Wireless systems of today face the dual challenge of both supporting large traffic flows and providing reliable quality of service to different delay-sensitive applications. For such applications, it is essential to derive meaningful performance measures such as queue-length distribution and packet loss probability, while providing service guarantees. The concepts of effective bandwidth and effective capacity offer a powerful cross-layer approach that provides suitable performance metrics for the bandwidth and capacity of wireless channels supporting delay-sensitive traffic. Many wireless systems rely on multihop forwarding to reach destinations outside the direct range of the source. This work extends part of the methodology available for the design of wireless systems to the multihop paradigm. It describes the analysis of a communication system with two hops using this cross-layer approach. A framework is developed to study the interplay between the allocation of physical resources across the wireless hops and overall service quality as defined by a queueing criterion based on large deviations. Decoupling techniques introduce simple ways of analyzing the queues independently. Numerical analysis helps identify fundamental performance limits for Rayleigh block fading wireless channel models with independent and identically distributed blocks. Simulation studies present comparable results akin to that obtained using the analytical framework. These results suggest that it is imperative to account for queueing aspects while analyzing delay-sensitive wireless communication systems.

The objective of this research is to develop distributed approaches to optimizing network traffic. Two problems are studied, which include exploiting social networks in routing packets (coupons) to desired network nodes (users in the social network), and developing a rate based transport protocol, which will guarantee that all the flows in a network (e.g. Internet) meet a delay constraint per packet. Firstly, we will study social networks as a means of obtaining information about a system. They are increasingly seen as a means of obtaining awareness of user preferences. Such awareness could be used to target goods and services at them. We consider a general user model, wherein users could buy different numbers of goods at a marked and at a discounted price. Our first objective is to learn which users would be interested in a particular good. Second, we would like to know how much to discount these users such that the entire demand is realized, but not so much that profits are decreased. We develop algorithms for multihop forwarding of such discount coupons over an online social network, in which users forward coupons to each other in return for a reward. Coupling this idea with the implicit learning associated with backpressure routing (originally developed for multihop wireless networks), we would like to demonstrate how to realize optimal revenue. We will then propose a simpler heuristic algorithm and try to show, using simulations, that its performance approaches that of backpressure routing. As the second problem, we look at the traditional formulation of the total value of information transfer, which is a multi-commodity flow problem. Here, each data source is seen as generating a commodity along a fixed route, and the objective is to maximize the total system throughput under some concept of fairness, subject to capacity constraints of the links used. This problem is well studied under the framework of network utility maximization and has led to several different distributed congestion control schemes. However, this idea of value does not capture the fact that flows might associate value, not just with throughput, but with link-quality metrics such as packet delay, jitter and so on. The traditional congestion control problem is redefined to include individual source preferences. It is assumed that degradation in link quality seen by a flow adds up on the links it traverses, and the total utility is maximized in such a way that the quality degradation seen by each source is bounded by a value that it declares. Decoupling source-dissatisfaction and link-degradation through an ?effective capacity? variable, a distributed and provably optimal resource allocation algorithm is designed, to maximize system utility subject to these quality constraints. The applicability of our controller in different situations is illustrated, and results are supported through numerical examples.

Data intensive applications, requiring reliability and strict delay constraints, have emerged recently and they necessitate a different approach to analyzing system performance. In my work, I establish a framework that relates physical channel parameters to the queueing performance for a single-user wireless system. I then seek to assess the potential benefits of multirate techniques, such as hybrid-ARQ (Automatic Repeat reQuest), in the context of delay-sensitive communications. Present methods of analysis in an information theoretic paradigm define capacity assuming that long codewords can be used to take advantage of the ergodic properties of the fading wireless channel. This definition provides only a limited characterization of the channel in the light of delay constraints. The assumption of independent and identically distributed channel realizations tends to over-estimate the system performance by not considering the inherent time correlation. A finite-state continuous time Markov channel model that I formulate enables me to partition the instantaneous data-rate received at the destination into a finite number of states, representing layers in a hybrid-ARQ scheme. The correlation of channel has been incorporated through level crossing rates as transition rates in the Markov model. The large deviation principle governing the buffer overflow of the Markov model, is very sensitive to channel memory, is tractable, and gives a good estimate of the system performance. Metrics such as effective capacity and probability of buffer overflow, that are obtained through large deviations have been related to the wireless physical layer parameters through the model. Using the above metrics under QoS constraints, I establish the quantitative performance advantage of using hybrid-ARQ over traditional systems. I conduct this inquiry by restricting attention to the case where the expected transmit power is fixed at the transmitter. The results show that hybrid-ARQ helps us in obtaining higher effective capacity, but it is very difficult to support delay sensitive communication over wireless channel in the absence of channel knowledge and dynamic power allocation strategies.

This dissertation addresses some of the most important issues in delay-sensitive communication over wireless systems and networks. Traditionally, the design of communication networks adopts a layered framework where each layer serves as a “black box” abstraction for higher layers. However, in the context of wireless networks with delay-sensitive applications such as Voice over Internet Protocol (VoIP), on-line gaming, and video conferencing, this layered architecture does not offer a complete picture. For example, an information theoretic perspective on the physical layer typically ignores the bursty nature of practical sources and often overlooks the role of delay in service quality. The purpose of this dissertation is to take on a cross-disciplinary approach to derive new fundamental limits on the performance, in terms of capacity and delay, of wireless systems and to apply these limits to the design of practical wireless systems that support delay-sensitive applications. To realize this goal, we consider a number of objectives. 1. Develop an integrated methodology for the analysis of wireless systems that support delay-sensitive applications based, in part, on large deviation theory. 2. Use this methodology to identify fundamental performance limits and to design systems which allocate resources efficiently under stringent service requirements. 3. Analyze the performance of wireless communication networks that takes advantage of novel paradigms such as user cooperation, and multi-antenna systems. Based on the proposed framework, we find that delay constraints significantly influence how system resources should be allocated. Channel correlation has a major impact on the performance of wireless communication systems. Sophisticated power control based on the joint space of channel and buffer states are essential for delaysensitive communications.

This dissertation studies fundamental limits of modern digital communication systems in presence/absence of delay and secrecy constraints. In the first part of this dissertation, we consider a typical time-division wireless communication system wherein the channel strengths of the wireless users vary with time with a power constraint at the base station and which is not subject to any delay constraint. The objective is to allocate resources to the wireless users in an equitable manner so as to achieve a specific throughput. This problem has been looked at in different ways by previous researchers. We address this problem by developing a systematic way of designing scheduling schemes that can achieve any point on the boundary of the rate region. This allows us to map a desired throughput to a specific scheduling scheme which can then be used to service the wireless users. We then propose a simple scheme by which users can cooperate and then show that a cooperative scheduling scheme enlarges the achievable rate region. A simple iterative algorithm is proposed to find the resource allocation parameters and the scheduling scheme for the cooperative system. In the second part of the dissertation, a downlink time-division wireless sys- tem that is subject to a delay constraint is studied, and the rate region and optimal scheduling schemes are derived. The result of this study concludes that the achievable throughput of users decrease as the delay constraint is increased. Next, we consider a problem motivated by cognitive radio applications which has been proposed as a means to implement efficient reuse of the licensed spectrum. Previous research on this topic has focussed largely on obtaining fundamental limits on achievable throughput from a physical layer perspective. In this dissertation, we study the impact of im- posing Quality of Service constraints (QoS) on the achievable throughput of users. The result of this study gives insights on how the cognitive radio system needs to be operated in the low and high QoS constraint regime. Finally, the third part of this dissertation is motivated by the need for commu- nicating information not only reliably, but also in a secure manner. To this end, we study a source coding problem, wherein multiple sources needs to be communicated to a receiver with the stipulation that there is no direct channel from the transmitter to the receiver. However, there are many \agents" that can help carry the information from the transmitter to the receiver. Depending on the reliability that the transmit- ter has on each of the agents, information is securely encoded by the transmitter and given to the agents, which will be subsequently given to the receiver. We study the overhead that the transmitter has to incur for transmitting the information to the receiver with the desired level of secrecy. The rate region for this problem is found and simple achievable schemes are proposed. The main result is that, separate secure coding of sources is optimal for achieving the sum-rate point for the general case of the problem and the rate region for simple case of this problem.

Background. Coverless steganography is a relatively unexplored area of steganography where the message is not embedded into a cover media. Instead the message is derived from one or several properties already existing in the carrier media. This renders steganalysis methods used for traditional steganography useless. Early coverless methods were applied to images or texts but more recently the possibilities in the video and audio domain have been explored. The audio domain still remains relatively unexplored however, with the earliest work being presented in 2022. In this thesis, we narrow the existing research gap by proposing an audio-compatible method which uses the timestamp that marks when a carrier media was received to generate a time-based key which can be applied to the hash produced by said carrier. This effectively allows one carrier to represent a range of different hashes depending on the timestamp specifying when it was received, increasing capacity. Objectives. The objectives of the thesis are to explore what features of audio are suitable for steganographic use, to establish a method for finding audio clips which can represent a specific message to be sent and to improve on the current state-of-the-art method, taking capacity, robustness and cost into consideration. Methods. A literature review was first conducted to gain insight on techniques used in previous works. This served both to illuminate features of audio that could be used to good effect in a coverless approach, and to identify coverless approaches which could work but had not been tested yet. Experiments were then performed on two datasets to show the effective capacity increase of the proposed method when used in tandem with the existing state-of-the-art method for coverless audio steganography. Additional robustness tests for said state-of-the-art method were also performed. Results. The results show that the proposed method could increase the per-message capacity from eight bits to 16 bits, while still retaining 100% effective capacity using only 200 key permutations, given a database consisting of 50 one-minute long audio clips. They further show that the time cost added by the proposed method is in total less than 0.1 seconds for 2048 key permutations. The robustness experiments show that the hashing algorithms used in the state-of-the-art method have high robustness against additive white gaussian noise, low-pass filters, and resampling attacks but are weaker against compression and band-pass filters.  Conclusions. We address the scientific gap and complete our objectives by proposing a method which can increase capacity of existing coverless steganography methods. We demonstrate the capacity increase our method brings by using it in tandem with the state-of-the-art method for the coverless audio domain. We argue that our method is not limited to the audio domain, or to the coverless method with which we performed our experiments. Finally, we discuss several directions for future works. / Bakgrund. Täcklös steganografi är ett relativt outforskat område inom steganografi där meddelandet, istället för att gömmas i ett medium, representeras av en eller flera egenskaper som kan erhållas från mediet. Detta faktum hindrar nuvarande steganalysmetoder från att upptäcka bruk av täcklös steganografi. Tidiga studier inom området behandlar bilder och text, senare studier har utökat området genom att behandla video och ljud. Den första studien inom täcklös ljudsteganografi publicerades år 2022. Målet med examensarbetet är att utöka forskningen med en föreslagen ljudkompatibel metod som använder tidsstämpeln då ett meddelande mottagits för att skapa en tidsbaserad nyckel som kan appliceras på en hash erhållen från ett steganografiskt medium. Detta tillåter mediet att representera olika hashar beroende på tiden, vilket ökar kapaciteten.   Syfte. Syftet med examensarbetet är att utforska vilka egenskaper i ett ljudmedia som lämpar sig åt steganografiskt bruk, att skapa en metod som kan hitta ljudklipp som representerar ett efterfrågat meddelande, samt att förbättra nuvarande state-of-the-art inom täcklös ljudsteganografi genom att finna en bra balans mellan kapacitet, robusthet och kostnad.   Metod. En litteraturstudie utfördes för att få förståelse för metoder använda i tidigare studier. Syftet var att hitta egenskaper i ljud som lämpar sig åt täcklös ljudsteganografi samt identifiera icke-täcklösa metoder som skulle kunna anpassas för att fungera som täcklösa. Experiment utfördes sedan på två dataset för att påvisa den ökning i effektiv kapacitet den föreslagna metoden ger när den appliceras på state-of-the-art-metoden inom täcklös ljudsteganografi. Experiment utfördes även för att utöka tidigare forskning på robustheten av state-of-the-art-metoden inom täcklös ljudsteganografi. Resultat. Resultaten visar att den föreslagna metoden kan öka kapaciteten per meddelande från åtta till 16 bits med 100% effektiv kapacitet med 200 nyckelpermutationer och en databas bestående av 50 stycken en-minut långa ljudklipp. De visar även att tidskostnaden för den föreslagna metoden är mindre än 0,1 sekund för 2048 nyckelpermutationer. Experimenten på robusthet visar att state-of-the-art-metoden har god robusthet mot additivt vitt gaussiskt brus, lågpassfilter och omsampling men är svagare mot kompression och bandpassfilter. Slutsatser. Vi fullbordar målen och utökar forskningen inom området genom att föreslå en metod kan öka kapaciteten av befintliga täcklösa metoder. Vi demonstrerar kapacitetsökningen genom att applicera vår metod på den senaste täcklösa ljudsteganografimetoden. Vi presenterar argument för vår metods tillämpning i områden utanför ljuddomänen och utanför metoden som den applicerades på. Slutligen diskuteras riktningar för framtida forskning.

Due to the highly-varying wireless channels over time, frequency, and space domains, statistical QoS provisioning, instead of deterministic QoS guarantees, has become a recognized feature in the next-generation wireless networks. In this dissertation, we study the adaptive wireless resource allocation problems for statistical QoS provisioning, such as guaranteeing the specified delay-bound violation probability, upper-bounding the average loss-rate, optimizing the average goodput/throughput, etc., in several typical types of mobile wireless networks. In the first part of this dissertation, we study the statistical QoS provisioning for mobile multicast through the adaptive resource allocations, where different multicast receivers attempt to receive the common messages from a single base-station sender over broadcast fading channels. Because of the heterogeneous fading across different multicast receivers, both instantaneously and statistically, how to design the efficient adaptive rate control and resource allocation for wireless multicast is a widely cited open problem. We first study the time-sharing based goodput-optimization problem for non-realtime multicast services. Then, to more comprehensively characterize the QoS provisioning problems for mobile multicast with diverse QoS requirements, we further integrate the statistical delay-QoS control techniques — effective capacity theory, statistical loss-rate control, and information theory to propose a QoS-driven optimization framework. Applying this framework and solving for the corresponding optimization problem, we identify the optimal tradeoff among statistical delay-QoS requirements, sustainable traffic load, and the average loss rate through the adaptive resource allocations and queue management. Furthermore, we study the adaptive resource allocation problems for multi-layer video multicast to satisfy diverse statistical delay and loss QoS requirements over different video layers. In addition, we derive the efficient adaptive erasure-correction coding scheme for the packet-level multicast, where the erasure-correction code is dynamically constructed based on multicast receivers’ packet-loss statuses, to achieve high error-control efficiency in mobile multicast networks. In the second part of this dissertation, we design the adaptive resource allocation schemes for QoS provisioning in unicast based wireless networks, with emphasis on statistical delay-QoS guarantees. First, we develop the QoS-driven time-slot and power allocation schemes for multi-user downlink transmissions (with independent messages) in cellular networks to maximize the delay-QoS-constrained sum system throughput. Second, we propose the delay-QoS-aware base-station selection schemes in distributed multiple-input-multiple-output systems. Third, we study the queueaware spectrum sensing in cognitive radio networks for statistical delay-QoS provisioning. Analyses and simulations are presented to show the advantages of our proposed schemes and the impact of delay-QoS requirements on adaptive resource allocations in various environments.