Fuzzy-Neural Web Switch Supporting Differentiated Service

Intelligent Engineering Systems

In a server cluster providing massively multiplayer online games, it is critical that the game workload is balanced among servers. This study develops a new mechanism, named Game Connection Handoff in operating system, which is able to transparently hand off a client's live game connection between servers. The responding server for a player can be dynamically changed and the game workload thus can be dynamically adjusted to balance the server loads. An effective request distribution mechanism is also proposed, named game server locality-aware request distribution. This takes into account the information of game maps and server loading to dynamically distribute game maps and players among game servers. The proposed mechanisms have been implemented in the Linux kernel of LVS-CAD cluster. Practical experiments using a real multiplayer online game--Stendhal--show that the proposed mechanisms not only effectively increase the service capacity, but also transform the game server architecture from the original with a single server into one with multiple servers.

The differentiated services (DS) architecture provides a framework for the scalable provisioning of multiple service levels in the Internet. Its definition and initial work have concentrated mainly on per-hop behaviors (PHB) and mechanisms at each DS domain. Equally important is what happens between DS domains. Thanks to the flexibility of the architecture, it is the providers’ choice how to interconnect with peers. In principle, traditional, static peering agreements work together with DS, but they do not offer the flexibility and dynamics needed in an electronic market for network capacity. Therefore, we look at dynamic service level agreements at the interdomain level. Such agreements are established by software entities called traders. These traders follow market-based principles to decide which contracts will be beneficial. In particular, traders compare the offers made by neighbor providers and select the most interesting ones. This selection of peer services creates competition among providers and integrates route selection based on service level and destination. We describe and implement a framework for service level agreement trading. We show the basic workings and first performance results of SLA trading using specific traders and an experimentally defined PHB in a simulation environment. Keywords— Network architecture, differentiated services architecture, service level agreement, pricing, trading, interdomain QoS routing.

This paper describes the development of fuzzy systems for modeling the hysteresis behavior of shape memory alloy (SMA) actuators. Due to their simplicity and ease of actuation, SMA actuators are very attractive for applications such as miniature robots for micro manufacturing. However, SMAs have not been widely used for motion control applications due to their nonlinear behavior and control difficulties. One approach to design a position controller for SMA systems is to employ an inverse-model of the system in the control loop to compensate the hysteresis properties of the material. Fuzzy systems, due to their nonlinear learning and adaptation abilities, are good candidates for obtaining inverse-models. In this paper two fuzzy modeling approaches are employed and compared to develop a model for a SMA wire actuator. A set of experiments are conducted to generate the training data. The test stand includes a Nickel-Titanium (TiNi) SMA wire, a position sensor, a bias spring and a current amplifier. By comparing the performance of the two employed fuzzy modeling techniques, it is revealed that the approach based on fuzzy Gustafson-Kessel (GK) clustering shows a better performance in the modeling of the hysteresis in the SMA wire. Thus, GK clustering algorithm is employed to develop the inverse-model for the SMA. The reported results demonstrate the ability of the employed fuzzy algorithm for modeling the hysteresis in the system, and the merits of the introduced inverse-model in the control of the position of the SMA.

A request distribution method for clustered VOD servers considering buffer sharing effects

Vaginismus is an involuntary muscle contraction of the outer third of vaginal barrel causing sexual penetration almost impossible. It is generally classified under sexual pain disorder (SPD). In Diagnostic and Statistical Manual, 5th edition (DSM-5), it is classified under the new rubric of Genito-Pelvic Pain/Sexual Penetration Disorder. This fear-avoidance condition poses an ongoing significant challenge to the medical and health professionals due to the very demanding needs in health care despite its unpredictable prognosis. The etiology of vaginismus is complex: through multiple biopsycho- social processes, involving bidirectional connections between pelvic-genital (local) and higher mental function (central regulation). It has robust neural and psychological-cognitive loop feedback involvement. The internal neural circuit involves an inter-play of at least two-pathway systems, i.e. both "quick threat assessment" of occipital-limbic-occipital-prefrontal-pelvic-genital; and the chronic pain pathways through the genito-spinothalamic-parietal-pre-frontal system, respectively. In this review, a neurobiology root of vaginismus is deliberated with the central role of an emotional-regulating amygdala, and other neural loop, i.e. hippocampus and neo-cortex in the core psychopathology of fear, disgust, and sexual avoidance. Many therapists view vaginismus as a neglected art-and-science which demands a better and deeper understanding on the clinico-pathological correlation to enhance an effective model for the bio-psycho-social treatment. As vaginismus has a strong presentation in psychopathology, i.e. fear of penetration, phobic avoidance, disgust, and anticipatory anxiety, we highlighted a practical psychiatric approach to the clinical management of vaginismus, based on the current core knowledge in the perspective of neuroscience.

Mathematics is one of the subjects taught in formal educational institutions, which is an important part of efforts to improve the quality of education. Students' adaptive reasoning ability in mathematics is reflected through their critical, logical, and systematic thinking, and their objective, honest, and disciplined nature in solving problems, whether in mathematics, other subjects, or daily life. This study aims to evaluate the adaptive reasoning ability in mathematics of middle school students. This research employs a qualitative approach with a descriptive method. The subjects of this study were 14 students from SMPIT Al-Irsyad Al-Islamiyyah Karawang, grouped into five levels: very high, high, medium, low, and very low. Data were collected from the tests on adaptive reasoning ability in mathematics to understand the students' reasoning abilities. Scores were given to students who answered each question based on the indicators achieved. The highest percentage was obtained on the indicator of forming conjectures, at 48.21%, while the lowest rate was 3.57% on the indicator of finding patterns in a mathematical phenomenon. Overall, the average results for each indicator were: Indicator 1 - 48.21%, Indicator 2 - 26.79%, Indicator 3 - 35.71%, Indicator 4 - 8.93%, and Indicator 5 - 3.57%. Therefore, the average percentage of student achievement in answering the adaptive reasoning ability test in mathematics was 24.64%. It can be concluded that middle school students' mastery of adaptive reasoning ability in mathematics is categorized as low.

The human brain possesses a remarkable ability to memorize information with the assistance of a specific external environment. Therefore, mimicking the human brain's environment-enhanced learning abilities in artificial electronic devices is essential but remains a considerable challenge. Here, a network of Ag nanowires with a moisture-enhanced learning ability, which can mimic long-term potentiation (LTP) synaptic plasticity at an ultralow operating voltage as low as 0.01 V, is presented. To realize a moisture-enhanced learning ability and to adjust the aggregations of Ag ions, we introduced a thin polyvinylpyrrolidone (PVP) coating layer with moisture-sensitive properties to the surfaces of the Ag nanowires of Ag ions. That Ag nanowire network was shown to exhibit, in response to the humidity of its operating environment, different learning speeds during the LTP process. In high-humidity environments, the synaptic plasticity was significantly strengthened with a higher learning speed compared with that in relatively low-humidity environments. Based on experimental and simulation results, we attribute this enhancement to the higher electric mobility of the Ag ions in the water-absorbed PVP layer. Finally, we demonstrated by simulation that the moisture-enhanced synaptic plasticity enabled the device to adjust connection weights and delivery modes based on various input patterns. The recognition rate of a handwritten data set reached 94.5% with fewer epochs in a high-humidity environment. This work shows the feasibility of building our electronic device to achieve artificial adaptive learning abilities.

This paper introduces admission control models for accepting or denying requests to differentiated media streaming services to avoid overload and to use multiple scarce server resources efficiently. The decision models are based on deterministic and stochastic mathematical programming techniques and use a number of heuristics to determine model parameters at run-time. In contrast to existing admission control approaches, the models address the problem of allocating multiple scarce resources, allow for fine grained service differentiation, and for adaptive admission control as demand forecasts are updated on every incoming service request and feedback is used regarding current resource utilization levels to dynamically adjust admission control policies. Experiments are described to illustrate the benefits of using the adaptive admission control models over static admission control rules.

Edge computing servers have been widely deployed in recent years to address the requirements of diverse tasks that are sensitive to delays and computationally intensive. However, due to their independent nature and uneven distribution of service requests, certain clusters may be relatively idle, while others may be overloaded. This situation can result in increased latency for certain tasks, and it prevents the full utilization of resources in the edge clusters. To mitigate this problem, we design and implement a prototype testbed for task offloading, aimed at achieving computing and network convergence. This testbed facilitates collaboration among multiple edge computing clusters. We construct multiple clusters using Intel NUC mini computers and incorporate key enabling technologies into the system. We assess the testbed's performance by employing multiple video processing services that require low latency and high computational capacity. In scenarios with uneven service requests, load balancing can be achieved across the edge computing clusters, resulting in reduced response latency for user tasks.

Given the difficulty in hand-coding task schemes, an intellectualized architecture of the autonomous micro-mobile robot based-behavior for fault-repair was presented. Integrating the reinforcement learning and the group behavior evolution simulating the human’s learning and evolution, the autonomous micro-mobile robot will automatically generate the suited actions satisfied the environment. However, the designer only devises some basic behaviors, which decreases the workload of the designer and cognitive deficiency of the robot to the environment. The results of simulation have shown that the architecture endows micro robot with the ability of learning, adaptation and robustness, also with the ability of accomplishing the given task.

With the rapid development of internet technology, the field of education is undergoing profound changes. As the main force of future society, college students' adaptive learning ability is particularly important. This article aims to explore how to enhance college students' adaptive learning ability by combining large-scale models and knowledge graph technologies. By constructing a knowledge graph system enhanced by large-scale models, it can be achieved in-depth mining of learning content and personalized recommendations, providing college students with more accurate and efficient learning paths, thereby cultivating their ability for autonomous and lifelong learning.

Due to the explosive growth of the Internet and increasing service demands from all around the world, the cluster-based system that consists of one request-dispatching server and several request-handling servers has become a cost-effective way to serve the huge amount of service demands. Nowadays, Web servers have to handle more complex types of requests since requests from clients may be mixed with dynamic Web pages, database processing, or multimedia stream data. Therefore, a Web cluster should be designed with intelligent request dispatching policies for supporting various types of service requests.

Wireless networks manage performance by adjusting the contention window, as they cannot directly detect collisions. Traditional contention window adjustment algorithms, such as the Binary Exponential Backoff (BEB) algorithm, may lead to lower throughput when multiple services with varying bandwidth demands coexist. To address this issue, this study aims to enhance network throughput by enabling differentiated bandwidth allocation for various services. Using deep reinforcement learning, the state space, action space, and reward functions are defined to optimize this differentiation. These definitions are integrated into the Deep Deterministic Policy Gradient (DDPG) technique, implemented in the Access Point (AP) to intelligently adjust the contention window. Leveraging DDPG’s capability for continuous actions, the proposed method provides Quality of Service (QoS) differentiation, ensuring that each service at its respective priority level meets its transmission requirements. Compared to the BEB algorithm, the proposed approach offers improved traffic allocation and higher network bandwidth utilization.

By introducing the notion of constraint-oriented fuzzy inference, we will show that it provides us ways of fuzzy control methods that has abilities of adaptation, learning and self-organization. The basic supporting techniques behind these abilities are “hard” processing by Artificial Intelligence or traditional computational framework and “soft” processing by Neural Network, Genetic Algorithm and Reinforcement Learning techniques. In the former processing, Qualitative Reasoning and Instance Generalization by Symbolic Reasoning play important role, while by the latter processing, fuzzy control becomes capable of learning, adaptation and evolutional self-organization.