5G offers several new improvements compared to its predecessors, such as higher speeds, lower latency and larger capacity. In order to ensure this, time of arrival algorithms are used in the Radio Access Networks for several purposes. Some of them are 3D positioning of User Equipment and synchronization. A big challenge for Time of Arrival algorithms are environments with multi-path propagation, especially when they use two-way measurements, as the algorithm may pick different propagation paths depending on the direction. This will result in a measurement error caused by asymmetry which leads to a degraded algorithm performance. This thesis investigates how time of arrival algorithms can be improved when facing asymmetry. Two different improvements are tested, the first is how using the same beamforming weights in the transmitter and the receiver impacts the assumption of channel reciprocity and the other is how the selection of the crosscorrelation peak can be altered. The simulations were performed using different Tapped Delay Line channel models in order to mimic a multipath propagation environment. The results show that the suggested peak search criterion could reduce measurement errors originating from asymmetry but also that the simulator was not equipped to test and verify that using the same beamforming weights would reduce asymmetry errors.

Spatial diversity is an effective diversity technique used in multiple antenna wireless systems. It requires that antenna elements be sufficiently distributed by at least ten times the signal wavelength (10λ). Sufficient spatial distribution of antenna elements renders performance improvements through diversity gain. In consumer electronics, antenna elements are often located in close proximity, < λ, due to form size constraints. Therefore performance improvement through spatial diversity is limited even though multiple antenna systems are employed. The work in this dissertation addresses the use of power line communication as a method to spatially distribute antenna elements of a multiple antenna wireless system. Addressing this problem requires the cooperative use of both wireless and powerline channels to form a hybrid channel. A hybrid powerline-wireless channel is defined as the concatenation of a powerline channel and a wireless channel to establish a unified channel. A sequence-based channel sounding approach using maximum length sequences and software-defined radio was designed, implemented andapplied to attain channel transfer characteristics of the hybrid channel. The measurement campaign was carried out for an indoor low-voltage level powerline network. The attained results include the channel impulse responses, frequency responses and power delay profiles. Coherence bandwidth, average channel attenuation and power delay profile timing delays are channel parameters that were extracted from the measurements. The extracted channel parameters reveal that the hybrid channel is a challenging medium for data transfer and communication applications. Based on the attained hybrid channel characterisation parameters, a physical layer structure for a 1 × 2 single-input multiple-output orthogonal frequency division multiplexing (SIMOOFDM) hybrid powerline-wireless system was designed and implemented. An investigation of the bit-error-rate performance of single-carrier and multi-carrier modulation for a multipath powerline channel was carried out in simulation. Using OFDM as a suitable candidate for multi-carrier modulation, it was found to outperform single-carrier modulation. OFDM was used as the channel access method for the hybrid channel. Source encoding and decoding of the physical layer structure was designed to be robust against inherent interferences of the hybrid channel. Selection combining, equal gain combining and maximal ratio combining methods were designed and implemented. Estimation methods for error rate, data transfer rate and SNR of the SIMO-OFDM system were designed and implemented. Capacitive signal coupling was used to interface the powerline channel to the wireless propagation environment, hence establishing the hybrid channel. This method also allowed powerline transmission to be performed at a much lower frequency than wireless transmission. An experiment was conducted to investigate the effect of spatial distribution on the 1 × 2 SIMO-OFDM hybrid powerline-wireless system for indoor environments. The results of the experiment were error rate curves produced for different spatial distribution lengths through a powerline channel for each signal combining method. Error rate performance improvements in the SIMO-OFDM system were achieved with an increase of antenna element spatial distribution due to lowered signal envelope correlation. Small but yet notable diversity gains were observed by the increase in the slope of the produced error rate and signal-to-noise ratio curves for each signal combining method. Experimental parameters and apparatus placed a limitation on the achievable spatial distribution of diversity branches, hence the achievable diversity gains. This was a result of overwhelming inherent interferences of the hybrid channel / Dissertation (MEng)--University of Pretoria, 2020. / CeTEIS / Electrical, Electronic and Computer Engineering / MEng / Unrestricted

In real-time signal processing there is a constraint to finish processing of an audio signal before the next audio segment is received. This makes it important to have signal processing algorithms with low computational complexity while still maintaining high quality results. This thesis presents methods for audio signal processing used in real-time systems. The publications presented cover areas of noise reduction, network echo cancellation, noise dosimeter measurements and voice analysis. A method for speech enhancement is presented with low amounts of speech distortion. The audio signal is split into several subbands, covering different frequency regions. For each subband, the noise level is estimated. A signal gain is calculated by comparing the total signal level with the noise level for each subband. The method presented here, improves performance compared to previously similar methods. Improvement is especially found in multi-speaker and noise-only scenarios. When communicating on a telephone line, network echo is introduced by hybrids in the network. In cases where multiple echo sources exist, the time range for echoes can be quite long. In devices with limited storage, it is difficult to get good echo cancellation in such cases. This thesis presents a method for network echo cancellation suited for use in a device with a larger external memory. Exposure to high noise levels will have negative health effects and methods for measuring noise level exposure is important. Included in this thesis is a study that remove the influence of own voice in noise dose measurements. For certain medical conditions it is beneficial with daily voice exercises. Methods for grading voice in four different exercises is presented, based on pitch and loudness. Evaluation is done in real-time on test medical device.

This paper deals with the implementation and performance evaluation of adaptive filtering algorithms for noise cancellation without reference signal. Noise cancellation is a technique of estimating a desired signal from a noise-corrupted observation. If the signal and noise characteristics are unknown or change continuously over time, the need of adaptive filter arises. In contrast to the conventional digital filter design techniques, adaptive filters do not have constant filter parameters, they have the capability to continuously adjust their coefficients to their operating environment. To design an adaptive filter, that produces an optimum estimate of the desired signal from the noisy environment, different adaptive filtering algorithms are implemented and compared to each other. The Least Mean Square LMS, the Normalized Least Mean Square NLMS and the Recursive Least Square RLS algorithm are investigated. Three performance criteria are used in the study of these algorithms: the rate of convergence, the error performance and the signal-to-noise ratio SNR. The implementation results show that the adaptive noise cancellation application benefits more from the use of the NLMS algorithm instead of the LMS or RLS algorithm.

A magnetoencephalogram (MEG) is a non-invasive tool for measuring neuronal activity with millisecond temporal resolution. In this study, MEG measurements were recorded as a subject carried out a simple, repetitive, numerical task: deciding whether a number is even or odd. Signal processing techniques were applied to the MEG data so as to characterise the spatial and temporal dynamics of the brain during the decision-making process. The data is first preprocessed using Independent Component Analysis (ICA) and other semiautomated methods. The data is then segmented into trials. Evoked fields or event-related fields (ERFs), the classical measure of brain activity, are found by averaging all the trials in the time domain. These responses are typically phase locked to the stimulus. Induced potentials or oscillatory rhythms that are not necessarily phase-locked to the stimulus are found by averaging the time-frequency representations (TFRs) over all the trials. The TFRs were found using the Wavelet Transform. The results show that typical ERF components are present just after the onset of each stimulus. These waveforms indicate that the following sequence of cognitive events occur: mental matching of the stimulus with previously experienced stimuli (N100); higher-order perceptual processing modulated by attention (P200); and “Go-NoGo” control procedure which initiates or inhibits the motor response (N200). The P200 response also indicates that parity information may be retrieved directly from memory rather than being extracted by means of a mental calculation strategy. Time-frequency plots of the data show pronounced synchronisation in the beta-band as the subject is actively concentrating on the mental task. Thereafter, beta band desynchronisation occurs as the motor response is carried out. Activity is pronounced in the left general interpretive area with a latency of around 650ms. This confirms the fact that the brain is lateralised according to function. One important avenue for further research would be to explore source reconstruction using beamforming techniques. This would enable researchers to pinpoint neuronal sources with greater accuracy. Furthermore, functional connectivity analysis may be a useful means of elucidating how information is transmitted and integrated across brain networks. Overall, there is much scope for future work.

In order to ensure secure communication in digital military radio systems, multiple methods are used to protect the transmission from being intercepted by enemy electronic warfare systems. An intercepted transmission can be used to estimate several parameters of the transmitted signal such as its origin (position or direction) and of course the transmitted message itself. The methods used in traditional electronic warfare direction-finding systems have in general poor performance against wideband low power signals while the considered correlation-based time-difference of arrival (TDOA) methods show promising results. The output from a TDOA-based direction-finding system using two spatially separated receivers is the TDOA for the signal between the receiving sensors which uniquely describes a hyperbolic curve and the emitter is located somewhere along this curve. In order to measure a TDOA between two digital radio receivers both receiver systems must have the same time and frequency references to avoid degradation due to reference imperfections. However, in some cases, the receivers are separated up to 1000 km and can not share a common reference. This is solved by using a reference module at each of the receiver sites and high accuracy is achieved using the NAVSTAR-GPS system but, still, small differences between the outputs of the different reference modules occurs which degrades the performance of the system. In a practical electronic warfare system there is a number of factors that degrade the performance of the system, such as non-ideal antennas, analog receiver filter differences, and the analog to digital converter errors. In this thesis we concentrate on the problems which arises from imperfections in the reference modules, such as time and frequency errors.

Satellite precipitation estimation at high spatial and temporal resolutions is beneficial for research and applications in the areas of weather, flood forecasting, hydrology, and agriculture. In this research, image processing and pattern recognition tools are incorporated into the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks Cloud Classification System (PERSIANN-CCS) methodology to enhance satellite precipitation and rainfall estimation. The enhanced algorithm incorporates five main steps to derive precipitation estimates: 1) segmenting the satellite infrared cloud images into patches; 2) extracting features from the segmented cloud patches; 3) feature selection or dimensionality reduction; 4) categorizing the cloud patches into separate groups; and 5) obtaining a relationship between the brightness temperature of cloud patches and the rain- rate (T-R) for every cluster. In this study, in addition to the features utilized for cloud patch classification, wavelet and lightning features are also extracted. The lightning feature is defined as the number of flashes occurring within 15 minutes of the nominal IR image scan. Both feature selection and dimensionality reduction techniques are examined to reduce the dimensionality as well as diminish the effects of the redundant and irrelevant features. The feature selection technique includes a Feature Similarity Selection (FSS) method and a Filter-Based Feature Selection using Genetic Algorithm (FFSGA). The Entropy Index (EI) fitness function is used to evaluate the feature subsets. Furthermore, Independent Component Analysis (ICA) was examined and compared to other linear and nonlinear unsupervised dimensionality reduction techniques to reduce the dimensionality and increase the estimation performance. In addition to a Self Organizing Map (SOM) neural network, the link-based cluster ensemble method is also examined in this research. In the final step, the Median Merging (MM) and Selected Curve Fitting (SCF) techniques are incorporated. After applying a Probability Matching Method (PMM) to each single patch and obtaining the T-R for each patch, a Median Merging technique which computes the median rain-rate for a given temperature is applied. A Selected Curve Fitting (SCF) procedure is also used to obtain the T-R for each cluster. The results show that the enhanced algorithm incorporating the above techniques improves precipitation estimation.

Despite shifting towards mm-wave bands, the sub 6-GHz band will continue tobe a fundamental spectral band in 5G. Yet, the severe crowdedness of this bandmakes a well constrained spectrum one of the critical 5G requirements. A wellconstrained spectrum means that the communications regimes should dwell politelywithin their dedicated spectral bands and not interfere with other systems workingon neighboring bands. Consequently, communications community seeks convenientmodulation schemes.Accordingly, high Out Of Band (OOB) emission phenomenon in Orthogonal Fre-quency Division Multiplexing (OFDM) is unfavorable for some regimes operating in5G. Therefore, to legitimize OFDM with all 5G regimes, we need to suppress OFDMOOB emission.Since the discontinuous nature of the OFDM signal is the main reason for the highOOB emission, one solution is to render the discontinuous OFDM signal continuous.Two factors control this discontinuity: the physical shape of the modulated signaland the correlation property of the data symbols that modulate the OFDM signal.While most of the traditional approaches focus on reshaping the OFDM signalto render it continuous, in this work we give our attention to the spectral precod-ing approaches. These approaches manipulate the correlation property of the datasymbols to control the high OOB emission in OFDM.On the other hand, tweaking the correlation property of the modulating datasymbols will violate their orthogonality. This violation will yield in-band interfer-ence within the OFDM signal which would degrade the bit error performance of thereceived data.The thesis explains the spectral precoding techniques from conceptual and math-ematical point of view. We discuss the OOB emission suppression capability of theprecoding techniques and study their drawbacks and limitations. We provide ana-lytical trade o study between precoding approaches and classical OFDM treatmentapproaches at the level of OOB emission suppression and in-band interference. Weshow that the in-band interference in precoding techniques is independent on thecommunications channel behavior contrary to that of classical techniques. More-over, we dene the optimal precoder that minimizes the in-band interference. Con-sequently, we design a novel practical precoder that approaches the performanceof the optimal precoder. Furthermore, we analyze the complexity of the precodingapproaches and study the implementation computational requirements.Finally, we test the real time performance of these precoding techniques usingSoftware Designed Radio (SDR) Universal Software Radio Peripherals (USRPs). Wespotlight the hardware limitations and show that despite these limitations, the spec-tral precoder is able to suppress the OOB emissions by tens of decibels. We check the reliability of spectral precoding in practical over air communications systems bysetting up the rst spectral precoding proof of concept prototype. The prototypeproves that precoded OFDM systems cause less OOB interference on neighboringcommunications systems.

The number of Nano satellites launched into space is increasing rapidly each year. All satellites need a ground station with a baseband to communicate with. Current commercial basebands for TT&amp;C (telemetry, tracking and command) are using proven designs and are too expensive to be used for Nano satellites. This thesis investigates the feasibility to use a SDB (Software-Defined Baseband) for TT&amp;C with Nano satellites in low earth orbit. It describes the making of a transceiver which is validated with simulated data,recorded data and live satellite passes. It also describes the implementation of a GPS (Global PositioningSystem), TCR (Time Code Receiver) and TCP/IP (Transmission Control Protocol/Internet Protocol)interface as well as the process of testing for EMC (Electromagnetic Compatability) requirements. A half-duplex transmit/receive system is also proposed. The SDB functioned acceptable for telemetry with Nano satellites with up to 90% of the data packets being received depending on which satellite that wasbeing tracked. The SDB needs further development, validation and certification to compete with today’scommercial basebands.

Digital communications continues to be a relevant Field of study as new technologies appear and old methodologies get revisited or renovated. The goal of this research is to look into the old digital communication scheme of Bell 202 [67] used by APRS and improve software based demodulation performance. Improved performance is defined by being able to correctly decode more packets in an efficient, real time, manner. Most APRS demodulation is currently done using specialized hardware since that yields the best performance. This research shows that through using Sivan Toledo's javAX25 [72] software package, new demodulation algorithms can be implemented that decode more Bell 202 encoded AX.25 packets than the existing software could. These improvements may help drive the adoption of software demodulation since it is a low cost alternative to specialized hardware.