Wong Tsz-Chun Stanley. / Thesis submitted in: July 2004. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 58-61). / Abstracts in English and Chinese. / 摘要 --- p.III / ACKNOWLEDGEMENT --- p.IV / LIST OF FIGURES --- p.VIII / Chapter CHAPTER 1 --- INTRODUCTION --- p.2 / Chapter 1.1 --- Introduction --- p.2 / Chapter 1.2 --- Organization of the Thesis --- p.6 / Chapter CHAPTER 2 --- BACKGROUND STUDIES --- p.7 / Chapter 2.1 --- Overview of IEEE 802.11 --- p.7 / Chapter 2.3 --- 802.11b Network Type --- p.10 / Chapter 2.4 --- IEEE 802.11 MAC overview --- p.12 / Chapter 2.4.1 --- Distributed Coordination Function (DCF) / Point Coordination Function (PCF) --- p.13 / Chapter 2.4. --- Request-to-Send (RTS) / Clear-to-Send (CTS) --- p.14 / Chapter CHAPTER 3 --- QUEUE MANAGEMENT IN ACCESS POINTS --- p.16 / Chapter 3.1 --- Introduction --- p.16 / Chapter 3.2 --- Packet Delay Variation in Access Points --- p.17 / Chapter 3.2 --- Simulations Settings and Configuration --- p.19 / Chapter 3.2.1 --- Mobile Networking in NS2 --- p.19 / Chapter 3.2.2 --- Input Parameter of Shadowing Radio Propagation model --- p.22 / Chapter 3.2.3 --- Configuration of the simulation --- p.25 / Chapter CHAPTER 4 --- PERFORMANCE ANALYSIS OF 802.11B DOWNLINKS --- p.30 / Chapter 4.1 --- Introduction --- p.30 / Chapter 4.5 --- Assumptions of the analysis --- p.31 / Chapter 4.3 --- Proposed Queue Management in Access Point --- p.32 / Chapter 4.4 --- Channel error modeling --- p.34 / Chapter 4.5 --- The analytical model of downlinks --- p.35 / Chapter 4.6 --- Performance Calculation --- p.37 / Chapter 4.4 --- Simulation Result --- p.41 / Chapter 4.4.1 --- Extension of NS2 --- p.42 / Chapter 4.4.2 --- Configuration of the simulation --- p.44 / Chapter CHAPTER 5 --- RESOURCE ASSIGNMENT IN ACCESS POINTS --- p.48 / Chapter 5.1 --- Introduction --- p.48 / Chapter 5.2 --- Packet Delay Balancing --- p.49 / Chapter 5.2 --- Simulation Result --- p.51 / Chapter CHAPTER 6 --- CONCLUSION --- p.56 / BIBLIOGRAPHY --- p.58

Liu Haiping. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 78-82). / Abstracts in English and Chinese. / Acknowledgments --- p.i / Abstract --- p.ii / Table of Contents --- p.vi / List of Figures --- p.vii / List of Tables --- p.ix / Chapter Chapter I --- Introduction to Wireless LAN --- p.1 / Chapter 1.1 --- Wireless LAN Netwrok Architecture --- p.2 / Chapter 1.2 --- IEEE 802.11 protocol family --- p.4 / Chapter 1.3 --- The Major factors influencing the System Performance --- p.7 / Chapter 1.4 --- Research Objectives --- p.13 / Chapter 1.5 --- Overview --- p.13 / Chapter Chapter II --- The Major Deficiency of DCF and Motivation --- p.14 / Chapter 2.1 --- DCF --- p.15 / Chapter 2.2 --- The Major Deficiencies in the DCF --- p.20 / Chapter 2.3 --- Improvement directions --- p.24 / Chapter Chapter III --- Proposed MAC Protocol --- p.27 / Chapter 3.1 --- The Design Idea --- p.28 / Chapter 3.2 --- The Number of Active Nodes --- p.30 / Chapter 3.3 --- Optimizaition Method for CW --- p.32 / Chapter 3.4 --- CW and Counter value Updating --- p.35 / Chapter 3.5 --- Procedure Flow and Simulation Results --- p.37 / Chapter Chapter IV --- Advanced Proposed Protocol with QoS issues --- p.44 / Chapter 4.1 --- "QoS requirement, EDCF solution and others' work" --- p.45 / Chapter 4.2 --- Frame structure changes in the Advanced Proposed MAC Protocol --- p.47 / Chapter 4.3 --- Recursivley Balance Optimization Method for CW --- p.48 / Chapter 4.4 --- Decision Algorithm --- p.54 / Chapter 4.5 --- Model Validation and Simulation Results --- p.66 / Chapter Chapter V --- Further Discussion about CW design --- p.70 / Chapter 5.1 --- Influence of the ranges of CW --- p.71 / Chapter 5.2 --- Proposal for adjusting CW --- p.73 / Chapter Chapter VI --- Conclusion --- p.75 / Bibliography --- p.78

An active sensor network is a wireless network comprising a large number of mobile sensor nodes. This dissertation deals with a general model of active sensor network, heterogeneous sensor network, where sensor nodes may have different sensing ranges. The deployment problem aims at relocating a large set of sensor nodes from arbitrary locations to give a connected, hole-free and locally maximized sensing coverage of the entire network. / The major contribution of this dissertation lies in four aspects. First, it is the first work that solves the deployment problem for optimal deterministic coverage of heterogeneous sensor networks, while most existing works limit their problems on stochastic coverage and homogeneous sensing model. Second, this work envisages the use of a generalized Voronoi diagram, the power diagram, as a novel solution to geometrically analyse and visualise the coverage of a heterogeneous sensor network. Third, it an original work that applies circle packing on sensor network deployment, and analyses and proves a number of geometrical properties of circle packing. Fourth, all methods provided in this dissertation are based on localized and distributed computation; no centralized processor or common data fusion platform is assumed to exist. / This dissertation gives an algorithm to solve the self-deployment problem. It is composed three parts. In the first part, the logical topology of the sensor network is constructed as triangulation by three distributed protocols: localized Delaunay triangulation, redundant boundary edge pruning and local edge swapping. Second, the sensor nodes self-deploy to new locations that are calculated using a circle packing algorithm. The dissertation shows that the homomorphism between Voronoi and power diagrams is necessary and sufficient for the equivalences of power Delaunay triangles to Delaunay triangles. This result allows the network to preserve a unit Delaunay triangulation by localized re-triangulations among a small number of nodes. Third, the sensor nodes further relocate themselves based on a virtual force approach to eliminate all existing coverage holes and redundant overlaps. / This dissertation studies the problem of active sensor network deployment. It focuses on self-deployment, localized and distributed computation and coverage maximization of heterogeneous sensor networks. / Lam, Miu Ling. / "February 2008." / Adviser: Yun-hui Liu. / Source: Dissertation Abstracts International, Volume: 70-03, Section: B, page: 1757. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (p. 152-158). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

As a subtopic in this thesis, we also investigate how to deal with uncertainty in the context of the traditional Straightforward Network Coding (SNC) scheme. With the proposed scheme, Soft Network Coding, only simple symbol-level network coding operation at the physical layer of the relay node can achieve even better performance than the traditional SNC scheme, which needs the complicated channel decoding and re-encoding operation. / Network coding is a promising upper layer technique first proposed in the context of wired networks. In this thesis, we investigate the application of network coding at the physical layer of wireless networks to take into account the unique properties of wireless networks. Specifically, we propose a new network coding scheme referred to as Physical layer Network Coding (PNC). PNC effects network-coding operation directly at the physical layer by proper interpretation of EM (Electromagnetic) signal received simultaneously from multiple sources. From the network point of view, this scheme can approach the min-cut throughput for both bi-direction and uni-direction linear relay networks; from the information theory point of view, this scheme can approach the capacity of two-way relay channel in the low and high SNR regions. When channel coding is considered, we could classify PNC into two classes, end-to-end coded PNC and link-by-link coded PNC. For end-to-end coded PNC, we further classify it into subclasses: PNC over infinite field (PNCI) and PNC over finite field (PNCF). For each subclass, we propose and analyze new PNC mapping schemes. For link-by-link coded PNC, we focus on the transformation from the received packet Y3 to the network coded form of unchannel-coded packet S1⊕S2, referred to as the Channel-decoding-Network-Coding process (CNC). Among three CNC designs, a matched CNC, CNC3, is of great interest due to its superior performance. Therefore, we design a new decoding algorithm at the relay node to make CNC3 feasible. Simulation result shows that the matched CNC with our new decoding algorithm outperforms the two straightforward CNC designs significantly in terms of BER without added complexity. Overall, this thesis lays down the fundamentals and foundation of PNC. And through theoretical analysis and implementation constructions, we provide insights on how good performance in wireless networks can be achieved with PNC. / Zhang, Shengli. / Adviser: Soung Chang Lieu. / Source: Dissertation Abstracts International, Volume: 70-06, Section: B, page: 3709. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 178-183). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Access scheduling is essentially power control with only on-off options. The second part is dealing with situations where user transmissions can exist simultaneously. Here, a problem of distributed power control for time varying systems is investigated in order to coordinate transmit power among users for their respective quality-of-service (QoS). We deal with a class of power control problems where the system link gains are assumed to be time varying and SIR estimates are allowed to be corrupted with bounded noises. A simple control algorithm is devised by applying a distributed, fixed-step approach. The feedback algorithm requires only local information. By modifying the fixed-step power control algorithm proposed by Sung and Wong, here we obtain a more robust version that can handle time varying link gains and measurement noises. The result extends the Foschini and Miljanic model to allow fading and measurement errors. Convergence property of the new algorithm is established. Simulation studies have been conducted and results show that it is effective. / In integrated wireless multimedia service, isochronous traffic of different connections can be scheduled by using a most regular binary sequence (MRBS). Such a sequence can schedule traffic in an evenly spaced manner to achieve any arbitrary rate asymptotically while avoiding excessive delay and buffering requirement. Flexible time slot assignment that can match requests exactly improves the bandwidth utilization efficiency in supporting multi-rate operations for traffics of various classes. The most regular binary sequence provides such a distributed solution for multi-access control that only requires a limited information exchange. Generally, the idea can be developed to support flexible resource allocation in various communication systems such as hybrid TD-CDMA and MC-CDMA systems. It results in an overall capacity gain. More interestingly, the MRBS transmission scheduling is applicable in a network sense. Deterministic end-to-end performance guarantees such as packet delay and buffering requirement can be investigated in a systematic way. Discussions of periodic binary sequences with interesting characteristics are presented in succession. / Chen Chung Shue. / "September 2005." / Adviser: Wing Shing Wong. / Source: Dissertation Abstracts International, Volume: 67-07, Section: B, page: 3984. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 154-167). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract in English and Chinese. / School code: 1307.

While the bandwidth available for wireless networks is limited, the world has seen an unprecedented growth in the number of mobile subscribers and an ever increasing demand for high data rates. Therefore efficient utilisation of bandwidth to maximise link spectral efficiency and number of users that can be served simultaneously are primary goals in the design of wireless systems. To achieve these goals, in this thesis, a new non-orthogonal uplink multiple access scheme which combines the functionalities of adaptive modulation and multiple access called collaborative modulation multiple access (CMMA) is proposed. CMMA enables multiple users to access the network simultaneously and share the same bandwidth even when only a single receive antenna is available and in the presence of high channel correlation. Instead of competing for resources, users in CMMA share resources collaboratively by employing unique modulation sets (UMS) that differ in phase, power, and/or mapping structure. These UMS are designed to insure that the received signal formed from the superposition of all users' signals belongs to a composite QAM constellation (CC) with a rate equal to the sum rate of all users. The CC and its constituent UMSs are designed centrally at the BS to remove ambiguity, maximize the minimum Euclidian distance (dmin) of the CC and insure a minimum BER performance is maintained. Users collaboratively precode their transmitted signal by performing truncated channel inversion and phase rotation using channel state information (CSI ) obtained from a periodic common pilot to insure that their combined signal at the BS belongs to the CC known at the BS which in turn performs a simple joint maximum likelihood detection without the need for CSI. The coherent addition of users' power enables CMMA to achieve high link spectral efficiency at any time without extra power or bandwidth but on the expense of graceful degradation in BER performance. To improve the BER performance of CMMA while preserving its precoding and detection structure and without the need for pilot-aided channel estimation, a new selective diversity combining scheme called SC-CMMA is proposed. SC-CMMA optimises the overall group performance providing fairness and diversity gain for various users with different transmit powers and channel conditions by selecting a single antenna out of a group of L available antennas that minimises the total transmit power required for precoding at any one time. A detailed study of capacity and BER performance of CMMA and SC-CMMA is carried out under different level of channel correlations which shows that both offer high capacity gain and resilience to channel correlation. SC-CMMA capacity even increase with high channel correlation between users' channels. CMMA provides a practical solution for implementing the multiple access adder channel (MAAC) in fading environments hence a hybrid approach combining both collaborative coding and modulation referred to as H-CMMA is investigated. H-CMMA divides users into a number of subgroups where users within a subgroup are assigned the same modulation set and different multiple access codes. H-CMMA adjusts the dmin of the received CC by varying the number of subgroups which in turn varies the number of unique constellation points for the same number of users and average total power. Therefore H-CMMA can accommodate many users with different rates while flexibly managing the complexity, rate and BER performance depending on the SNR. Next a new scheme combining CMMA with opportunistic scheduling using only partial CSI at the receiver called CMMA-OS is proposed to combine both the power gain of CMMA and the multiuser diversity gain that arises from users' channel independence. To avoid the complexity and excessive feedback associated with the dynamic update of the CC, the BS takes into account the independence of users' channels in the design of the CC and its constituent UMSs but both remain unchanged thereafter. However UMS are no longer associated with users, instead channel gain's probability density function is divided into regions with identical probability and each UMS is associated with a specific region. This will simplify scheduling as users can initially chose their UMS based on their CSI and the BS will only need to resolve any collision when the channels of two or more users are located at the same region. Finally a high rate cooperative communication scheme, called cooperative modulation (CM) is proposed for cooperative multiuser systems. CM combines the reliability of the cooperative diversity with the high spectral efficiency and multiple access capabilities of CMMA. CM maintains low feedback and high spectral efficiency by restricting relaying to a single route with the best overall channel. Two possible variations of CM are proposed depending on whether CSI available only at the users or just at the BS and the selected relay. The first is referred to Precode, Amplify, and Forward (PAF) while the second one is called Decode, Remap, and Forward (DMF). A new route selection algorithm for DMF based on maximising dmin of random CC is also proposed using a novel fast low-complexity multi-stage sphere based algorithm to calculate the dmin at the relay of random CC that is used for both relay selection and detection.

The idea of the smart home has been discussed for over three decades, but it has yet to achieve mass-market adoption. This thesis asks the question Why is my home not smart? It highlights four main areas that are barriers to adoption, and concentrates on a single one of these issues: usability. It presents an investigation that focuses on design, prototyping and evaluation of mobile interfaces for domestic environments resulting in the development of a novel framework. A smart home is the physical realisation of a ubiquitous computing system for domestic living. The research area offers numerous benefits to end-users such as convenience, assistive living, energy saving and improved security and safety. However, these benefits have yet to become accessible due to a lack of usable smart home control interfaces. This issue is considered a key reason for lack of adoption and is the focus for this thesis. Within this thesis, a framework is introduced as a novel approach for the design, prototyping and evaluation of mobile interfaces for domestic environments. Included within this framework are three components. Firstly, the Reconfigurable Multimedia Environment (RME), a physical evaluation and observation space for conducting user centred research. Secondly, Simulated Interactive Devices (SID), a video-based development and control tool for simulating interactive devices commonly found within a smart home. Thirdly, iProto, a tool that facilitates the production and rapid deployment of high fidelity prototypes for mobile touch screen devices. This framework is evaluated as a round-tripping toolchain for prototyping smart home control and found to be an efficient process for facilitating the design and evaluation of such interfaces.

With the expectation of over a 1000-fold increase in the number of connected devices by 2020, efficient utilization of the limited bandwidth has become ever more important in the design of mobile wireless systems. Furthermore, the ever-increasing demand for higher data rates has made it necessary for a new waveform design that satisfies not only throughput demands, but network capacity as well. One such technique recently proposed is the non-orthogonal multiple access (NOMA) which utilizes the distance-dependent power domain multiplexing, based on the principles of signal superposition. In this thesis, new spectrally efficient non-orthogonal signal techniques are proposed. The goal of the schemes is to allow simultaneous utilization of the same time frequency network resources. This is achieved by designing component signals in both power and phase domain such that users are precoded or preformed to form a single and uniquely decodable composite signal. The design criteria are based on maximizing either the sum rate or spectral efficiency, minimizing multi-user interference and detection ambiguity, and maximizing the minimum Euclidean distance between the composite constellation points. The design principles are applied in uplink, downlink and coordinated multipoint (CoMP) scenarios. We assume ideal channel state with perfect estimation, low mobility and synchronization scenarios so as to prove the concept and serve as a bound for any future work in non-ideal conditions. Extensive simulations and numerical analysis are carried to show the superiority and compatibility of the schemes. First, a new NOMA signal design called uplink NOMA with constellation precoding is proposed. The precoding weights are generated at the eNB based on the number of users to be superposed. The eNB signals the precoding weights to be employed by the users to adjust their transmission. The adjustments utilize the channel state information estimated from common periodic pilots broadcasted by the eNB. The weights ensure the composite received signal at the eNB belongs to the pre-known constellation. Furthermore, the users precode to the eNB antenna that requires the least total transmit power from all the users. At the eNB, joint maximum likelihood (JML) detection is employed to recover the component signals. As the composite constellation is as that of a single user transmitting that same constellation, multiple access interference can be viewed as absent, which allows multiple users to transmit at their full rates. Furthermore, the power gain achieved by the sum of the component signals maximizes the sum rate. Secondly, the constellation design principle is employed in the downlink scenario. In the scheme, called downlink NOMA with constellation preforming, the eNB preforms the users signal with power and phase weights prior to transmission. The preforming ensures multi-user interference is eliminated and the spectral efficiency maximized. The preformed composite constellation is broadcasted by the eNB which is received by all users. Subsequently, the users perform JML detection with the designed constellation to extract their individual component signals. Furthermore, improved signal reliability is achieved in transmit and receive diversity scenarios in the schemes called distributed transmit and receive diversity combining, respectively. Thirdly, the constellation preforming on the downlink is extended to MIMO spatial multiplexing scenarios. The first MIMO scheme, called downlink NOMA with constellation preforming, each eNB antenna transmits a preformed composite signal composed of a set of multiple users' streams. This achieves spatial multiplexing with diversity with less transmit antennas, reducing costs associated with multiple RF chains, while still maximizing the sum rate. In the second MIMO scheme, a highly spectrally efficient MIMO preforming scheme is proposed. The scheme, called group layer MIMO with constellation preforming, the eNB preforms to a specific group of users on each transmit antenna. In all the schemes, the users perform JML detection to recover their signals. Finally, the adaptability of the constellation design is shown in CoMP. The scheme, called CoMP with joint constellation processing, the additional degrees of freedom, in form of interfering eNBs, are utilized to enable spatial multiplexing to a user with a single receive antenna. This is achieved by precoding each stream from the coordinating eNB with weights signalled by a central eNB. Consequently, the inter-cell interference is eliminated and the sum-rate maximized. To reduce the total power spent on precoding, an active cell selection scheme is proposed where the precoding is employed on the highest interferers to the user. Furthermore, a power control scheme is applied the design principle, where the objective is to reduce cross-layer interference by adapting the transmission power to the mean channel gain.

The research contributes to work in Rapid Deployment of a National Public Security and Emergency Communications Network using Communication Satellite Broadband. Although studies in Public Security Communication networks have examined the use of communications satellite as an integral part of the Communication Infrastructure, there has not been an in-depth design analysis of an optimized regional broadband-based communication satellite in relation to the envisaged service coverage area, with little or no terrestrial last-mile telecommunications infrastructure for delivery of satellite solutions, applications and services. As such, the research provides a case study of a Nigerian Public Safety Security Communications Pilot project deployed in regions of the African continent with inadequate terrestrial last mile infrastructure and thus requiring a robust regional Communications Satellite complemented with variants of terrestrial wireless technologies to bridge the digital hiatus as a short and medium term measure apart from other strategic needs. The research not only addresses the pivotal role of a secured integrated communications Public safety network for security agencies and emergency service organizations with its potential to foster efficient information symmetry amongst their operations including during emergency and crisis management in a timely manner but demonstrates a working model of how analogue spectrum meant for Push-to-Talk (PTT) services can be re-farmed and digitalized as a “dedicated” broadband-based public communications system. The network's sustainability can be secured by using excess capacity for the strategic commercial telecommunication needs of the state and its citizens. Utilization of scarce spectrum has been deployed for Nigeria's Cashless policy pilot project for financial and digital inclusion. This effectively drives the universal access goals, without exclusivity, in a continent, which still remains the least wired in the world.

With the proliferation of smartphones and their advanced connectivity capabilities, opportunistic networks have gained a lot of traction during the past years; they are suitable for increasing network capacity and sharing ephemeral, localised content. They can also offload traffic from cellular networks to device-to-device ones, when cellular networks are heavily stressed. Opportunistic networks can play a crucial role in communication scenarios where the network infrastructure is inaccessible due to natural disasters, large scale terrorist attacks or government censorship. Geocasting, where messages are destined to specific locations (casts) instead of explicitly identified devices, has a large potential in real world opportunistic networks, however it has attracted little attention in the context of opportunistic networking. In this thesis, we propose Geocasting Spray And Flood (GSAF), a simple but efficient and flexible geocasting protocol for opportunistic, delay tolerant networks. GSAF follows a simple but elegant and flexible approach where messages take random walks towards the destination cast. Messages that follow directions away from the cast are extinct when the device buffer gets full, freeing space for new messages to be delivered. In GSAF, casts do not have to be pre-defined; instead users can route messages to arbitrarily defined casts. Also, the addressed cast is flexible in comparison to other approaches and can take complex shapes in the network. DA-GSAF as the direction aware version of the GSAF is proposed as well which use location information to aid routing decisions in the GSAF. Extensive evaluation shows that GSAF and DA-GSAF are significantly more efficient than existing solutions, in terms of message delivery ratio and latency as well as network overhead.