Network QoS, fairness and resource sharing control are not completely solved problems. Available solutions lack scalability due to maintaining flow state, require re-tuning if traffic changes, focus on a limited set of networking scenarios or require complex, centralized controllers and feedback loops. In this paper, we propose a core-stateless solution for closed network domains like access, enterprise and data center networks that handles resource sharing and provides guarantees for per-hop latency, independently. The proposed method enables controlled resource sharing by encoding the utility function of flows to Packet Value markings. This allows expressing resource sharing policies for all possible congestion situations, while operation is completely flow unaware inside the network. In addition, it also satisfies per-hop delay requirements for traffic flows independently. The separation of the delay requirements of the packets from their importance has not generally been possible by existing methods so far. The performance of the proposed method has thoroughly been analyzed by large number of simulations covering both static and dynamic scenarios and was implemented in a cloud-native virtual router implementing all the policies needed for a Broadband Network Gateway, showing good performance and better scalability than existing weighted queuing-based solutions.

Network QoS, fairness and resource sharing control are open challenges of network slicing and virtualization in 5G and future networks providing ultra-high speed Internet access. Traditional stateful solutions either employ the one-size-fits-all approach to provide services to end-users, regardless of the requirements of vertical services, or require real-time network monitoring and complex feedback loops for ensuring appropriate resource allocation at any time. In the past years, different core-stateless resource sharing solutions as scalable alternatives to traditional stateful approaches have recently emerged for closed networking domains like access, enterprise and data center networks. In this paper, we extend the core-stateless Per Packet Value (PPV) framework with the support of resource sharing policies across multiple layers of virtualization where policies defined by both physical and virtual network operators should be taken into account at the same time. To this end, we propose a re-marking mechanism that re-calculates packet markings during the transition from one virtualization layer to another, ensuring the desired resource sharing among end-users, network slices and physical networks without the need of real-time monitoring and complex feedback loops in the network core. To the best of our knowledge, this is the first core-stateless proposal that offers resource sharing for multiple layers of virtualization.

Governing how network resources are shared among various traffic flows is not a completely solved problem. The ideal system can implement a wide variety of detailed and flexible policies; enforces those policies under all possible offered traffic combinations and scales well with the number of flows. In this paper we introduce a framework that approximates the ideal system well. It consists of flow-aware Packet Value marking at the edges and flow-unaware operation inside the network. We show that inside the network relatively simple scheduling and AQM algorithms operating solely on Packet Value are suitable to implement the policies set at the network edge. We also discuss, how the framework can be used to support resource sharing over radio networks (where the cost of transmission varies) and how it enables load balancing, resource balancing and optimal network-wide load distribution.

In this paper we introduce a novel framework for traffic identification that employs machine learning techniques focusing on the estimation of multiple traffic influencing factors. The effect of these factors is handled with the training of several machine learning models. We utilize the outcome of the multiple models via a recombination algorithm to achieve high overall true positive and true negative and low overall false positive and false negative classification ratio. The proposed method can improve the performance of every kind of machine learning based traffic identification engine making them capable of efficient operation in changing network environment i.e., when the probing node is trained and tested in different sites.

Experimental system of IP/sup 2/ mobility management was designed and implemented in order to confirm feasibility of the protocol and identify its bottlenecks. The experimental system is based on a prototype implementation of IP/sup 2/ mobility management which runs on PC servers. Series of experiments were conducted in order to check validity of protocol sequence and impact of load status on system performance. From the feasibility tests results, it was confirmed that there was no fundamental error in IP/sup 2/ mobility management and it could interwork with other IP protocols. Throughout the bottleneck tests, degradation in performance due to limitation of the transport: mechanism was observed when the MN had relatively large number of active peers. Experimental results indicated that alternate mechanism to improve reliability of signalling messages was highly needed. From user-plane viewpoint, additional processing that is specific to IP/sup 2/ mobility management put negligible effect on the packet forwarding delay of the AR.

This paper proposes a middleware that creates a dynamically changeable, but uniform computing environment for personal area networks (PANs). The middleware hides the physical scatteredness and the dynamically changing configuration of the PAN and presents its capabilities as a single computer. Using the uniform application programming interface (API) offered by the middleware, developers creating distributed applications do not have to take care of various PAN configurations or PAN dynamics. The solution provides easy set-up of PANs and transparent redirection of ongoing dataflows.

The IPv4+4 architecture presented provides an evolutionary approach to the extension of IPv4 address space. It builds on the existence of network address translators (NATs) and private address realms. During and after the transition from IPv4 to IPv4+4 there is no need to change routers, only NATs and end-hosts. End-to-end address transparency is regained incrementally as deployment progresses. The transition process is simple and it provides incentives for networks with both private and public addresses to upgrade and increase transparent reachability. Early implementation experiences are also included.

In our present work we model the maximum throughput of stationary ad hoc networks. The routing process we apply is ideal in the sense that it includes collision avoidance and transmits using shortest paths between random source-destination pairs. Our model not only confirms the n-1/2 decay of the throughput that was published earlier in the literature, but also provides an approach to acquire the function relations between various network parameters. This way the effects of changes in surface topology occupied by the network, traffic generation algorithm or other network characteristics can be predicted using first order approximation

The emergence of Bluetooth as a default radio interface allows handheld electronic devices to be rapidly interconnected into ad hoc networks. Bluetooth allows large numbers of piconets to form a scatternet using designated nodes that participate in multiple piconets. We study the performance implications of forming scatternets from piconets. The contribution of our work is twofold. First, we establish a network model and define performance metrics for Bluetooth scatternets. Our model is derived from constraints specific to the Bluetooth technology, but is sufficiently abstract to relate to the more general field of ad hoc networking. Second, using a number of simulation studies, we relate scatternet parameters to performance metrics and discover correlations between scatternet formation rules and performance. These results reveal some important performance implications of scatternet design decisions and can serve as guidelines for future scatternet formation algorithms.

In this paper we consider an output port of an ATM switch, where cell streams belonging to different TCP and UDP sessions arrive. During congestion the switch applies early packet discard (EDP) in order to reduce bandwidth waste. In order to guarantee fairness in the sense that only misbehaving sources get affected by the packet drop mechanism and to ensure high server utilization we propose an algorithm which also features simplicity. A salient feature of this algorithm is that it allows a predefined share /spl alpha//sub i/ to be associated with stream i. The algorithm (which we call the weighted fair EPD, WFEPD) attempts to provide this weighted share of the bandwidth for the streams in the long time average. A sliding window implementation of WFEPD allows us to demonstrate its efficiency in terms of fairness and bandwidth utilization.