be very satisfactory.

Software-defined networking (SDN) has gained a tremendous attention in the recent years, both in academia and industry. This revolutionary networking paradigm is an attempt to bring the advances in computer science and software engineering into the information and communications technology (ICT) domain. The aim of these efforts is to pave the way for completely programmable networks and control-data plane separation. Recent studies on feasibility and applicability of SDN concepts in cellular networks show very promising results and this trend will most likely continue in near future. In this work, we study the benefits of SDN on the radio resource management (RRM) of future-generation cellular networks. Our considered cellular network architecture is in line with the recently proposed Long-Term Evolution (LTE) Release 12 concepts, such as control-data plane split, heterogeneous networks (HetNets) environment, and network densification through deployment of small cells. In particular, the aim of our RRM scheme is to enable the macro base station (BS) to efficiently allocate radio resources for small cell BSs in order to assure quality-of-service (QoS) of moving users/vehicles during handoffs. We develop an approximate, but very time- and space-efficient algorithm for radio resource allocation within a HetNet. Experiments on commodity hardware show algorithm running times in the order of a few seconds, thus making it suitable even in cases of fast moving users/vehicles. We also confirm a good accuracy of our proposed algorithm by means of computer simulations.

Optical Code Division Multiple Access (OCDMA) provides increased security communications with large dedicated bandwidth to end users and simplified network control. We analyse the call-level performance of an OCDMA Passive Optical Network (PON) configuration, which accommodates multiple service-classes with finite traffic source population. The considered user activity is in accordance with the bursty nature of traffic, so that calls may alternate between active (steady transmission of a burst) and passive states (no transmission at all). Parameters related to multiple access interference, additive noise, user activity and number of traffic sources are incorporated to our analysis, which is based on a two-dimensional Markov chain. An approximate recursive formula is derived for efficient calculation of call blocking probability. Furthermore, we determine the burst blocking probability; burst blocking occurs when a burst delays its returning from passive to active state. The accuracy of the model is completely satisfactory and is verified through simulation. Moreover, we reveal the consistency and necessity of the proposed model.

In this paper, we consider a single link that supports both elastic and adaptive traffic of Batch Poisson arriving calls, under the Bandwidth Reservation (BR) policy, whereby we can achieve specific QoS per service-class. Arriving batches have a generally distributed batch size, and can be serviced either as a whole or in part (partial batch blocking discipline), depending on the available link bandwidth. Blocked calls are lost. Accepted calls of a batch can compress or expand their bandwidth; elastic calls expand or compress their service time accordingly, while adaptive calls do not alter their service time. This system does not have a Product Form Solution. For the efficient calculation of time and call congestion probabilities as well as link utilization, we derive approximate but recursive formulas. The accuracy of the model is completely satisfactory and is verified together with the model's consistency, through simulation. Comparison of the new model with existing models reveals its necessity.

We present a new loss model for the call-level analysis of a single link, which accommodates calls of different service-classes with elastic bandwidth requirements. Calls arrive in the link according to a Batch Poisson process, a process that can be used to model traffic, which is more 'peaked' and 'bursty' than the Poisson process. The available link bandwidth is shared to calls according to the Bandwidth Reservation policy, whereby we can guarantee certain Quality-of-Service for each service-class. In the proposed model, we assume a general batch size distribution and the Partial Batch Blocking discipline. According to this discipline, one or more calls of an arriving batch can be accepted, while the rest can be discarded, depending on the available link bandwidth. New and in-service calls tolerate bandwidth compression/expansion. The analysis of the system is based on Markov chains. Since no Product Form Solution exists, for an efficient solution, we propose an approximate reversible Markov chain. Based on it, we derive a recursive formula for the calculation of link occupancy distribution and consequently time and call congestion probabilities (important call-level performance metrics). The proposed model's accuracy and its consistency are verified by simulation and found to be quite satisfactory.

We present a new model, named Wireless Finite Connection-Dependent Threshold Model, for the call-level analysis of W-CDMA networks that support both elastic and stream traffic. Calls generated by service-classes of finite source population (quasi-random call arrival process) compete for their acceptance to a W-CDMA cell, under the complete sharing policy. An arriving call can be accepted with one of several contingency Quality-of-Service (QoS) requirements, depending on the resource availability in the cell; the latter is indicated by thresholds. We present an approximate but recurrent formula for the efficient calculation of the system state probabilities; consequently, the call blocking (time congestion) probabilities and other performance metrics in the uplink direction are provided. The model's accuracy is verified by simulation and found to be quite satisfactory. Moreover, the proposed model performs much better than the corresponding model of infinite number of sources.

We propose a new teletrafflc model for the calculation of link occupancy distribution and determine the call blocking probabilities and link utilization in a single-link multi-rate loss system. We considerKdifferent service-classes, either of elastic or of adaptive type, that compete for the available link bandwidth under the complete sharing policy. Calls arrive to the link with several contingency bandwidth requirements that depend on thresholds, which indicate the total link occupied bandwidth. It is possible for a call, while in service with a certain bandwidth to reduce it, in order for new calls to be accepted in the link. The accuracy of the new model is verified by simulation results. Moreover, the proposed model performs much better than the existing model for elastic and adaptive traffic.

We propose a new teletraffic model, named Wireless Engset Multi-rate Loss Model (W-EnMLM) for the call-level analysis of W-CDMA networks, supporting heterogeneous service-classes of finite population of traffic sources. Quasi- random arriving calls (generated by mobile users), compete for their admission to a W-CDMA cell under the Complete bandwidth Sharing (CS) policy. We present an approximate but recurrent formula for the efficient calculation of the system state probabilities and the call blocking probabilities in the uplink direction. The accuracy of the proposed model is verified by simulation results and found to be completely satisfactory. Moreover, the proposed model performs much better than the corresponding model of infinite number of traffic sources.

We consider an ON-OFF traffic model of a single link which accommodates service-classes of finite population. Calls arrive according to a quasi-random process and, if accepted, enter the system via state ON; then calls may alternate between ON-OFF states. When a call is transferred to state OFF, it releases the bandwidth held in state ON, while when it tries to return to state ON it re-requests its bandwidth. If it is available a new ON-period (burst) begins; otherwise burst blocking occurs and the call remains in state OFF. We prove that the proposed finite source ON-OFF model (f-ON-OFF) has a product form solution and provide an accurate recursive formula for the call blocking probabilities calculation. For the burst blocking probabilities calculation we propose an approximate formula. Finally, we generalize the f-ON-OFF model to include a mixture of service-classes of finite and infinite number of sources. Simulation results validate our analytical methodology.

The call-level QoS assessment in ATM networks remains an open issue, due to elastic services. We consider the coexistence of available bit rate (ABR) services with QoS guarantee services in a virtual path (VP) link and evaluate the call blocking probabilities (CBP). To this end, firstly, we review two extensions of the Erlang multirate loss model and examine their applicability on ABR services. In the first extension, the retry models, blocked calls can retry with reduced resource requirements and increased arbitrary mean residency requirements. In the second extension, the threshold models, request sizes and residency times are state dependent. For both models, we present expressions for efficient calculation of the final CBP. Secondly, we propose the connection-dependent threshold model, which resembles the threshold models, but the state dependency is individualized among call connections. Finally we evaluate the above-mentioned models by comparing them based on the resultant CBP. Our investigation shows that the retry models can hardly approach the behavior of ABR services, while the threshold models perform better than the retry models. The proposed connection-dependent threshold model performs much better than the threshold models. Simulation results validate the proposed model.