Network resources (such as bandwidth on a link) are not unlimited, and must be shared by all networked applications in some manner of fairness. This calls for the development and implementation of effective strategies that enable optimal utilization of these scarce network resources among the various applications that share the network. Although several rate controllers have been proposed in the literature to address the issue of optimal rate allocation, they do not appear to capture other factors that are of critical concern. For example, consider a battlefield data fusion application where a fusion center desires to allocate more bandwidth to incoming flows that are perceived to be more accurate and important. For these applications, network users should consider transmission rates of other users in the process of rate allocation. Hence, a rate controller should consider application specific rate coordination directives given by the underlying application. The work reported herein addresses this issue of how a rate controller may establish and maintain the desired application specific rate coordination directives. We identify three major challenges in meeting this objective. First, the application specific performance measures must be formulated as rate coordination directives. Second, it is necessary to incorporate these rate coordination directives into a rate controller. Of course, the resulting rate controller must co-exist with ordinary rate controllers, such as TCP Reno, in a shared network. Finally, a mechanism for identifying those flows that require the rate allocation directives must be put in place. The first challenge is addressed by means of a utility function which allows the performance of the underlying application to be maximized. The second challenge is addressed by utilizing the Network Utility Maximization (NUM) framework. The standard utility function (i.e. utility function of the standard rate controller) is augmented by inserting the application specific utility function as an additive term. Then the rate allocation problem is formulated as a constrained optimization problem, where the objective is to maximize the aggregate utility of the network. The gradient projection algorithm is used to solve the optimization problem. The resulting solution is formulated and implemented as a window update function. To address the final challenge we resort to a machine learning algorithm. We demonstrate how data features estimated utilizing only a fraction of the flow can be used as evidential input to a series of Bayesian Networks (BNs). We account for the uncertainty introduced by partial flow data through the Dempster-Shafer (DS) evidential reasoning framework.

This thesis presents two modern control methods to address the Internet traffic congestion control issues. They are based on a distributed traffic management framework for the fast-growing Internet traffic in which routers are deployed with intelligent or optimal data rate controllers to tackle the traffic mass. The first one is called the IntelRate (Intelligent Rate) controller using the fuzzy logic theory. Unlike other explicit traffic control protocols that have to estimate network parameters (e.g., link latency, bottleneck bandwidth, packet loss rate, or the number of flows), our fuzzy-logic-based explicit controller can measure the router queue size directly. Hence it avoids various potential performance problems arising from parameter estimations while reducing much computation and memory consumption in the routers. The communication QoS (Quality of Service) is assured by the good performances of our scheme such as max-min fairness, low queueing delay and good robustness to network dynamics. Using the Lyapunov’s Direct Method, this controller is proved to be globally asymptotically stable. The other one is called the OFEX (Optimal and Fully EXplicit) controller using convex optimization. This new scheme is able to provide not only optimal bandwidth allocation but also fully explicit congestion signal to sources. It uses the congestion signal from the most congested link, instead of the cumulative signal from a flow path. In this way, it overcomes the drawback of the relatively explicit controllers that bias the multi-bottlenecked users, and significantly improves their convergence speed and throughput performance. Furthermore, the OFEX controller design considers a dynamic model by proposing a remedial measure against the unpredictable bandwidth changes in contention-based multi-access networks (such as shared Ethernet or IEEE 802.11). When compared with the former works/controllers, such a remedy also effectively reduces the instantaneous queue size in a router, and thus significantly improving the queueing delay and packet loss performance. Finally, the applications of these two controllers on wireless local area networks have been investigated. Their design guidelines/limits are also provided based on our experiences.

This thesis presents two modern control methods to address the Internet traffic congestion control issues. They are based on a distributed traffic management framework for the fast-growing Internet traffic in which routers are deployed with intelligent or optimal data rate controllers to tackle the traffic mass. The first one is called the IntelRate (Intelligent Rate) controller using the fuzzy logic theory. Unlike other explicit traffic control protocols that have to estimate network parameters (e.g., link latency, bottleneck bandwidth, packet loss rate, or the number of flows), our fuzzy-logic-based explicit controller can measure the router queue size directly. Hence it avoids various potential performance problems arising from parameter estimations while reducing much computation and memory consumption in the routers. The communication QoS (Quality of Service) is assured by the good performances of our scheme such as max-min fairness, low queueing delay and good robustness to network dynamics. Using the Lyapunov’s Direct Method, this controller is proved to be globally asymptotically stable. The other one is called the OFEX (Optimal and Fully EXplicit) controller using convex optimization. This new scheme is able to provide not only optimal bandwidth allocation but also fully explicit congestion signal to sources. It uses the congestion signal from the most congested link, instead of the cumulative signal from a flow path. In this way, it overcomes the drawback of the relatively explicit controllers that bias the multi-bottlenecked users, and significantly improves their convergence speed and throughput performance. Furthermore, the OFEX controller design considers a dynamic model by proposing a remedial measure against the unpredictable bandwidth changes in contention-based multi-access networks (such as shared Ethernet or IEEE 802.11). When compared with the former works/controllers, such a remedy also effectively reduces the instantaneous queue size in a router, and thus significantly improving the queueing delay and packet loss performance. Finally, the applications of these two controllers on wireless local area networks have been investigated. Their design guidelines/limits are also provided based on our experiences.

Network utility maximization has emerged as a powerful tool in studying flow control, resource allocation and other cross-layer optimization problems. In this work, we study a flow control problem in the optimization framework. The objective is to maximize the sum utility of the users subject to the flow constraints of the network. The utility maximization is solved in a distributed setting; the network operator does not know the user utility functions and the users know neither the rate choices of other users nor the flow constraints of the network. We build upon a popular decomposition technique proposed by Kelly [Eur. Trans. Telecommun., 8(1), 1997] to solve the utility maximization problem in the aforementioned distributed setting. The technique decomposes the utility maximization problem into a user problem, solved by each user and a network problem solved by the network. We propose an iterative algorithm based on this decomposition technique. In each iteration, the users communicate to the network their willingness to pay for the network resources. The network allocates rates in a proportionally fair manner based on the prices communicated by the users. The new feature of the proposed algorithm is that the rates allocated by the network remains feasible at all times. We show that the iterates put out by the algorithm asymptotically tracks a differential inclusion. We also show that the solution to the differential inclusion converges to the system optimal point via Lyapunov theory. We use a popular benchmark algorithm due to Kelly et al. [J. of the Oper. Res. Soc., 49(3), 1998] that involves fast user updates coupled with slow network updates in the form of additive increase and multiplicative decrease of the user flows. The proposed algorithm may be viewed as one with fast user update and fast network update that keeps the iterates feasible at all times. Simulations suggest that our proposed algorithm converges faster than the aforementioned benchmark algorithm. When the flows originate or terminate at a single node, the network problem is the maximization of a so-called d-separable objective function over the bases of a polymatroid. The solution is the lexicographically optimal base of the polymatroid. We map the problem of finding the lexicographically optimal base of a polymatroid to the geometrical problem of finding the concave cover of a set of points on a two-dimensional plane. We also describe an algorithm that finds the concave cover in linear time. Next, we consider the minimization of a more general objective function, i.e., a separable convex function, over the bases of a polymatroid with a special structure. We propose a novel decomposition algorithm and show the proof of correctness and optimality of the algorithm via the theory of polymatroids. Further, motivated by the need to handle piece-wise linear concave utility functions, we extend the decomposition algorithm to handle the case when the separable convex functions are not continuously differentiable or not strictly convex. We then provide a proof of its correctness and optimality.

We present a novel bilevel programming formulation that aims to solve a resource allocation problem for request-response systems. Our formulation is motivated by potential inefficiencies in the allocation of computational resources to incoming user requests in such systems. In our experience, systems often operate with a surplus of resources despite potentially incurring unjustifiable cost. Our work attempts to optimize the tradeoff between the financial cost of resources and the opportunity cost of unfulfilled user demand. Our bilevel formulation consists of an \textit{upper} problem which has a constraint value appearing in the \textit{lower} problem. We derive efficient methods for finding global solutions to the upper problem in two settings; first with logarithmic utility functions, and then with a particular type of sigmoidal utility function. A solution to the model we describe (1) determines the optimal number of total resources to allocate and (2) determines the optimal distribution of such resources across the set of user requests.

Providing end-to-end quality-assured services implies many challenges, which go beyond technical ones, involving as well economic and even cultural or political issues. In this thesis we first focus on a technical problem and then intent a more holistic regard to the whole problem, considering at the same time Network Service Providers (NSPs), stakeholders and buyers' behaviour and satisfaction. One of the most important problems when deploying interdomain path selection with Quality of Service (QoS) requirements is being able to rely the computations on metrics that hold for a long period of time. Our proposal for solving that problem is to compute bounds on the metrics, taking into account the uncertainty on the traffic demands. We then move to a NSP-alliance scenario, where we propose a complete framework for selling interdomain quality-assured services, and subsequently distributing revenues. At the end of the thesis we adopt a more holistic approach and consider the interactions with the monitoring plane and the buyers' behaviour. We propose a simple pricing scheme and study it in detail, in order to use QoS monitoring information as feedback to the business plane, with the ultimate objective of improving the seller's revenue.

<p>It is predicted that there would be 41.6 billion IoT devices by 2025, which has kindled new interests on the timing coordination between sensors and controllers, i.e., how to use the waiting time to improve the performance. Sun et al. showed that a <i>controller</i> can strictly improve the data freshness, the so-called Age-of-Information (AoI), via careful scheduling designs. The optimal waiting policy for the <i>sensor</i> side was later characterized in the context of remote estimation. The first part of this work develops the jointly optimal sensor/controller waiting policy. It generalizes the above two important results in that not only do we consider joint sensor/controller designs, but we also assume random delay in both the forward and feedback directions. </p> <p> </p> <p>The second part of the work revisits and significantly strengthens the seminal results of Sun et al on the following fronts: (i) When designing the optimal offline schemes with full knowledge of the delay distributions, a new <i>fixed-point-based</i> method is proposed with <i>quadratic convergence rate</i>; (ii) When the distributional knowledge is unavailable, two new low-complexity online algorithms are proposed, which provably attain the optimal average AoI penalty; and (iii) the online schemes also admit a modular architecture, which allows the designer to <i>upgrade</i> certain components to handle additional practical challenges. Two such upgrades are proposed: (iii.1) the AoI penalty function incurred at the destination is unknown to the source node and must also be estimated on the fly, and (iii.2) the unknown delay distribution is Markovian instead of i.i.d. </p> <p> </p> <p>With the exponential growth of interconnected IoT devices and the increasing risk of excessive resource consumption in mind, the third part of this work derives an optimal joint cost-and-AoI minimization solution for multiple coexisting source-destination (S-D) pairs. The results admit a new <i>AoI-market-price</i>-based interpretation and are applicable to the setting of (i) general heterogeneous AoI penalty functions and Markov delay distributions for each S-D pair, and (ii) a general network cost function of aggregate throughput of all S-D pairs. </p> <p> </p> <p>In each part of this work, extensive simulation is used to demonstrate the superior performance of the proposed schemes. The discussion on analytical as well as numerical results sheds some light on designing practical network utility maximization protocols.</p>