Heterogeneous Wireless Networks (HWNs) are composite networks made of different wireless access technologies, possibly with overlapping coverage. Users with multi-mode terminals in HWNs will be able to initiate connectivity in the access technology that best suits their attributes and the requirements of their applications. The true potential of HWNs, however, is only realized through allowing users to maintain their sessions when toggling from one access technology to another. Such inter-technology handoffs, called vertical handoffs, will enable users to persistently select the most appropriate network, and not just at session initiation. For operators, HWNs pave the road to higher profitability through more capable networks where the complementary advantages of individual access technologies are combined. However, the characteristics of HWNs challenge traditional arguments for designing Radio Resource Management (RRM) frameworks. Managing the resources of an access technology in an HWN independently of other networks with which its overlaid risks underutilization and resource mismanagement. The dynamic nature of user demands in HWNs also calls for RRM modules with controlled operational cost. More importantly, the unique characteristics of HWNs call for non-traditional solutions that exploit the ``complementarity" of the individual networks. In this thesis, we address these issues through proposing a framework for RRM in HWNs. Our framework comprises three key components. The first component is aimed at improving allocation policies in HWNs through joint allocation policies involving provisioning and admission control. In addition, we outline the basis for achieving robust provisioning that accommodates variability in user demands, but also in network capabilities. The second component is concerned with controlling the operational cost of RRM modules. As a case study, we choose bandwidth adaptation algorithms and optimize their performance. We also introduce the notion of stochastic triggers which enables operators to direct the operation of a RRM module based on the operator's objectives and network conditions. In the third component, we introduce a new module that exploits vertical handoffs to the benefit of network operators. Such operator motivated vertical handoffs can be utilized in instances of congestion control. They can also be used proactively to achieve long-term objectives such as load balancing or service delivery cost reduction. / Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2007-09-28 04:54:46.819

Thesis (M.S.)--University of Texas at Dallas, 2008. / Includes vita. Includes bibliographical references (leaves 39-40)

The Internet of Things (IoT) is a promising networking technology that will interconnect a plethora of heterogeneous wireless devices. To support the connectivity across a massive-scale IoT, the scarce wireless communication resources must be appropriately allocated among the IoT devices, while considering the technical challenges that arise from the unique properties of the IoT, such as device heterogeneity, strict communication requirements, and limited device capabilities in terms of computation and memory. The primary goal of this dissertation is to develop novel resource management frameworks using which resource-constrained IoT devices can operate autonomously in a dynamic environment. First, a comprehensive overview on the use of various learning techniques for wireless resource management in an IoT is provided, and potential applications for each learning framework are proposed. Moreover, to capture the heterogeneity among IoT devices, a framework based on cognitive hierarchy theory is discussed, and its implementation with learning techniques of different complexities for IoT devices with varying capabilities is analyzed. Next, the problem of dynamic, distributed resource allocation in an IoT is studied when there are heterogeneous messages. Particularly, a novel finite memory multi-state sequential learning is proposed to enable diverse IoT devices to reallocate the limited communication resources in a self-organizing manner to satisfy the delay requirement of critical messages, while minimally affecting the delay-tolerant messages. The proposed learning framework is shown to be effective for the IoT devices with limited memory and observation capabilities to learn the number of critical messages. The results show that the performance of learning framework depends on memory size and observation capability of IoT devices and that the learning framework can realize low delay transmission in a massive IoT. Subsequently, the problem of one-to-one association between resource blocks and IoT devices is studied, when the IoT devices have partial information. The one-to-one association is formulated as Kolkata Paise Restaurant (KPR) game in which an IoT device tries to choose a resource block with highest gain, while avoiding duplicate selection. Moreover, a Nash equilibrium (NE) of IoT KPR game is shown to coincide with socially optimal solution. A proposed learning framework for IoT KPR game is shown to significantly increase the number of resource blocks used to successful transmit compared to a baseline. The KPR game is then extended to consider age of information (AoI), which is a metric to quantify the freshness of information in the perspective of destination. Moreover, to capture heterogeneity in an IoT, non-linear AoI is introduced. To minimize AoI, centralized and distributed approaches for the resource allocation are proposed to enable the sharing of limited communication resources, while delivering messages to the destination in a timely manner. Moreover, the proposed distributed resource allocation scheme is shown to converge to an NE and to significantly lower the average AoI compared to a baseline. Finally, the problem of dynamically partitioning the transmit power levels in non-orthogonal multiple access is studied when there are heterogeneous messages. In particular, an optimization problem is formulated to determine the number of power levels for different message types, and an estimation framework is proposed to enable the network base station to adjust power level partitioning to satisfy the performance requirements. The proposed framework is shown to effectively increase the transmission success probability compared to a baseline. Furthermore, an optimization problem is formulated to increase sum-rate and reliability by adjusting target received powers. Under different fading channels, the optimal target received powers are analyzed, and a tradeoff between reliability and sum-rate is shown. In conclusion, the theoretical and performance analysis of the frameworks proposed in this dissertation will prove essential for implementing an appropriate distributed resource allocation mechanisms for dynamic, heterogeneous IoT environments. / Doctor of Philosophy / The Internet of Things (IoT), which is a network of smart devices such as smart phones, wearable devices, smart appliances, and environment sensors, will transform many aspects of our society with numerous innovative IoT applications. Those IoT applications include interactive education, remote healthcare, smart grids, home automation, intelligent transportation, industrial monitoring, and smart agriculture. With the increasing complexity and scale of an IoT, it becomes more difficult to quickly manage the IoT devices through a cloud, and a centralized management approach may not be viable for certain IoT scenarios. Therefore, distributed solutions are needed for enabling IoT devices to fulfill their services and maintain seamless connectivity. Here, IoT device management refers to the fact that the system needs to decide which devices access the network and using which resources (e.g., frequencies). For distributed management of an IoT, the unique challenge is to allocate scarce communication resources to many IoT devices appropriately. With distributed resource management, diverse IoT devices can share the limited communication resources in a self-organizing manner. Distributed resource management overcomes the limitations of centralized resource management by satisfying strict service requirements in a massive, complex IoT. Despite the advantages and the opportunities of distributed resource management, it is necessary to address the challenges related to an IoT, such as analyzing intricate interaction of heterogeneous devices, designing viable frameworks for constrained devices, and quickly adapting to a dynamic IoT. Furthermore, distributed resource management must enable IoT devices to communicate with high reliability and low delay. In this regard, this dissertation investigates these critical IoT challenges and introduces novel distributed resource management frameworks for an IoT. In particular, the proposed frameworks are tailored to realistic IoT scenarios and consider different performance metrics. To this end, mathematical frameworks and effective algorithms are developed by significantly extending tools from wireless communication, game theory, and machine learning. The results show that the proposed distributed wireless resource management frameworks can optimize key performance metrics and meet strict communication requirements while coping with device heterogeneity, massive scale, dynamic environment, and scarce wireless resources in an IoT.

Energy consumption has become a major research topic from both environmental and economical perspectives. The telecommunications industry is currently responsible for 0.7% of the total global carbon emissions, a figure which is increasing at rapid rate. By 2020, it is desired that CO2 emissions can be reduced by 50%. Thus, reducing the energy consumption in order to lower carbon emissions and operational expenses has become a major design constraint for future communication systems. Therefore, in this thesis energy efficient resource allocation methods have been studied taking the Long Term Evolution (LTE) standard as an example. Firstly, a theoretical analysis, that shows how improvements in energy efficiency can directly be related with improvements in fairness, is provided using a Shannon theory analysis. The traditional uplink power control challenge is re-evaluated and investigated from the view point of interference mitigation rather than power minimization. Thus, a low complexity distributed resource allocation scheme for reducing the uplink co-channel interference (CCI) is presented. Improvements in energy efficiency are obtained by controlling the level of CCI affecting vulnerable mobile stations (MSs). This is done with a combined scheduler and a two layer power allocation scheme, which is based on non-cooperative game theory. Simulation results show that the proposed low complexity method provides similar performance in terms of fairness and energy efficiency when compared to a centralized signal interference noise ratio balancing scheme. Apart from using interference management techniques, by using efficiently the spare resources in the system such as bandwidth and available infrastructure, the energy expenditure in wireless networks can also be reduced. For example, during low network load periods spare resource blocks (RBs) can be allocated to mobile users for transmission in the uplink. Thereby, the user rate demands are split among its allocated RBs in order to transmit in each of them by using a simpler and more energy efficient modulation scheme. In addition, virtual Multiple-input Multiple-output (MIMO) coalitions can be formed by allowing single antenna MSs and available relay stations to cooperate between each other to obtain power savings by implementing the concepts of spatial multiplexing and spatial diversity. Resource block allocation and virtual MIMO coalition formation are modeled by a game theoretic approach derived from two different concepts of stable marriage with incomplete lists (SMI) and the college admission framework (CAF) respectively. These distributed approaches focus on optimizing the overall consumed power of the single antenna devices rather than on the transmitted power. Moreover, it is shown that when overall power consumption is optimized the energy efficiency of the users experiencing good propagation conditions in the uplink is not always improved by transmitting in more than one RB or by forming a virtual MIMO link. Finally, it is shown that the proposed distributed schemes achieve a similar performance in bits per Joule when compared to much more complex centralized resource allocation methods.

動態頻譜共享(dynamic spectrum sharing) 允許不具有執照的無線電用戶(坎級用戶)擇機使用具有執照的無線電用戶(主用戶)的頻譜，因此被認為是一種有效解決頻譜低效利用問題的方案。本論文研究次級用戶如何智能地實現高效率的動態頻譜共享。我們考慮兩種智能共享模式:社交智能(social intelligence) 以及個體智能(individual intelligence) 。 / 對於社交智能，次級用戶基於社交互動(social interactions) 來協作地共享頻譜。受到電子商務工業的推薦系統(recommendation sYstem) 的啟發，我們提出了一種基於推薦的社交頻譜共享機制。其中，次級用戶相互協作，彼此推薦良好的信道， 并動態接入信道。我們設計了種基於馬爾科夫決策過程( Markovdecision process) 的自適應信道推薦算法。該算法可突現良好的系統通信性能。同時，我們也提出種基於模仿(imitation) 的社交頻譜分享機制。其中，次級用戶根據自身觀察來估計自己的期望通信速率并彼此分享。如果鄰近用戶的期望通信速率更高，該用戶則模仿鄰近用戶的信道接入。我們證明該機制能夠有效地收斂到模仿均衡。如果次級用戶的數目較多，收斂的模仿均衡即是納什均衡(Nashequilibrium) 。該均衡是個次級用戶相互滿意的頻譜共享結果。 / 對於個體智能，次級用戶基於策略互動(strategic interactions) 來競爭地共享頻譜。對於基於空間複用(spatial reuse) 的競爭性頻譜共享，我們提出了種新穎的空間頻譜接入博弈框架。我們研究了不同的干擾圖形結構對於納什均衡的存在性的影響。同時，我們設計了種基於用戶自身觀察的分佈式學習算法。該算法適用於所有空間頻譜接入博弈，并能夠有效地收斂到近似納什均衡(approximateNash equilibrium) 。對於基於數據庫的電視頻譜(white-space spectrum) 無線AP(access point)網絡，我們運用博弈理論方法為分佈式AP 信道選擇問題以及分佈式次級用戶AP 連接問題建立理論模型。我們證明了分佈式AP信道選擇博奔以及分佈式次級用戶AP 連接博弈屬於勢博弈(potential game) 的範疇。基於勢博莽的有限改進性質(finite improvement property) ，我們設計了分佈式算法能夠有效地收斂到納什均衡。 / Dynamic spectrum sharing enables unlicensed secondary wireless users to opportunistically share the spectrum with licensed primary users, and thus is envisioned as a promising solution to address the spectrum under-utilization problem. This thesis explores the intelligence of secondary users for achieving efficient distributed spectrum sharing. We consider two types of intelligences: social intelligence and individual intelligence. / For the social intelligence, secondary users share the spectrum collaboratively based on social interactions. Inspired by the recommendation system in the electronic commerce industry, we propose a recommendation-based social spectrum sharing mechanism, where secondary users collaboratively recommend "good" channels to each other and access accordingly. We devise an adaptive channel recommendation algorithm based on Markov decision process, which achieves a good system communication performance. We then propose an imitation-based social spectrum sharing mechanism, where each secondary user estimates its expected throughput based on local observations, and imitates another neighboring user’s channel selection if neighbor’s estimated throughput is higher. We show that the mechanism can converge to an imitation equilibrium. When the number of users is large, the convergent imitation equilibrium corresponds to a Nash equilibrium, which is a mutually satisfactory spectrum sharing solution. / For the individual intelligence, secondary users share the spectrum competitively based on strategic interactions. To formulate the competitive spectrum sharing with spatial reuse, we propose a framework of spatial spectrum access game on general directed interference graphs. We investigate the impact of the underlying interference graph structure on the existence of a Nash equilibrium. We also design a distributed learning algorithm based on local observations that can converge to an approximate Nash equilibrium for any spatial spectrum access games. We then apply the game theoretic approach for modeling the distributed channel selection problem among the APs and distributed AP association problem among the secondary users in database-assisted white-space AP networks. We show that both the distributed AP channel selection game and the distributed AP association game are potential games. We then design distributed algorithms for achieving Nash equilibria by utilizing the finite improvement property of potential game. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Chen, Xu. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 180-188). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese. / Abstract --- p.i / Acknowledgement --- p.v / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation and Overview --- p.1 / Chapter 1.2 --- Thesis Outline --- p.5 / Chapter I --- Social Intelligence For Distributed Spectrum Sharing --- p.7 / Chapter 2 --- Recommendation-based Social Spectrum Sharing --- p.8 / Chapter 2.1 --- Introduction --- p.8 / Chapter 2.2 --- System Model --- p.12 / Chapter 2.3 --- Introduction To Channel Recommendation --- p.13 / Chapter 2.3.1 --- Review of Static Channel Recommendation --- p.14 / Chapter 2.3.2 --- Motivations For Adaptive Channel Recommendation --- p.16 / Chapter 2.4 --- Adaptive Channel Recommendation With Channel Homogeneity --- p.18 / Chapter 2.4.1 --- MDP Formulation For Adaptive Channel Recommendation --- p.19 / Chapter 2.4.2 --- Existence of Optimal Stationary Policy --- p.21 / Chapter 2.5 --- Model Reference Adaptive Search For Optimal Spectrum Access Policy --- p.22 / Chapter 2.5.1 --- Model Reference Adaptive Search Method --- p.23 / Chapter 2.5.2 --- Model Reference Adaptive Search For Optimal Spectrum Access Policy --- p.24 / Chapter 2.5.3 --- Convergence of Model Reference Adaptive Search --- p.29 / Chapter 2.6 --- Adaptive Channel Recommendation With Channel Heterogeneity --- p.30 / Chapter 2.7 --- Numerical Results --- p.33 / Chapter 2.7.1 --- Simulation Setup --- p.33 / Chapter 2.7.2 --- Homogeneous Channel Recommendation --- p.34 / Chapter 2.7.3 --- Heterogenous Channel Recommendation --- p.35 / Chapter 2.8 --- Chapter Summary --- p.38 / Chapter 2.9 --- Appendix --- p.39 / Chapter 2.9.1 --- Proof of Lemma 2.1 --- p.39 / Chapter 2.9.2 --- Derivation of Transition Probability --- p.40 / Chapter 2.9.3 --- Proof of Theorem 2.1 --- p.41 / Chapter 2.9.4 --- Proof of Theorem 2.2 --- p.42 / Chapter 2.9.5 --- Proof of Theorem 2.3 --- p.47 / Chapter 2.9.6 --- Proof of Theorem 2.4 --- p.50 / Chapter 3 --- Imitation-based Social Spectrum Sharing --- p.52 / Chapter 3.1 --- Introduction --- p.52 / Chapter 3.2 --- Spectrum Sharing System Model --- p.55 / Chapter 3.3 --- Imitative Spectrum Access Mechanism --- p.58 / Chapter 3.3.1 --- Expected Throughput Estimation --- p.59 / Chapter 3.3.2 --- Information Sharing Graph --- p.63 / Chapter 3.3.3 --- Imitative Spectrum Access --- p.63 / Chapter 3.4 --- Convergence of Imitative Spectrum Access --- p.65 / Chapter 3.4.1 --- Cluster-based Representation of Information Sharing Graph --- p.65 / Chapter 3.4.2 --- Dynamics of Imitative Spectrum Access --- p.67 / Chapter 3.4.3 --- Convergence of Imitative Spectrum Access --- p.71 / Chapter 3.5 --- Imitative Spectrum Access with Innovation --- p.73 / Chapter 3.6 --- Imitative Spectrum Access With User Heterogeneity --- p.75 / Chapter 3.7 --- Simulation Results --- p.77 / Chapter 3.7.1 --- Large User Population --- p.78 / Chapter 3.7.2 --- Small User Population --- p.82 / Chapter 3.7.3 --- Markovian Channel Environment --- p.85 / Chapter 3.7.4 --- Imitative Spectrum Access With User Heterogeneity --- p.88 / Chapter 3.8 --- Chapter Summary --- p.88 / Chapter 3.9 --- Appendix --- p.89 / Chapter 3.9.1 --- Proof of Theorem 3.1 --- p.89 / Chapter 3.9.2 --- Proof of Theorem 3.2 --- p.91 / Chapter II --- Individual Intelligence For Distributed Spectrum Sharing --- p.93 / Chapter 4 --- Spatial Spectrum Access Game --- p.94 / Chapter 4.1 --- Introduction --- p.94 / Chapter 4.2 --- System Model --- p.97 / Chapter 4.3 --- Spatial Spectrum Access Game --- p.101 / Chapter 4.4 --- Existence of Nash Equilibria --- p.102 / Chapter 4.4.1 --- Existence of Pure Nash Equilibria on Directed Interference Graphs --- p.103 / Chapter 4.4.2 --- Existence of Pure Nash Equilibria on Undirected Interference Graphs --- p.108 / Chapter 4.5 --- Distributed Learning For Spatial Spectrum Access --- p.113 / Chapter 4.5.1 --- Expected Throughput Estimation --- p.114 / Chapter 4.5.2 --- Distributed Learning Algorithm --- p.115 / Chapter 4.5.3 --- Convergence of Distributed Learning Algorithm --- p.117 / Chapter 4.6 --- Numerical Results --- p.121 / Chapter 4.7 --- Chapter Summary --- p.126 / Chapter 4.8 --- Appendix --- p.127 / Chapter 4.8.1 --- Proof of Theorem 4.2 --- p.127 / Chapter 4.8.2 --- Proof of Theorem 4.3 --- p.129 / Chapter 4.8.3 --- Proof of Lemma 4.4 --- p.131 / Chapter 4.8.4 --- Proof of Lemma 4.5 --- p.133 / Chapter 4.8.5 --- Proof of Theorem 4.5 --- p.136 / Chapter 4.8.6 --- Proof of Theorem 4.6 --- p.139 / Chapter 5 --- Distributed AP Channel Selection Game --- p.141 / Chapter 5.1 --- Introduction --- p.141 / Chapter 5.2 --- Distributed AP Channel Selection --- p.144 / Chapter 5.2.1 --- Problem Formulation --- p.144 / Chapter 5.2.2 --- Distributed AP Channel Selection Game --- p.146 / Chapter 5.3 --- Distributed AP Channel Selection Algorithms --- p.149 / Chapter 5.3.1 --- Distributed AP Channel Selection Algorithm With Information Exchange --- p.149 / Chapter 5.3.2 --- Distributed AP Channel Selection Algorithm Without Information Exchange --- p.151 / Chapter 5.4 --- Numerical Results --- p.157 / Chapter 5.4.1 --- Distributed AP Channel Selection With Information Exchange --- p.157 / Chapter 5.4.2 --- Distributed AP Channel Selection Without Information Exchange --- p.159 / Chapter 5.5 --- Chapter Summary --- p.161 / Chapter 5.6 --- Appendix --- p.162 / Chapter 5.6.1 --- Proof of Theorem 5.2 --- p.162 / Chapter 6 --- Distributed AP Association Game --- p.165 / Chapter 6.1 --- Introduction --- p.165 / Chapter 6.2 --- Distributed AP Association --- p.166 / Chapter 6.2.1 --- Channel Contention Within an AP --- p.167 / Chapter 6.2.2 --- Distributed AP Association Game --- p.168 / Chapter 6.2.3 --- Distributed AP Association Algorithm --- p.170 / Chapter 6.3 --- Numerical Results --- p.172 / Chapter 6.4 --- Chapter Summary --- p.175 / Chapter 7 --- Conclusions and Future Work --- p.176 / Bibliography --- p.180

A radio resource management scheme for WCDMA uplinks is proposed that manages quality of service (QoS) for heterogeneous services whilst maintaining high channel utilisation efficiency. The proposed system is partitioned into the 3 modules, viz. a QoS-sensitive rate scheduler, an inter-service and intra-service user prioritisation schemes, and a frame admission controller for dynamic resource reallocation. Users are allocated the minimum resources required to manage their QoS requirements through just-in-time delivery of payload, leaving more room for best-effort service users. The transmission urgency of each user is estimated by the rate scheduler based on a target transmission delay - a unique parameter used in the proposed resource management strategy to enable just-in-time payload delivery, service differentiation, and uncomplicated mapping of application requirements to QoS parameters. Transmission rate change requests from the rate schedulers are collectively processed through inter-service and intra-service priority queuing in a manner that is shown to exhibit fairness in allocation of resources amongst users of a heavily loaded network. The performance of the proposed strategy is explored through discrete-event simulations for 3 classes of traffic - voice, video and data, over the WCDMA uplink in the presence of short-term Rayleigh fading, ARQ, FEC, target transmission delays and FER targets in a multi-cell environment. Two alternatives for distributed resource management have been studied, with the UE or Node-B in control of resource allocation. The UE controlled resource management system is shown to achieve higher channel utilisation efficiency at the cost of fairness. The Node-B controlled resource manager respects the priority of speech, video and data traffic in heavily loaded systems, as reflected in 95 percentile packet transmission delays. / PhD Doctorate

Radio resource management (RRM) is an important component of a mobile wireless network that efficiently utilizes the limited radio resources such as spectrum, transmission power, and network infrastructure. Unfortunately, current RRM schemes do not support cooperative multiple point (CoMP), a promising technology that extends coverage, increases capacity, and improves the spectral efficiency of the next generation broadband network, i.e., 4G network. Specifically, to coordinate with CoMP, a RRM scheme should be aware of three main properties of CoMP - cooperative transmitting information, coordinated scheduling transmission, and single interference noise ratio (SINR) improvement. However, few of the existing RRM schemes consider these properties, since they were designed based on the conventional mobile data networks without CoMP technology. In this dissertation, I present a series of new CoMP aware RRM schemes for ensuring users' throughput and maximizing network capacity in an integrated PON-OFDM network, which is a norm of the 4G network and can best implement the CoMP technology. I call the PON-OFDM network with CoMP a CoMP Network (CoMPNet). I provide two classes of RRM schemes for two practical CoMP technologies, cooperative transmission (CT) and coordinated scheduling (CoS), respectively. In the first class, I propose two groups of RRM schemes using the CT technology. In the first group, three OFDM-TDMA based RRM schemes are designed for three different users' moving speeds. The objective of these schemes is to minimize time slot consumption. The RRM schemes in the third group are contrived for an OFDM-FDMA based CoMPNet. I provide four linear programming (LP) based optimal schemes, one for minimizing bandwidth usage, one for minimizing transmission power consumption, and two for balancing resource costs. An optimized resource allocation solution can be obtained by flexibly choosing one of the schemes according to network load. In the second class, I present a sub-optimal RRM scheme for an OFDM-FDMA based CoMPNet. The CoS technology is applied for ICI mitigation. I formulate the system optimal task into constrained optimization problems for maximizing network capacity. To improve the computation efficiency, fast yet effective heuristic schemes are introduced for divide-and-conquer. The proposed heuristic schemes are featured by CoS based timeslots/subcarriers assignment mechanisms, which are further incorporated with intelligent power control schemes. Through simulations, I study the proposed RRM schemes performances and discuss the effect of the CoMP technology. The performance benefits of CoMP on bandwidth saving and capacity increasing are shown by comparing the new schemes with the conventional schemes without CoMP.

Opportunistic spectrum access (OSA) promises to significantly improve the utilization of the RF spectrum. Under OSA, an unlicensed secondary user (SU) is allowed to detect and access under-utilized portions of the licensed spectrum, provided that such operation does not interfere with the communication of licensed primary users (PUs). Cognitive radio (CR) is a key enabling technology of OSA. In this dissertation, we propose several radio resource management and routing mechanisms that optimize the discovery and utilization of spectrum opportunities in a cognitive radio network (CRN). First, we propose a sequential channel sensing and probing mechanism that enables a resource-constrained SU to efficiently identify the optimal transmission opportunity from a pool of potentially usable channels. This mechanism maximizes the SUs expected throughput by accounting for the tradeoff between the reward and overhead of scanning additional channels. The optimal channel sensing and probing process is modeled as a maximum-rate-of-return problem in optimal stopping theory. Operational parameters, such as sensing and probing times, are optimized by exploiting the problem's special structure. Second, we study the problem of coordinated spectrum access in CRNs to maximize the CRNs throughput. By exploiting the geographic relationship between an SU and its surrounding PUs, we propose the novel concept of microscopic spectrum opportunity, in which active SUs and PUs are allowed to operate in the same region, subject to power constraints. Under this framework, we formulate the coordinated channel access problem as a joint power/rate control and channel assignment optimization problem. Centralized and distributed approximate algorithms are proposed to solve this problem efficiently. Compared with its macroscopic counterpart, we show that the microscopic-spectrum-opportunity framework offers significant throughput gains. Finally, at the network layer, we study the problem of truthful least-priced-path (LPP) routing for profit-driven CRNs. We design a route selection and pricing mechanism that guarantees truthful spectrum cost reporting from profit-driven SUs and that finds the cheapest route for end users. The problem is investigated with and without capacity constraints at individual nodes. In both cases, polynomial-time algorithms are developed to solve the LPP problem. Extensive simulations are conducted to verify the validity of the proposed mechanisms.

In recent years, there has been a rapid evolution and deployment of wireless networks. In populated areas, high-rate data access is enabled anywhere and anytime with the pervasive wireless infrastructure such as the fourth-generation (4G) cellular systems, IEEE 802.11-based wireless local area networks (WLANs), and IEEE 802.16-based wireless metropolitan area networks (WMANs). In such a heterogeneous wireless access medium, multi-radio devices become a trend for users to conveniently explore various services offered by different wireless systems. This thesis presents radio resource management mechanisms, for bandwidth allocation, call admission control (CAC), and mobile terminal (MT) energy management, that can efficiently exploit the available resources in the heterogeneous wireless medium and enhance the user perceived quality-of-service (QoS). Almost all existing studies on heterogeneous networking are limited to the traditional centralized infrastructure, which is inflexible in dealing with practical scenarios, especially when different networks are operated by different service providers. In addition, in most current wireless networks, mobile users are simply viewed as service recipients in network operation, with passive transceivers completely or partially under the control of base stations or access points. In this thesis, we present efficient decentralized bandwidth allocation and CAC mechanisms that can support single-network and multi-homing calls. The decentralized architecture gives an active role to the MT in the resource management operation. Specifically, an MT with single-network call can select the best wireless network available at its location, while an MT with multi-homing call can determine a required bandwidth share from each network to satisfy its total required bandwidth. The proposed mechanisms rely on cooperative networking and offer a desirable flexibility between performance measures (in terms of the allocated bandwidth per call and the call blocking probability), and between the performance and the implementation complexity. With the increasing gap between the MT demand for energy and the offered battery capacity, service degradation is expected if the MT cannot efficiently manage its energy consumption. Specifically, for an uplink multi-homing video transmission, the existing studies do not guarantee that the MT available energy can support the entire call, given the battery energy limitation. In addition, the energy management mechanism should take account of video packet characteristics, in terms of packet distortion impact, delay deadline, and precedence constraint, and employ the available resources in the heterogeneous wireless medium. In this thesis, we present MT energy management mechanisms that can support a target call duration, with a video quality subject to the MT battery energy limitation. In addition, we present a statistical guarantee framework that can support a consistent video quality for the target call duration with minimum power consumption.

Radio resource management (RRM) is an important component of a mobile wireless network that efficiently utilizes the limited radio resources such as spectrum, transmission power, and network infrastructure. Unfortunately, current RRM schemes do not support cooperative multiple point (CoMP), a promising technology that extends coverage, increases capacity, and improves the spectral efficiency of the next generation broadband network, i.e., 4G network. Specifically, to coordinate with CoMP, a RRM scheme should be aware of three main properties of CoMP - cooperative transmitting information, coordinated scheduling transmission, and single interference noise ratio (SINR) improvement. However, few of the existing RRM schemes consider these properties, since they were designed based on the conventional mobile data networks without CoMP technology. In this dissertation, I present a series of new CoMP aware RRM schemes for ensuring users' throughput and maximizing network capacity in an integrated PON-OFDM network, which is a norm of the 4G network and can best implement the CoMP technology. I call the PON-OFDM network with CoMP a CoMP Network (CoMPNet). I provide two classes of RRM schemes for two practical CoMP technologies, cooperative transmission (CT) and coordinated scheduling (CoS), respectively. In the first class, I propose two groups of RRM schemes using the CT technology. In the first group, three OFDM-TDMA based RRM schemes are designed for three different users' moving speeds. The objective of these schemes is to minimize time slot consumption. The RRM schemes in the third group are contrived for an OFDM-FDMA based CoMPNet. I provide four linear programming (LP) based optimal schemes, one for minimizing bandwidth usage, one for minimizing transmission power consumption, and two for balancing resource costs. An optimized resource allocation solution can be obtained by flexibly choosing one of the schemes according to network load. In the second class, I present a sub-optimal RRM scheme for an OFDM-FDMA based CoMPNet. The CoS technology is applied for ICI mitigation. I formulate the system optimal task into constrained optimization problems for maximizing network capacity. To improve the computation efficiency, fast yet effective heuristic schemes are introduced for divide-and-conquer. The proposed heuristic schemes are featured by CoS based timeslots/subcarriers assignment mechanisms, which are further incorporated with intelligent power control schemes. Through simulations, I study the proposed RRM schemes performances and discuss the effect of the CoMP technology. The performance benefits of CoMP on bandwidth saving and capacity increasing are shown by comparing the new schemes with the conventional schemes without CoMP.