No description available.

The thesis is concerned with the detection of digital data transmitted over voice frequency channels such as telephone circuits and HF radio links, where the main impairment is additive noise and intersymbol interference, and the latter may be time-invariant or may vary slowly with time. The characteristics of these channels are briefly reviewed and a survey of the most important known detection techniques is presented. The thesis includes also a detailed study of quadrature amplitude modulated (QAM) signals transmitted over voice-channels, both, when the transmission path has time-invariant characteristics and when it introduces Rayleigh fading into the transmitted data signal. Based on this study, baseband models of QAM systems are suggested for use when these are to be computer simulated. A systematic study of channel models is carried out here. The transmission and detection of baseband signals over telephone circuits in the presence of frequency offset is investigated and a baseband signal generated by means of Hilbert transform pairs is suggested for this purpose. It is shown that this signal exhibits theoretical and experimental equivalence to a QAM signal. Several near-maximum likelihood detection techniques have been developed for the detection of digital data signals serially transmitted at 19200 bit/s over telephone lines and at 9600 bit/s over HF radio links. The performance of the detection systems has been evaluated by computer simulation and is given in terms of their tolerance to additive white Gaussian noise.

The promise of digital hardware economies (due to recent advances in VLSI technology), has focussed much attention on more complex and sophisticated speech coding algorithms which offer improved quality at relatively low bit rates. This thesis describes the results (obtained from computer simulations) of research into various efficient (time and frequency domain) speech encoders operating at a transmission bit rate of 16 Kbps. In the time domain, Adaptive Differential Pulse Code Modulation (ADPCM) systems employing both forward and backward adaptive prediction were examined. A number of algorithms were proposed and evaluated, including several variants of the Stochastic Approximation Predictor (SAP). A Backward Block Adaptive (BBA) predictor was also developed and found to outperform the conventional stochastic methods, even though its complexity in terms of signal processing requirements is lower. A simplified Adaptive Predictive Coder (APC) employing a single tap pitch predictor considered next provided a slight improvement in performance over ADPCM, but with rather greater complexity. The ultimate test of any speech coding system is the perceptual performance of the received speech. Recent research has indicated that this may be enhanced by suitable control of the noise spectrum according to the theory of auditory masking. Various noise shaping ADPCM configurations were examined, and it was demonstrated that a proposed pre-/post-filtering arrangement which exploits advantageously the predictor-quantizer interaction, leads to the best subjective performance in both forward and backward prediction systems. Adaptive quantization is instrumental to the performance of ADPCM systems. Both the forward adaptive quantizer (AQF) and the backward oneword memory adaptation (AQJ) were examined. In addition, a novel method of decreasing quantization noise in ADPCM-AQJ coders, which involves the application of correction to the decoded speech samples, provided reduced output noise across the spectrum, with considerable high frequency noise suppression. More powerful (and inevitably more complex) frequency domain speech coders such as the Adaptive Transform Coder (ATC) and the Sub-band Coder (SBC) offer good quality speech at 16 Kbps. To reduce complexity and coding delay, whilst retaining the advantage of sub-band coding, a novel transform based split-band coder (TSBC) was developed and found to compare closely in performance with the SBC. To prevent the heavy side information requirement associated with a large number of bands in split-band coding schemes from impairing coding accuracy, without forgoing the efficiency provided by adaptive bit allocation, a method employing AQJs to code the sub-band signals together with vector quantization of the bit allocation patterns was also proposed. Finally, 'pipeline' methods of bit allocation and step size estimation (using the Fast Fourier Transform (FFT) on the input signal) were examined. Such methods, although less accurate, are nevertheless useful in limiting coding delay associated with SRC schemes employing Quadrature Mirror Filters (QMF).

This thesis is on a new method for improving the convergence speed of the normalised least mean square (NLMS) algorithm when the input is a coloured signal, such as speech, that can be decorrelated using linear prediction. The proposed method gives a significant improvement in the convergence speed and requires very little additional computation in terms of arithmetic operations and memory space. It is also very easy to implement. An important aspect of the proposed method is its inherent stability irrespective of the order of the prediction-error filter or the manner in which it is adapted. This allows the proposed method to be used without any restrictions beyond those of the conventional NLMS algorithm. Following the stability analysis, a further improvement to the basic proposed method is suggested. This improvement is restricted to the cases where the input signal is decorrelated by a prediction-error filter of up to order two. The proposed method finds immediate application in acoustic echo cancellation in hands-free telephones where the impulse response of the system (echo path) to be identified is very long and the identification has to be done in real time. Such an application requires an algorithm with low computational complexity and a fast rate of convergence. In general, the proposed method can be used where the input coloured signal can be decorrelated using linear prediction.

The work described in this thesis proposes and investigates further use of scrambling in industrial analogue and digital monochrome video systems. This scrambling inevitably entails some signal modification. Providing the receiver is able to distinguish between the original and the scrambled signal, regardless of which one was transmitted, more efficient signal exploitation is possible. This more efficient signal exploitation is performed at the expense of the inherent redundancy present in the analogue and di ital signals. Analogue video signals are usually of a highly correlative nature and, this characteristic is exploited, in this thesis, by enabling them to be unwitting data carriers. The video signal is made the data carrier while the data gets a free ride. Each scan-line of the video signal is sampled, and blocks of pels are scrambled or not by modulo masking, depending on whether the data necessary for transmission is a logical 1 11 or 1 01 respectively. Prior to transmission the combined data and video sequence is converted into a continuous signal with a bandwidth that is no greater than that of the original video signal. From the knowledge of the original and the modified signal statistics, the receiver is able to perform the inverse operation of the transmitter, recovering the video signal and the data. Three novel systems are proposed for embedding data into analogue pictures. Two of these systems are capable of supporting an average of 17430 and 8713 bits per (256x256) image LV respectively, with excellent recovered picture quality. The third system produced a constant number of bits per image, with a slight degradation in the recovered picture quality but, with a capability of conveying up to about 0.5 mega bits/sec of data. The idea and technique of embedding data into analogue signals was then carried on to the digital method of coding video signals using differential pulse code modulation. However, the scrambling technique here was used to obtain a novel switched quantization scheme, with forward estimation, without the necessity of sending any side information. Scrambling was performed on the quantizer output levels by inversion. Initially, experiments were carried out using fixed length code words with one and two dimensional predictors. Blocks of quantizer output levels are scrambled, or not, depending on which quantizer was used in encoding the video signal. Hence the switching information was carried by the quantized block of error signals. This type of set-up produced only modest improvements. The quantizers were then altered by using a different number of levels and the switching information was carried one block of quantized error signals in advance. As a result, the average bit rate was reduced to about 2.7 bits/pel using a one dimensional predictor with exceptionally good subjective picture quality. When used with a two dimensional predictor, the scheme produced an average bit rate of about 1.7 bits/pel with excellent subjective picture quality.

This thesis describes the design, implementation, and performance evaluation of a communications software architecture designed in the first place for the construction of distributed computing systems based on high bandwidth local area networks, typified by the Cambridge Ring and Ethernet. This architecture consists of protocols and program interfaces, within the communication software and to the network, intended to support conveniently the development of a wide variety of distributed applications. However the architecture has been designed so as to allow its use also over multiple and varied data communication facilities, including wide area networks. The differences between this architecture and that conventionally found in wide area networks are discussed; reliability and performance issues concerning adequate structuring of the communication software suitable for local networks as well as wide area networks are examined. The implementation of this architecture developed for a local area network is evaluated, and its use on a wide area network is discussed.

Petri nets were introduced by C.A. Petri as a theoretical model of concurrency in which the causal relationship between actions, rather than just their temporal ordering, can be represented. As a theoretical model of concurrency, Petri nets have been widely successful. Moreover, Petri nets are popular with practitioners, providing practical tools for the designer and developer of real concurrent and distributed systems. However, it is from this second context that perhaps the most widely voiced criticism of Petri nets comes. It is that Petri nets lack any algebraic structure or modularity, and this results in large, unstructured models of real systems, which are consequently often intractable. Although this is not a criticism of Petri nets per se, but rather of the uses to which Petri nets are put, the criticism is well taken. We attempt to answer this criticism in this work. To do this we return to the view of Petri nets as a model of concurrency and consider how other models of concurrency counter this objection. The foremost examples are then the synchronisation trees of Milner, and the traces of Hoare, (against which such criticism is rarely, if ever, levelled). The difference between the models is clear, and is to be found in the richness of the algebraic characterisations which have been made for synchronisation trees in Milner's Calculus of Communicating Systems (CCS), and for traces in Hoare's Communicating Sequential Processes (CSP). With this in mind we define, in this thesis, a class of high level Petri nets, High Level Petri Boxes, and provide for them a very general algebraic description language, the High Level Petri Box Algebra, with novel ideas for synchronisation, and including both refinement and recursion among its operators. We also begin on the (probably open-ended task of the) algebraic characterisation of High Level Petri Boxes. The major contribution of this thesis is a full behavioural characterisation of the High Level Petri Boxes which form the semantic domain of the algebra. Other contributions are: a very general method of describing communication protocols which extend the synchronisation algebras of Winskel; a recursive operator that preserves finiteness of state (the best possible, given the generality of the algebra); a refinement operator that is syntactic in nature, and for which the recursive construct is a behavioural fix-point; and a notion of behavioural equivalence which is a congruence with respect to a major part of the High Level Petri Box Algebra.

No description available.

No description available.

No description available.