Patrick-Benjamin Bok

Context-Aware QoS Control for Wireless Mesh Networks of UAVs

Unmanned Aerial Vehicles (UAVs) can be used in a wide application range. For example, UAVs are utilized to observe critical areas in disaster situations without jeopardizing individuals or to extend the transmission range of communication networks. Connecting a set of UAVs with each other within a Wireless Mesh Network helps to increase the coverage of the observable area. Due to their limited performance and energy resources and, thereby, restricted communication capabilities, a tailored QoS management scheme has to be used to optimize occurring data flows in these networks. We present a QoS control scheme that works on the basis of process-patterns, each describing the context-dependent behavior of UAVs according to the execution order of services for different situations. Furthermore, the logical communication path is optimized within the mesh network based on each nodes areal position using a dynamic hierarchical communication structure. Thereby, performance and fairness within a network of UAVs can be increased demand-actuated.

A business process aware semantic QoS provisioning scheme

The execution of business processes is supported by running many applications within a corporate network. Each business process includes several tasks which have different priorities expressing each tasks relevance in helping to achieve the related business objectives. The provisioning of a certain level of QoS according to the requirements of an entire business process can hardly be accomplished using existing QoS provisioning schemes because these do not account for the dynamic requirements introduced by business processes. The definition of a certain level of QoS using the existing models is just driven by technical aspects of the running applications. In this paper we present a novel business process aware semantic QoS provisioning scheme that accounts for the dynamic requirements and permits application-independent usage. In the business world the priority of a task depends on the current context it is executed in. So equal tasks may need different conditions of QoS in different situations, which can be characterized by the tasks that ran before or run in parallel. Therefore, the semantic behavior is based on the entire business processes as input for the QoS provisioning process. With the proposed QoS provisioning scheme a fair application conditioning can be achieved that is much closer to the business requirements on QoS. For the validation of the scheme, it has been implemented as a QoS-driver in different operating systems.

Trust-based resource allocation and evaluation of workflows in distributed computing environments

An optimal resource allocation is desired to increase the efficiency of systems processing business or scientific workflows. These systems include the processing of workflows in distributed computing environments such as grid or cloud computing. Current approaches used in this systems consider QoS-requirements as quality, speed or costs in the resource allocation process and just select the resources that satisfy these requirements. However, the selection of a resource should not be based solely on whether a resource meets the QoS-requirements or not, because it does not imply that a task is processed satisfactory by the selected resource according to its content. To evaluate resources regarding this aspect, trust models can be used. But current trust models are not specifically designed to model trust in distributed computing environments, hence handling the dynamics of business or scientific workflows that constantly change their requirements and conditions. In this paper, we present approaches which improve current trust models according to the problems mentioned. Thereby, the degree of automation in the trust evaluation process will be increased as well. Finally, in combination with an adjusted trust level workflow, the presented approach allows an optimal resource allocation for grid or cloud computing service providers.

A distributed fair congestion avoidance protocol

Congestion is a particular problem in computer networks because all nodes compete with each other for the available resources. If congestion occurs, packets cannot be delivered anymore and the achieved data rate decreases significantly for all nodes. Different approaches for congestion avoidance and control exist. Nevertheless, a new proactive approach that works independent of the congestion control or avoidance capabilities of Transport Layer protocols and the available resources may be superior. In this paper, we present DiFCAP, a proactive Distributed Fair Congestion Avoidance Protocol that works independent of the congestion handling capabilities of Transport Layer protocols and the available resources. DiFCAP uses a transmission notification mechanism that leads to fair throttling of outgoing transmissions of all nodes within a predefined group. Thereby, congestion can be partially avoided and the packet drop rate can be reduced. The performance improvements that can be achieved using DiFCAP are shown through simulations.

Dynamic Link Classification Based on Neuronal Networks for QoS Enabled Access to Limited Resources

In this paper, the authors present a study of a network fingerprinting classification using monotone multilayer perceptron neuronal networks. It is part of an overall performance engineering approach. The classification is used to increase the performance of an active queue management on the quality of service for a next generation public safety communication system based on an IP overlay network. This network combines heterogeneous communication networks and technologies to increase the overall systems performance. Public safety users have higher requirements regarding coverage, data rates and quality of service than standard commercial ones. Main challenge for this study is the optimization of the overall system for voice group communication, which is still the most important communication within public safety scenarios. This paper shows that with the given parametrization, an ensemble of multi-layer perceptrons gives a satisfactory classification probability, if a setup of three technologies (EDGE, UMTS and LTE) is assumed to be in usage as communication technologies. This setup is practicable enough to have a chance to be implemented in a future system.

Evaluating the Impact of Transmission Power on QoS in Wireless Mesh Networks

Wireless interference is a natural problem of single radio Wireless Mesh Networks (WMN) that operate on a shared medium. Noise induced by simultaneously sending nodes within the same transmission range leads to less throughput and a higher latency. These are critical for the overall quality of a network and regulate the level of achieved QoS. Reducing the transmission power implies less interference on the medium, but also less robust links. Therefore finding an optimal transmission power level may be one solution to increase the efficiency of single radio WMN. We analyze the impact of transmission power control (TPC) on wireless interference within a real testbed environment, and show that transmission power control leads to a better throughput and latency.

I-DWRR — An insolvency enabled scheduling scheme extending Deficit Weighted Round Robin

The allocation of network resources to flows within a computer network is performed using scheduling algorithms with the aim of enabling every flow to receive the required resources to perform well. One of several algorithms proposed in the last decades is the well-known Deficit (Weighted) Round Robin (D(W)RR) scheduling algorithm proposed by Shreedhar and Varghese [1]. It has a low complexity, but at the same time allows fair and weighted sharing of limited resources. Nevertheless, DWRR does not work deadline compliant and active handling of nearly over-delayed packets is not considered. In this paper, we present a novel insolvency enabled extension of the DWRR scheduling algorithm called I-DWRR. It reduces the number of deadline violations by preventing them using an innovative mechanism called Queue Insolvency (QI). The proposed extension provides a lower packet-loss rate for multi-queue scenarios in comparison to DWRR while keeping the occurring latency equal to DWRR. The algorithm and its benefits have been validated through a broad simulation study.

On the performance of host-based business process aware QoS management

In general, network traffic is classified and marked at the edge- or ingress-routers of a transit network to permit different levels of QoS when using the DiffServ architecture. Thus, the core-nodes of a network are able to handle the traffic accordingly. Problems in the process of identification of traffic flows and their associated applications, for example because of encryption, avoid the classification of network traffic. This can be improved by relocating the functions of classifying and marking towards the network traffic generating hosts. The great benefit of the relocation is that applications can be identified in any case. In addition, pre-shaping of network traffic becomes possible. Unfortunately, the set of rules for classification and marking on hosts is always static in its behavior so that it would not be possible to map dynamics, for example, of business processes to the set of rules or its interpreter. Therefore, it has to be possible that the marking of packets depends on their temporal execution context and, thereby, on the business processes. In this paper, we present an efficient solution that considers business processes in host-based QoS management systems. The applied QoS management system considers the dynamic requirements of business processes and works application-independent. The performance of the system, introduced by its efficient design, is analyzed and its benefits are discussed.

Distributed Flow Permission Inspection for Mission-Critical Communication of Untrusted Autonomous Vehicles

Swarms of Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) are used for cooperative sensing, clarification and communication relaying in disaster situations and, thereby, allow to reduce jeopardizing the health of rescue personnel. Having meshed interconnection, a swarm maintains a communication network dedicated for applications with context-dependent performance requirements that require reliable traffic classification and conditioning. Nevertheless, these systems are vulnerable because UAVs/UGVs may act incorrectly resulting in an erroneous network behavior which may affect the performance of mission-critical flows. This is a major problem if a couple of untrusted UAVs and UGVs from different rescue organizations are connected. In this paper, the reliability of traffic conditioning within untrusted swarms is enhanced by the Distributed Flow Permission Inspection (DiFPIN) scheme. DiFPIN is used to verify the flow classification to circumvent the usage of erroneous entities on the transmission path. Thereby, the negative impairment of mission-critical flows can be reduced.

ADeM: Active Delay Management for critical group communication over heterogeneous public cellular networks

As the basis of Next Generation Public Safety Communication (PSC) systems for critical group communication, 4G (LTE) technologies have been chosen from the different stakeholder. Existing 2G and 3G equivalent technologies are and will be used to allow for a sufficient network coverage. Due to the long investment cycles within the public sector, for a long time small band 2G Professional Mobile Radio (PMR) networks will be the backbone of mission critical voice communication. To enable enhanced data exchange 4G networks will provide this capability as an add-on. It is a well known fact that every transmission technology has its own footprint of delay and delay-spread (jitter). This lead to challenges within heterogeneous system environments, because different behavior introduces unfairness into the system, negligent users in legacy networks. The proposed solution in this paper is based on the identification of the unique footprint of the used networks to increase the fairness. In this paper, the authors present an Active Delay Management (ADeM) PSC System that allows for critical group communication in next generation public safety communication network based on ×G technologies. Therefore, the performance limitations of an IP overlay for heterogeneous communication networks have been studied and analyzed for different mobile radio technologies, based on mobile field trials performed in commercial 2G, 3G and 4G networks. Build upon this study, the gathered information is used to classify the transmission networks to enable the proposed ADeM system to actively consider network dependent delays in its Active Queue Management (AQM) decisions. Thereby, enhanced fairness for the different information streams as well as an improved performance in critical communication scenarios can be achieved.