The transition to quantum-resistant cryptography is becoming increasingly urgent as advances in quantum computing threaten widely deployed public-key systems such as RSA and elliptic-curve cryptography. While the National Institute of Standards and Technology (NIST) has recently standardized several post-quantum cryptographic (PQC) algorithms, there remains limited empirical data on their practical performance and energy characteristics on commodity hardware.

Distributed ledger technologies (DLTs) form critical infrastructure for decentralized applications, yet their security relies heavily on classical asymmetric cryptographic primitives that are vulnerable to quantum attacks. Post-quantum cryptography (PQC) provides candidate algorithms designed to resist such threats, but integrating these schemes into operational blockchain systems introduces significant architectural and performance trade-offs.

The growing pressures of urbanization, vehicular proliferation, and fragmented parking infrastructure pose significant sustainability and mobility challenges in modern cities. In response, we present SPOSChain (Smart Parking Optimization with SDN and Blockchain), a novel Blockchain-enhanced and Software-Defined Networking (SDN)-based smart parking system that unifies independent parking providers under a decentralized, intelligent coordination framework. SPOSChain introduces a four-layer architecture integrating IoT, data, control, and blockchain layers, to ensure transparency, scalability, and real-time responsiveness. The core parking assignment task is formulated as a fairness-driven optimization problem, which is mathematically equivalent to a parallel job scheduling problem, known to be NP-hard, thereby necessitating the development of efficient heuristic strategies. To this end, we propose, adopt, and evaluate multiple heuristic and hybrid algorithms, including Local Search, Branch-and-Bound, and Genetic Search, culminating in a time-aware Hybrid Search model. Simulation results under diverse vehicle arrival distributions (uniform, normal, and exponential) demonstrate that our approach significantly reduces load imbalance, quantified via the Total of Differences metric, while improving responsiveness and maintaining scalability. SPOSChain not only enables equitable and efficient parking allocation but also supports sustainable urban mobility by reducing driver search time, CO 2 emissions, and network overhead. These results underscore the transformative potential of programmable, decentralized parking systems in future smart city infrastructures.

The rise in cyber attacks demands improved network security for resource-constrained IoT devices. We assess machine learning models on the CICIDS-2017 dataset, emphasizing the need for a balance between performance and resource efficiency. Principal component analysis (PCA) aids in creating efficient models, with XGBoost and Decision Tree as suitable options, providing decent performance without excessive resource requirements. Rebalancing techniques can enhance recall but may lower precision.

Driven by a cybercrime industry worth over $6 trillion, the demand for cybersecurity professionals is surging with a projected increase of over 35% in jobs over the next decade according to the US Department of Labor's BLS. Simultaneously, advancements in Artificial Intelligence (AI), Data Science (DS), and Machine Learning (ML) are reshaping industries and cre-ating challenges, notably job displacement for the under-skilled. This research-to-practice full paper addresses the critical need to integrate AI/DS/ML into cybersecurity education to meet these evolving demands. We envision a marriage of AI/DS/ML and cybersecurity (ADM4CYB) through a research-to-education paradigm. Despite extensive research, there remains a significant lack of publicly accessible labs that apply AI, DS, and ML to real-world cyber-security challenges. As such, readily available, easy to adopt, modular, and hands-on labs applying AI/DS/ML solutions to cybersecurity problems are highly desirable. At this juncture, we are developing a methodology that enables instructors and researchers to transform their work into practical, safe, and interactive teaching labs, enhancing student learning across various cybersecurity topics. Our initiative aims to integrate AI, DS, and ML techniques into interactive labs to effectively address current cybersecurity challenges. These modular labs provide students with both theoretical knowledge and hands-on experience, essential for adapting to the rapidly evolving field. By creating accessible labs, our methodology empowers students to develop skills across various cybersecurity topics. We also report results of our preliminary evaluation after using the first lab from this methodology in two classes that show increased learning, heightened levels of interest and motivation in the fields of networking, cybersecurity, AI/DS/ML, as well as coding.

Computing has become an essential field of study, with programming as its core component. However, pedagogical aspects of teaching programming remain under-explored and poorly understood. The selection of a student's first programming language has sparked considerable debate within the community, often referred to as a “culture war”, and presents curricular challenges that extend to subsequent language choices. This research full paper investigates the transitions between different programming languages and measures students' learning experiences as they progress through a single language. We utilize Mindshift Learning Theory (MLT) to examine the transfer of knowledge across programming languages as students go through courses in a pre-requisite chain in our curriculum. Our study focuses on three aspects: novelty (encountering new, unique concepts), change (differently applying previously exposed topics), and carryover (directly transferring concepts to new contexts). Our methodology includes an online survey of current and past students transitioning through these courses. We aim to understand how perceived transitions impact mastery and confidence in programming languages. We designed the survey questions, identified target classes, and implemented incentive mechanisms to ensure a statistically significant response rate. Our results are partially aligned with earlier findings in the literature, which correlated carryover, changed, and novel concepts with knowledge scores of object oriented (OO) developers, despite a totally different demographic. Namely, our student participants transitioning to C/C++ (procedural programming) and OO C++ mostly have high, moderate, and low knowledge scores on the programming concepts they perceive as carryover, novel, and changed respectively. On the other hand, those transitioning to OO Java have higher scores for the changed concepts.

This innovative practice full paper introduces an automated assessment system, Michigan System of Automated Assessment for Accreditation (MSA3), designed for computing and engineering programs to meet both external accreditation and internal university requirements while minimally increasing faculty and staff workload. The system integrates with Canvas, enabling instructors to input assessment data directly during grading, streamlining the process by eliminating the need for separate data entry and report dissemination. Key features include automated data export and the generation of visual analytics, securely hosted on university servers. Additionally, the paper presents a preliminary study on how this system affects faculty attitudes towards the assessment workload required by ABET. Our preliminary survey results indicate a favorable perception by faculty, which should increase their buy-in and engagement with assessment, supporting assertions made in the literature. Our findings also suggest that the system may help align faculty efforts with accreditation demands more seam-lessly. Its code will be made available on GitHub, allowing for easy adaptation by other institutions using Canvas. Our future work includes increasing the level of automation, extending our methodology to other learning management systems, and conducting a larger evaluation to gauge MSA3's efficacy on the perception of the accreditation and faculty/staff buy-in.

Software-Defined Networking (SDN) and Blockchain (BC) are transformative technologies reshaping network management and security, utilizing their synergies. SDN’s centralized control enhances flexibility and efficiency but introduces vulnerabilities due to its single point of control. With its decentralized and immutable ledger, BC offers a solution by distributing control and providing a tamper-proof audit trail. This paper integrates these technologies to address blackhole attacks—a critical vulnerability where compromised SDN controllers disrupt network performance. We propose Blockchain-Enhanced SDN for Adaptive Path Finding (BeS4APF) algorithm against blackhole attacks in the multidomain SDNs. The algorithm maintains a high Packet Delivery Ratio (PDR) by dynamically adjusting network paths in response to node failures. The BeS4APF algorithm presented effectively maintains a high PDR by dynamically adjusting paths in response to node failures. The methodology involves monitoring the network, detecting compromised nodes, and recalculating optimal paths using Dijkstra’s algorithm and node-disjoint path selection. Experiments with synthetic networks of varying sizes (from 60 to 120 domains) demonstrate that the algorithm successfully handles domain-compromising node attacks, maintaining high PDR even with increasing stochastic disruptions. Our preliminary results show that, while PDR slightly deviates and recovers in smaller networks, it stabilizes towards almost 100% in larger networks after initial adjustments. This work advances the integration of SDN and BC, offering a robust approach to securing modern networks against evolving stochastic attack scenarios and threats.

Machine Learning (ML) has proven itself to be one of the most powerful tools to detect and prevent cyber-attacks. In this study, ML is used for anomaly detection in an Internet of Things (IoT) network. This paper proposes a Two-Tier Hybrid Learning Model for Clustering and Classification (TT-HLMCC) of unlabeled loT data. The proposed method uses two algorithms for clustering: K-means and DBSCAN (Density Based Spatial Clustering for Applications with Noise). Random Forest is used for the final classification stage on the newly labeled data. The TT-HLMCC model (Proposed Model) is compared to the performance metrics of three other algorithms. While the TT-HLMCC model matches and slightly beats the performance of [3], our computational time is significantly lower.

Huge potentials of optical networks provide an excellent choice in meeting the exponential growth in bandwidth demand. In particular, Elastic Optical Networks (EONs), with its promising potentials in increased availability, failure resilience, load-balancing, new use cases for on-demand opportunities, and efficient resource allocations, are good candidates for the future high-speed networks. When EONs are augmented with the emerging Software Defined Networking (SDN), that decouples the data and control planes, an agile and synergistic combination emerges under a general term of SDN-based optical networking (SDON), albeit with operational challenges. In this paper, we introduce a blockchain-enabled QoS-based inter-Internet Service Provider (ISP) routing coordination framework for SDONs, SpectrumChain (SC), to eliminate centralized mediators and reduce QoS signaling overhead. Experimental results show that the SC framework can effectively handle the QoS-enabled inter-ISP routing in SDON architecture in terms of flow setup time, message overhead, and requests serviced as compared to the traditional approaches.

Internet of Things (IoT) networks consist of sensors, actuators, mobile and wearable devices that can connect to the Internet. With billions of such devices already in the market which have significant vulnerabilities, there is a dangerous threat to the Internet services and also some cyber-physical systems that are also connected to the Internet. Specifically, due to their existing vulnerabilities IoT devices are susceptible to being compromised and being part of a new type of stealthy Distributed Denial of Service (DDoS) attack, called Mongolian DDoS, which is characterized by its widely distributed nature and small attack size from each source. This article proposes a novel anomaly-based Intrusion Detection System (IDS) that is capable of timely detecting and mitigating this emerging type of DDoS attacks. The proposed IDS’s capability of detecting and mitigating stealthy DDoS attacks with even very low attack size per source is demonstrated through numerical and testbed experiments.

Data flows of various applications, such as video conferencing, telesurgery, online-gaming, etc., require particular treatment regarding Quality of Service (QoS) because of their sensitivity to such parameters as high bandwidth, low delay, and so on. However, provisioning of QoS-based routing services across Autonomous Systems (ASes) poses gruelling challenges due to the distributed nature and business confidentially hesitations. This study presents a novel blockchain-aided QoS-enabled inter-AS routing framework,QoSChain (QC), by blending merits of Software-Defined Networking (SDN) and Blockchain technologies.QCframework introduces a coordination framework with mutually distrusting participants that eliminates centralized mediators in provisioning QoS among SDN-enabled ASes. Also,QCreduces the time to set up an end-to-end (E2E) path and the number of messages exchanged and processed by SDN controllers for a QoS flow. Furthermore, it alleviates the concerns of network operators regarding potentially sensitive information disclosure resulting from the required information sharing for QoS-based cross-AS paths by masking details of any confidential information. Experimental results verify the feasibility of the proposed framework on E2E QoS-based routing among SDN networks. This work aims at being a useful primer for researchers in providing insights on how blockchain can be deployed in QoS-enabled routing.

The stakeholders in the computer science education have been trying to pinpoint the most relevant factors and variables contributing significantly to student recruitment, success, and retention in computing degrees over the last several decades. The goal of these efforts is to intervene using these controllable variables as early as possible and at key transition junctions for the most benefit to the students. Yet, there is still not a general agreement, let alone a consensus, neither on the most important factors nor on the degree of their contributions. While we cannot provide a panacea to this elusive and challenging problem, in this paper, we report our initial study, analyses, and results of what we intend to transform into a longitudinal undertaking to delineate success factors in computing programs in general, and in CS2 (data structures) courses in particular. Our analyses confirm some of the findings from the literature, such as CS1 performance as the most important contributor for CS2 success as well as prior programming experience, and perception of computing. At the same time we diverge from some of the literature on such factors as high school GPA or CSO performance. Our conclusions corroborate one of the rare consensus in the computing education literature that more efforts should still be exerted in this important topic. We also provide a new and novel taxonomy of the factors for success in computing programs as well as a short summary of the ACM Guidelines over time.

The impact of cybersecurity attacks on the Smart Grid may cause cyber as well as physical damages, as clearly shown in the recent attacks on the power grid in Ukraine where consumers were left without power. A set of recent successful Distributed Denial-of-Service (DDoS) attacks on the Internet, facilitated by the proliferation of the Internet-of-Things powered botnets, shows that it is just a matter of time before the Smart Grid, as one of the most attractive critical infrastructure systems, becomes the target and likely victim of similar attacks, potentially leaving catastrophic disruption of power service to millions of people. It is in this context that we propose a scalable mitigation approach, referred to as Minimally Invasive Attack Mitigation via Detection Isolation and Localization (MIAMI-DIL), under a hierarchical data collection infrastructure. We provide a proofof- concept by means of simulations which show the efficacy and scalability of the proposed approach.

A key enabler for the smart grid is the fine-grained monitoring of power utilization. Although such a mechanism is helpful in the optimization of the whole electricity generation, distribution, and consumption cycle, it also creates opportunities for the potential adversaries in deducing the activities and habits of the subscribers. In fact, by utilizing the standard and readily available tools of nonintrusive load monitoring (NILM) techniques on the metered electricity data, many details of customers' personal lives can be easily discovered. Therefore, prevention of such adversarial exploitations is of utmost importance for privacy protection. One strong privacy preservation approach is the modification of the metered data through the use of on-site storage units in conjunction with renewable energy resources. In this study, we introduce a novel mathematical programming framework to model eight privacy-enhanced power-scheduling strategies inspired and elicited from the literature. We employ all the relevant techniques for the modification of the actual electricity utilization (i.e., on-site battery, renewable energy resources, and appliance load moderation). Our evaluation framework is the first in the literature, to the best of our knowledge, for a comprehensive and fair comparison of the load-shaping techniques for privacy preservation. In addition to the privacy concerns, we consider monetary cost and disutility of the users in our objective functions. Evaluation results show that privacy preservation strategies in the literature differ significantly in terms of privacy, cost, and disutility metrics.

As many countries embark on a major transitioning to the Smart Grid (SG), realizing the benefits of the new paradigm depends on generation, collection, and usage of unprecedented amount of data. With such proliferated data processing, there are legitimate concerns that call for sound privacy measures to ensure proper safekeeping. A significant portion of the potential benefits of the SG initiative, thus, relies on devising convincing mechanisms to strengthen the privacy. As a new distribution subsystem, microgrids are expected to play a crucial role in the SG. In this paper, we develop a privacy-aware microgrid power scheduling formulation with renewable sources and energy storage where five different classes of appliances are prioritized by smart meters. Our analysis shows that there is a tradeoff between maximizing the power usage and privacy levels. Hence, it is not possible to find a solution that maximizes both objectives simultaneously. However, it is possible to achieve significantly higher levels of privacy preservation with moderate sacrifice from the power usage, especially when the number of users is high.

Smart Cities include the concept of large, intelligent systems coming together cooperatively to enable better living conditions in future urban areas. Reduction in fuel/electricity usage, crime, improved economies, less waste, and overall better living conditions are primary goals. A key enabling technology is the Intelligent Transportation System (ITS), in which various transportation mechanisms cooperate to improve quality of life by reducing fuel/electricity usage, decreasing congestion, and decreasing travel time. Studies have shown that enormous resources are wasted by congestion alone in transportation systems. In this position paper, we present the ITS as a major factor in facilitating the smart city of the future.

At the tipping of a paradigm shift in the way energy is produced, transmitted and delivered, the research efforts have not been paralleled by the curricular development. With rapid pace of changes in the field of Smart Grid (SG), the traditional research and educational efforts have been a major domain for electrical engineers. As a mode of discovery and education, interdisciplinarity facilitates broadened perspectives, ability to synthesize, analyze, integrate, and apply knowledge, and out-of-the-box thinking. The major contributions of this paper are discussion of requirements of the educational efforts in SG with special emphasis on multidisciplinarity, survey of the related work, which is the first in the literature to the best of our knowledge, and a discussion of the content for such an effort. A mixed team of power engineers and computer scientists are developing a layered curriculum starting from the introductory material from a variety of SG topics. A distinguishing advantage is the availability of many software tools and the state-of-the-art testbed as a result of years of research.

The energy sector worldwide has embarked on a transformational process to modernize the over-a-century-old power grid under an umbrella term of the Smart Grid. This vast infrastructural upgrade and operational change involves integration of a variety of advanced digital computing, communications and industrial control technologies. This brings new capabilities, but also necessitates a re-education of the aging workforce and training of the emerging workforce. While training does exist, the training approach and the accessibility of that training is often at odds with the needs of the utilities. To support this education and training need, in this paper we introduce the beginning of a modular, hands-on and open Smart Grid cybersecurity educational training platform and supporting materials together with an assessment of a preliminary version leveraged at the Trustworthy Cyber Infrastructure for the Power Grid (TCIPG) Summer School held in 2013. We base pedagogical pillars onto: (1) Active Learning that promotes analysis, synthesis, and evaluation of the content from Bloom's taxonomy, (2) The theory of project-based learning, (3) Piaget's learn-by-doing posture, and (4) Constructivist perspective of education. The main goal of this effort is to develop a complete, phased, and modular learning platform to provide the essential base knowledge and hands-on training exercises for understanding and demonstrating competency in Smart Grid cybersecurity.

With the proliferation of the application areas of the Wireless Sensor Networks (WSNs) and the technological advances in sensor electronics, target tracking has become one of the most important functions in WSNs. The overwhelming majority of moving target tracking studies in WSNs has assumed either one-dimensional space (lineland), or two-dimensional domain (flat-land). However, many existing, such as environmental monitoring of rugged terrain, structural health monitoring of buildings, underground and underwater WSN applications, and emerging applications, such as search-and-rescue via Micro Aerial Vehicles, reconnaissance by insect-sized flying robots, need target tracking in three-dimensional space. In this study, we assume a Wireless Aerial Sensor Network (WASN) with conventional low-power and low-complexity features and devise a low spatial and temporal complexity moving target tracking mechanism in 3D based on g-h filter that considers measurement errors. Our prefatory simulations show promising preliminary results for our approach in terms of very low inaccuracy rates between the true location of the target and the forecasts at the sink node as well as negligible estimate errors in terms of Euclidean distance.

With classroom instruction undergoing a massive transformation to incorporate online learning techniques at unprecedented levels, technological advances have facilitated a range of mechanisms that improve teaching and learning. At the same time, these technological advances have also facilitated different forms of cheating in classes.

The ever-increasing complexity and challenges of the higher education coupled with more public scrutiny for its efficacy necessitate more in depth analysis and evaluation of the educational best practices. One such practice is the cooperative (co-op) education. While co-op education has been around for more than a century, its implementation is still barely more than nominal in higher education. Even though there are studies published to demonstrate its effectiveness, more in depth studies, both spatial and temporal, seem to be warranted to narrow the widening gap between students' jobs-first and institutions' intellectual-experience-first expectations. In this paper, we summarize the first and highly successful co-op program in Turkey at TOBB University of Economics and Technology (TOBB ETU). Our analysis include a taxonomy of different co-op programs to contextualize TOBB ETU's. Further, we touch upon the theoretical underpinnings from a variety of generally accepted foundational work. As part of the evaluation of the program, we provide a preliminary assessment of a longitudinal study as the first from Turkey and one of a very few from the non-US, international institutions. We believe that ours is the first to report an evaluation of co-op programs in terms of pre-co-op, post-co-op and graduated students separately.

The impact of the computing field is becoming more profound everyday in all facets of modern society. Despite significant efforts from academia and relatively high demand for computing, technology producers are needed more than ever. This has been corroborated by a recent NSF Press Release [1], stating that computing is the only STEM discipline with more job openings than college graduates to fill them up. A major responsibility for such an outcome rests upon educators' ability to produce professionals capable of handling ever more complex technical challenges. In this challenging environment, the importance of practical work in science and engineering, supported by lab exercises, is widely known. While it is desirable to create physical laboratories for teaching these technologies, it is not always possible nor desirable. In this paper, we present an effective alternative approach, in the form of an affordable, virtualized laboratory environment that can be used in a variety of computing and engineering courses, as well as in other fields, in lieu of physical labs. Results of a survey supporting the validity of this approach are also presented. Students in four different classes were surveyed concerning the effectiveness and desirability of this virtual laboratory approach. The survey results indicate that student response to and interest in the virtual environment is very strong.

Mobility has been introduced to sensor networks through the deployment of movable nodes. In movable wireless networks, network connectivity among the nodes is a crucial factor in order to relay data to the sink node, exchange data for collaboration, and perform data aggregation. However, such connectivity can be lost due to a failure of one or more nodes. Even a single node failure may partition the network, and thus, eventually reduce the quality and efficiency of the network operation. To handle this connectivity problem, we present PADRA to detect possible partitions, and then, restore the network connectivity through controlled relocation of movable nodes. The idea is to identify whether or not the failure of a node will cause partitioning in advance in a distributed manner. If a partitioning is to occur, PADRA designates a failure handler to initiate the connectivity restoration process. The overall goal in this process is to localize the scope of the recovery and minimize the overhead imposed on the nodes. We further extend PADRA to handle multiple node failures. The approach, namely, MDAPRA strives to provide a mutual exclusion mechanism in repositioning the nodes to restore connectivity. The effectiveness of the proposed approaches is validated through simulation experiments.

Finding the most likely path satisfying a requested additive Quality-of-Service (QoS) value, such as delay, when link metrics are defined as random variables by known probability distributions is NP-Hard. We transform the probability distributions into the Laplace domain, find the Laplace Transform of their convolutions and numerically inverse to find the distribution function in the time domain. Picard's iterative method of successive approximations is used to find the solution. To the best of our knowledge, ours is the first to propose a transform-based approach for the QoS routing problem of finding the most likely path. Simulations show that our stochastic approach (1) Selects correct paths more frequently, (2) Incurs less overhead with respect to the dissemination and processing of state information, and (3) Reduces the churn by selecting more stable paths.

Network applications that are in need of some level of guarantees from the network to operate, such as multimedia programs, are not well served with the conventional best-effort service of the IP-based networks. The difficulty of finding preferential paths for those applications is compounded by the intrinsic inaccuracies of the network state information maintained by the nodes that have to make such decisions. We use a probabilistic modeling and a framework to select paths for applications that want more cooperation from the network to operate satisfactorily. The links are associated with additive link metrics. We represent the stochasticity of links by means of a new composite metric composed of an interval with a lower and upper bound and an associated probability. The interpretation and relevance of our metric is such that in the next decision time period the expected value of the resource is between the upper and the lower bound with the associated probability. Three simple and straightforward methods of computing our composite metric are presented. An algorithm, called Augmented-Dijkstra Additive Metric (ADAM), with the same complexity as the standard Dijkstra algorithm, provides an effective solution for statistical additive link metric (such as delay) guarantees. Simulation results conducted in ns2 evaluate and confirm the effectiveness of our approach.

Using multiple radios/channels in wireless mesh networks (WMNs) can significantly boost the throughput of the network and brings broadband wireless access to more users with reduced cost. However, this requires careful assignment of channels to each radio so that interference due to parallel communications can be minimized and the network is not partitioned. We propose a distributed channel assignment protocol for WMNs to maximize the total number of non-interfering concurrent active links and guarantee the network connectivity. The motivation for maximizing the number of links is to give as much leeway to the routing subsystem as possible. The problem is modeled as a list coloring problem where each channel corresponds to a color. The main idea is to determine the minimum degree spanning tree (MDST) of the network for providing more parallel transmissions and guaranteeing connectivity. The links that are not part of the MDST are assigned channels based on a 5-way handshake protocol to resolve arising conflicting requests in the distributed approach. We validate the performance by comparing our approach to a greedy channel assignment that does not guarantee connectivity.

Wireless sensor and actor networks (WSANs) additionally employ actor nodes within the wireless sensor network (WSN) which can process the sensed data and perform certain actions based on this collected data. In most applications, inter-actor coordination is required to provide the best response. This suggests that the employed actors should form and maintain a connected inter-actor network at all times. However, WSANs often operate unattended in harsh environments where actors can easily fail or get damaged. Such failures can partition the inter-actor network and thus eventually make the network useless. In order to handle such failures, we present a connected dominating set (CDS) based partition detection and recovery algorithm. The idea is to identify whether the failure of a node causes partitioning or not in advance. If a partitioning is to occur, the algorithm designates one of the neighboring nodes to initiate the connectivity restoration process. This process involves repositioning of a set of actors in order to restore the connectivity. The overall goal in this restoration process is to localize the scope of the recovery and minimize the movement overhead imposed on the involved actors. The effectiveness of the approach is validated through simulation experiments.

We study the problem of finding the most likely path satisfying a requested additive Quality-of-Service (QoS) value, such as delay. The link metrics are defined as random variables following Weibull probability distributions as empirically reported in [13] and analytically derived in [12]. The problem of finding the most likely path is NP-Hard [24]. Our approach involves reducing the complicated probability convolutions necessary to calculate the most probable path that satisfies a requested delay value. With the reduction of the objective function, an extended Bellman-Ford algorithm is devised to solve the problem. The resulting approach have the same complexity as the standard Bellman-Ford algorithm. Our reduced objective function only needs the location parameter of the Weibull distributions, hence avoiding the complexity of inferring the shape and scale parameters. We evaluate the performance of our approach by simulations and conclude with possible extensions of our work.

Not all network applications can make do with the best-effort service of the original Internet design. Many new applications, including multimedia and mission-critical ones, require a certain level of service from the network to operate properly. The difficulty of finding such preferential paths is compounded by the intrinsic inaccuracies of the network state information maintained by the nodes that have to make such decisions. We choose a stochastic framework to select paths for applications that want more cooperation from the network to operate satisfactorily. We represent the stochasticity of links by means of a new composite metric composed of an interval with a lower and upper bound and an associated probability. The interpretation and relevance of our metric is such that in the next decision time period the expected value of the resource is between the upper and the lower bound with the associated probability. Simple and straightforward methods of computing our composite metric are presented. An algorithm, called Augmented-Dijkstra, with the same complexity as the standard Dijkstra, provides an effective solution for statistical bandwidth guarantees. Simulation results evaluate and confirm the effectiveness of our approach.

In this paper we present results which suggest that Internet packet delay, when viewed as a time series, is self-similar in nature. A self-similar phenomenon displays the same or similar statistical properties across a wide range of time scales. Self-similarity has previously been observed in the magnitude of traffic transmitted on local-area and wide-area networks. Our research is the first to explore the self-similarity of traffic delay rather than traffic magnitude. Our results imply that the delay experienced by application layer protocols is extremely bursty and that traditional Poisson models cannot accurately predict the degree of burstiness. Furthermore, we present evidence that the degree of self-similarity for a round-trip path in the Internet may be correlated with the packet loss observed on that path.