This dissertation presents a novel approach to Internet congestion control known as Probabilistic Congestion Notification (PCN). Preliminary research has established the requirements and motivations for a more efficient Transmission Control Protocol (TCP) congestion control schemes. Because of the substantial increase in the the number of Internet users and the diversification of online services, the dynamics of the Internet are changing, making classical congestion control mechanisms inefficient. Traditionally, TCP has relied on packet loss as an indication of congestion. Though packet loss is a typical result of overflowing buffers, it cannot be used to prevent congestion before it occurs. Consequently, a novel protocol is proposed in this thesis, which allows congestion notification and control via packet marking. This new protocol is elegant as it makes use of the existing one-bit Explicit Congestion Notification (ECN) field in the Internet Protocol (IP) header. Each PCN data packet can be marked probabilistically by one router at most. Furthermore, PCN performance has been improved by pre-signalling the number of intervening queues along the path. The level of load factor (congestion measure) at each link is fed back to the PCN source, which then estimates the exact level of congestion at each intermediate queue, one-step ahead of time. By knowing this, the source can take avoiding action either by adjusting its sending rate or by using alternate routes. The estimation mechanism uses Multiple Linear Regression (MLR) and Time Series Analysis to improve the quality of the congestion estimate and to predict the level of congestion at each queue along the path. For this purpose, the work presented in this thesis also analyses the suitability and accuracy of such statistical methods in predicting future congestion levels. The new PCN congestion control protocol has been designed and simulated in the network simulator (ns-2). Moreover, this project evaluates PCN’s performance using real high speed Internet traffic traces. Results show that the methods can successfully predict the congestion level at any queue along the path. The proposed approach has low complexity and it is easy to implement. Furthermore, it can be easily deployed in existing TCP/IP networks as it does not require modifications to existing IP or TCP implementations.

Over the past century, due to the significant increase in recreational activities, transportation, construction in high altitude areas, mountain areas all around the world, have been seen substantial development. In these high altitude areas around mountain ranges, avalanche always brings huge threaten to human’s activities and lives. The rising demand for higher safety measures has given new pressure to the development of mitigation technology to protect human in certain areas, and driven rise to a new scientific area entirely devoted to avalanche. A RADAR system can provide superior penetration capability through any type of weather condition, and can be used in the day or night time. A RADAR system uses electromagnetic wave that does not require a medium like Sonar that is using water as medium. Radar also can be long range, because the electromagnetic wave is able to propagate at the speed of light. It is less susceptible to weather conditions compared with Lasers. It does not require target cooperation to emit any signals. The original contribution of this thesis contains four parts: the novel work of snow particles models (option models and HFT model); the simulation of a receiver module in an avalanche radar system; the development of an active baseband filter; and the development of an FPGA chirp generator. The option model and HFT model are built to give solutions and to map and predict the moving route of snow particles. The purposes of proposing these two models are different and the assumptions of both totally diverge. The option model is built on the assumption of knowing all the information about the stop location and start location of an airborne avalanche. The key concept of the option model is the introduction of a binary tree. By using the theories of a binary tree, normal distribution and the knowledge of the stop location and start location of an airborne avalanche, its route can be mapped and the behaviour can be further studied. On the other hand, an HFT model is based on the theories of stochastic process. It does not require any knowledge of the avalanches and can be used to predict the movement of an airborne avalanche. The FPGA chirp generator is built for more flexibility than the DDS waveform generator. The simulation is done in this thesis to help design the receiver module in avalanche radar. And the prototype of the FPGA chirp generator is based on Xilinx virtex-5 development board and avanet high-speed DAC. This new design is different from existing FPGA chirp generator, since it uses the onboard memory to store the chirp signal data, which gives the ability to store more data to significantly increase the sampling rate and resolution of chirp signal.

Disorders of the vestibular system result in loss of body balance and a steady vision in humans and animals. The most common consequences include vertigo, oscillopsia, postural instability and blurred vision. Currently, conventional medicine cannot cure the damage or restore the function of the vestibular system. Vestibular prosthesis may assist in restoring its function using electrical stimulation, which involves delivering current pulses into the nerves innervating the semi-circular canals in the inner ear. A vestibular prosthesis contains external electronics and an implantable medical device . The system delivers modulated electrical pulses and stimulates vestibular nerves with these pulses to inform the brain about the motion. Power transfer to, and communication with the implanted device, is provided by telemetry. In biomedical implanted devices the telemetry is usually implemented by radio-frequency induction using weakly coupled coils. Using a single set of coils for simultaneous power transfer and communication creates the challenge of contradicting requirements. For high data rates the inductive link must have a wide bandwidth but power transfer requires a low bandwidth. Moreover by modulating a carrier the power transfer is degraded. This thesis describes the design, implementation and experimental evaluation of a novel telemetry system for a three-dimensional vestibular prosthesis with neural recording. The developed telemetry system uses a single pair of inductively-coupled coils to power-up the implant and maintain bi-directional communication to control its operation. It also relays raw electroneurogram (ENG) data out of the body at high speed. For inductive power control two methods are combined: a geometrical approach and a feedback loop to maintain a constant level of delivered power. The communication to the implant (downlink) is obtained by amplitude modulation while the communication from the implant to the external transmitter (uplink) uses passive phase shift modulation. On-chip humidity sensing capabilities are facilitated in the implant microelectronics to monitor hermeticity of the package. The uplink achieves the highest data speed demonstrated in the literature of available methods using a single set of coils with combined power and communication links. The developed technique can be applied to other applications including RFID.

Next generation optical networks will evolve from static to dynamically reconfigurable architectures to meet the increasing bandwidth and service requirements. The benefits of dynamically reconfigurable networks (improved operations, reduced footprint and cost) have introduced new challenges, in particular the need for complex management which has put pressure on the engineering rules and transmission margins. This has provided the main drive to develop new techniques for optical performance monitoring (OPM) without using optical-to-electrical-to-optical conversions. When considering impairments due to chromatic dispersion in dynamic networks, each channel will traverse a unique path through the network thus the channels arriving at the monitoring point will, in general, exhibit different amounts of residual dispersion. Therefore, in a dynamic network it is necessary to monitor all channels individually to quantify the degradation, without the requirement of knowing the data path history. The monitoring feature can be used in conjunction with a dispersion compensation device which can either be optical or electrical or used to trigger real-time alarms for traffic re-routing. The proposed OPM technique is based on RF spectrum analysis and used for simultaneous and independent monitoring of power, chromatic dispersion (CD), polarisation mode dispersion (PMD) and optical signal-to-noise ratio (OSNR) in 40Gbit/s multi0channel systems. An analytical model is developed to describe the monitoring technique which allows the prediction of the measurement range. The experimental results are given for group velocity dispersion (GVD), differential group delay (DGD) and OSNR measurements. This technique is based on electro-optic down-conversion that simultaneously down-converts multiple channels, sharing the cost of the key components over multiple channels and making it cost effective for multi-channel operation. The measurement range achieved with this method is equal to 4742±100ps/nm for GVD, 200±4ps for DGD and 25±1dB for OSNR. To the knowledge of the author, these dispersion monitoring ranges are the largest reported to date for the bit-rate of 40Gbit/s with amplitude modulation formats.

A low jitter optical pulse source for applications including optical sampling and optical metrology was modelled and then experimentally implemented using photonic components. Dispersion and non-linear fibre effects were utilised to compress a periodic optical waveform to generate pulses of the order of 10 picoseconds duration, via soliton-effect compression. Attractive features of this pulse source include electronically tuneable repetition rates greater than 1.5 GHz, ultra-short pulse duration (10-15 ps), and low timing jitter as measured by both harmonic analysis and single-sideband (SSB) phase noise measurements. The experimental implementation of the modelled compression scheme is discussed, including the successful removal of stimulated Brillouin scattering (SBS) through linewidth broadening by injection dithering or phase modulation. Timing jitter analysis identifies many unwanted artefacts generated by the SBS suppression methods, hence an experimental arrangement is devised (and was subsequently patented) which ensures that there are no phase modulation spikes present on the SSB phase noise spectrum over the offset range of interest for optical sampling applications, 10Hz-Nyquist. It is believed that this is the first detailed timing jitter study of a soliton-effect compression scheme. The soliton-effect compression pulses are then used to perform what is believed to be the first demonstration of optical sampling using this type of pulse source. The pulse source was also optimised for use in a novel optical metrology (range finding) system, which is being developed and patented under European Space Agency funding as an enabling technology for formation flying satellite missions. This new approach to optical metrology, known as Scanning Interferometric Pulse Overlap Detection (SIPOD), is based on scanning the optical pulse repetition rate to find the specific frequencies which allow the return pulses from the outlying satellite, i.e. the measurement arm, to overlap exactly with a reference pulse set on the hub satellite. By superimposing a low frequency phase modulation onto the optical pulse train, it is possible to detect the pulse overlap condition using conventional heterodyne detection. By rapidly scanning the pulse repetition rate to find two frequencies which provide the overlapping pulse condition, high precision optical pulses can be used to provide high resolution unambiguous range information, using only relatively simple electronic detection circuitry. SIPOD’s maximum longitudinal range measurement is limited only by the coherence length of the laser, which can be many tens of kilometres. Range measurements have been made to better than 10 microns resolution over extended duration trial periods, at measurement update rates of up to 470 Hz. This system is currently scheduled to fly on ESA’s PROBA-3 mission in 2012 to measure the intersatellite spacing for a two satellite coronagraph instrument. In summary, this thesis is believed to present three novel areas of research: the first detailed jitter characterisation of a soliton-effect compression source, the first optical sampling using such a compression source, and a novel optical metrology range finding system, known as SIPOD, which utilises the tuneable repetition rate and highly stable nature of the compression source pulses.

The concept of cooperative communication has been proposed to improve link capacity, transmission reliability and network coverage in multiuser wireless communication networks. Different from conventional point-to-point and point-to-multipoint communications, cooperative communication allows multiple users or stations in a wireless network to coordinate their packet transmissions and share each other’s resources, thus achieving high performance gain and better service coverage. According to the IEEE 802.11 standards, Wireless Local Area Networks (WLANs) can support multiple transmission data rates, depending on the instantaneous channel condition between a source station and an Access Point (AP). In such a multi-rate WLAN, those low data-rate stations will occupy the shared communication channel for a longer period for transmitting a fixed-size packet to the AP, thus reducing the channel efficiency and overall system performance. This thesis addresses this challenging problem in multi-rate WLANs by proposing two cooperative Medium Access Control (MAC) protocols, namely Busy Tone based Cooperative MAC (BTAC) protocol and Cooperative Access with Relay’s Data (CARD) protocol. Under BTAC, a low data-rate sending station tries to identify and use a close-by intermediate station as its relay to forward its data packets at higher data-rate to the AP through a two-hop path. In this way, BTAC can achieve cooperative diversity gain in multi-rate WLANs. Furthermore, the proposed CARD protocol enables a relay station to transmit its own data packets to the AP immediately after forwarding its neighbour’s packets, thus minimising the handshake procedure and overheads for sensing and reserving the common channel. In doing so, CARD can achieve both cooperative diversity gain and cooperative multiplexing gain. Both BTAC and CARD protocols are backward compatible with the existing IEEE 802.11 standards. New cross-layer mathematical models have been developed in this thesis to study the performance of BTAC and CARD under different channel conditions and for saturated and unsaturated traffic loads. Detailed simulation platforms were developed and are discussed in this thesis. Extensive simulation results validate the mathematical models developed and show that BTAC and CARD protocols can significantly improve system throughput, service delay, and energy efficiency for WLANs operating under realistic communication scenarios.

Wireless sensor networks (WSNs) are used for event detection and data collection in a plethora of environmental monitoring applications. However a critical factor limits the extension of WSNs into new application areas: energy constraints. This thesis develops self-organising energy efficient data collection and dissemination protocols in order to support WSNs in event detection and data collection and thus extend the use of sensor-based networks to many new application areas. Firstly, a Dual Prediction and Probabilistic Scheduler (DPPS) is developed. DPPS uses a Dual Prediction Scheme combining compression and load balancing techniques in order to manage sensor usage more efficiently. DPPS was tested and evaluated through computer simulations and empirical experiments. Results showed that DPPS reduces energy consumption in WSNs by up to 35% while simultaneously maintaining data quality and satisfying a user specified accuracy constraint. Secondly, an Adaptive Detection-driven Ad hoc Medium Access Control (ADAMAC) protocol is developed. ADAMAC limits the Data Forwarding Interruption problem which causes increased end-to-end delay and energy consumption in multi-hop sensor networks. ADAMAC uses early warning alarms to dynamically adapt the sensing intervals and communication periods of a sensor according to the likelihood of any new events occurring. Results demonstrated that compared to previous protocols such as SMAC, ADAMAC dramatically reduces end-to-end delay while still limiting energy consumption during data collection and dissemination. The protocols developed in this thesis, DPPS and ADAMAC, effectively alleviate the energy constraints associated with WSNs and will support the extension of sensorbased networks to many more application areas than had hitherto been readily possible.

In this thesis investigations were performed into digital signal processing (DSP) algorithms for coherent optical fibre transmission systems, which provide improved performance with respect to conventional systems and algorithms. Firstly, an overview of coherent detection and coherent transmission systems is given. Experimental investigations were then performed into the performance of digital backpropagation for mitigating fibre nonlinearities in a dual-polarization quadrature phase shift keying (DP-QPSK) system over 7780 km and a dual-polarization 16- level quadrature amplitude modulation (DP-QAM16) system over 1600 km. It is noted that significant improvements in performance may be achieved for a nonlinear step-size greater than one span. An approximately exponential relationship was found between performance improvement in Q-factor and the number for required complex multipliers. DSP algorithms for polarization-switched quadrature phase shift keying (PS-QPSK) are then investigated. A novel two-part equalisation algorithm is proposed which provides singularity-free convergence and blind equalisation of PS-QPSK. This algorithm is characterised and its application to wavelength division multiplexed (WDM) transmission systems is discussed. The thesis concludes with an experimental comparison between a PS-QPSK transmission system and a conventional DP-QPSK system. For a 42.9 Gb/s WDM system, the use of PS-QPSK enabled an increase of reach of more than 30%. The resultant reach of 13,640 km was, at the time of publication, the longest transmission distance reported for 40 Gb/s transmission over an uncompensated link with standard fibre and optical amplification.

Research in the area of CVD single crystal diamond plates of which only recently has been made commercially available saw significant advancements during the last decade. In parallel to that, detonation nanodiamond (DND) particles also now widely made accessible for requisition are provoking a lot of scientific investigations. The remarkable properties of diamond including its extreme hardness, low coefficient of friction, chemical inertness, biocompatibility, high thermal conductivity, optical transparency and semiconducting properties make it attractive for a number of applications, among which electronic and micro electrical-mechanical systems devices for chemical and biological applications are few of the key areas. A detailed knowledge of diamond devices at the prototypical stage is therefore critical. The work carried out encapsulated in this thesis describes the employment of the nanometer-scale diamond structures for the design, fabrication and testing of electronic devices and micro electrical-mechanical system (MEMS) structures for chemical sensing applications. Two major approaches are used to achieve engineering novelty. The first type being devices based on single crystal diamond substrates, which include state of the art δ-doped single crystal diamond Ion Sensitive Field Effect Transistor with an intrinsic layer capping the delta-doped layer for pH sensing and the fabrication and characterization of a triangular-face single crystal diamond MEMS. A comprehensive set of characterisations was systematically performed on the delta ISFET devices. Cyclic Voltammetry has been used to determine the devices’ potential window determining the limits of the applied potential for the Current-Voltage measurements. In solutions of different pH levels, an improved sensitivity of 55mV/pH compared to cap-less design in a previous study is taken as the salient figure of merit. Electrochemical Impedance Spectroscopy sheds some light on device performance in terms of flatband voltages and conduction pathways through circuit modelling. Improved ISFET characteristics such as lower flat-band voltage at 3.74V, simpler conduction paths and drain current saturation onsets show the chosen design is correct and advances delta-doped diamond ISFET research and development work. For the single crystal diamond cantilever, the theoretical modelling supports the triangular-face design to be a better option, generating 3x greater deflections in relation to the conventional rectangular-face design, when operated as a static mode sensor. Based on experimental characterisation methods such as Raman and Energy Dispersive Spectroscopy, the focusedion beam only milling technique inflicts minimum damage to the beam structure. In the second approach, a novel hybrid device idea was conceived and implemented using off-the-shelf silicon ISFETs and cantilevers with a coat of nanodiamond particles on the ‘active area’ surfaces of the respective devices. These nanodiamond-coated silicon devices exhibit high sensitivity for tracing threat signatures such as explosive precursors and analogues with the former in both liquid and vapour medium, and the latter in the vapour phase. The nanodiamond-gated ISFET shows a voltage response of a commendable maximum voltage shift of ~90 mV throughout 0 to 0.1M concentration range of NO2 - and ClO3 - solutions. In the vapour phase detecting 2,4-DNT, a sensitivity of ~20mV/0.4ppm is observed. The nanodiamond-coated silicon cantilever demonstrates a performance advantage of 7.4 Hz/ppb to 1.7 Hz/ppb in a previous study. Fourier Transform Infra-red spectroscopy was carried out on the nanodiamond surfaces hosted by potassium bromide (KBr) discs to ascertain the vapour chemisorption. With the fabrication technique simplified, commercialisation of these proof-of-concept devices should be less time consuming thus enabling quicker deployment of diamond-based surface sensing technology.

The rapid growth in interactive bandwidth-hungry services demands ever higher capacity at various stages of the optical network, leading to a potential capacity exhaust, termed the capacity crunch. The main aim of the research work described in this thesis was to help solve the potential capacity crunch by exploring techniques to increase the data rate, spectral efficiency and reach of optical fibre systems. The focus was on the use of advanced signal modulation formats, including optical time-division multiplexing (OTDM), quadrature phase shift keying (QPSK), and 16-state quadrature amplitude modulation (QAM16). QPSK and QAM16 modulations formats were studied in combination with coherent detection and digital signal processing (DSP) for the compensation of transmission impairments. In addition, return-to-zero (RZ) pulses were explored to increase the tolerance towards nonlinearity for coherently detected signals, and nonlinearity compensation (NLC) through the DSP. Initially, to maximise the bit-rate, research was focused on the study of OTDM transmission at 80Gbit/s with the aim to optimise the phase difference between the adjacent OTDM channels. A new technique to achieve bit-wise phase control using a phase-stabilised fibre interferometer was proposed. Faced with a limited fibre capacity, the need to maximise the spectral efficiency became paramount, and thus the need to use phase, amplitude and polarisation domains for signal transmission. In combination with coherent detection the research focused on the performance of optical fibre systems using QPSK and QAM16 modulation formats, including their generation, transmission and detection in single-channel and WDM regimes. This included the study of the impact of pulse shapes, and the mitigation of linear and nonlinear transmission impairments with receiver-based DSP at bit-rates ranging from 42.7 to 224Gbit/s. The technique demonstrated for bit-wise phase control for OTDM was successfully used to demonstrate a new method for QAM16 signal generation. Longest transmission distances (up to 10160km in 112Gbit/s QPSK, 4240km in 112Gbit/s QAM16, and 2000km in 224Gbit/s QAM16) have been achieved with the use of NLC and RZ pulses. The efficiency of these two techniques is explored through a comprehensive set of experiments in both single-channel and WDM transmission experiments. The results can be used in the design of future optical transmission systems.