Search

The resultant relative permittivity of 3D printed polylactide (PLA) substrates with varying percentage volumes of PLA are measured using a two microstrip line method. The resultant relative permittivity of the PLA substrates is also predicted using five existing heterogenous mixture models. The measured and predicted values of relative permittivity are compared to identify the degree of fit of the mixture models to the measured values. The measured relative permittivity was found to increase as the percentage volume of PLA was increased. The measured values also increased with increasing signal frequency. The mixture models showed varying degrees of fit with two models, Rayleigh (RAY) and Refractive Index (RI), showing the best fit out of the five models. None of the models considered the frequency dependence of the relative permittivity. The results allow for the identification of opportunities for future work related to frequency dependent mixture models and detailed testing of 3D printed substrates.

Diabetes is a chronic disease where the body does not produce sufficient insulin hormone to control glucose levels. To prevent health complications arising, diabetics must continuously monitor their blood glucose. Currently, blood glucose levels are mainly monitored using invasive methods which can discourage regular testing. This paper proposes a non-invasive double U-slot patch antenna system to monitor blood glucose levels. COMSOL was used to simulate the port impedance and resonant frequency shifts for varying glucose blood vessel concentrations. The impact of sensor placement, relative to the blood vessel, on glucose level sensitivity was explored considering the relationship between blood glucose concentration and RF-based predictors. Results demonstrate that the lumped port magnitude resonant frequency shift correlates with the range of glucose concentrations investigated. The sensitivity is better at lower frequencies for the designed sensor, and the sensor is relatively insensitive to rotation shifts and horizontal shifts, while vertical shifts, representing more fatty tissue impact the sensitivity.

As infrastructure ages, grid operators across the world are becoming more cognizant of the need for monitoring their transmission infrastructure. The geographic extent of the transmission system, however, makes this a difficult and expensive task with thousands of components requiring visual inspection to identify faults which can lead to potentially catastrophic failures. This paper describes the use of Deep Neural Networks to automatically detect areas of concrete damage on utility poles in a European utility from photographs. This is beneficial to reduce time spent in the field as well as variability in between human assessment. The results show that even with a small dataset for training, the network is able to identify new damage with a high level of precision.

It is well known that electric loads exhibit repetitive short term behaviour which has been the focus of load forecasting research for many years. In this study a high-order non-causal iterative learning load frequency controller (HOILC) for a multi-source generation scheme in an intensive corporate microgrid setting, subject to repetitive electric loading is proposed. The HOILC strategy is compared to that of an optimally tuned integral strategy and fuzzy controller showing significantly better performance for the periodic disturbances imposed by the batch loading processes even under parameter variations.

Internet of Farming (IoF) is the application of Internet of Things (IoT) to agriculture use cases. Organic channels are any hydrocarbon based media. When organic channels are used to communicate data, they are termed Organic Communication Channels (OCCs). The accomplishments of Human Body Communication (HBC) in Body Area Networks (BANs) suggests OCCs’ great potential when applied to cyber-physical systems. For example, soil-based and vegetation-based OCCs can be integrated into IoF scenarios. However other OCCs are less studied. This paper characterizes vegetation-based and soil-based OCCs through analysis of channel measurements and derivation of path loss equations. Presented results suggest log-distance path median path loss models hold for herbaceous vegetation-based channels. Furthermore, the results suggest vegetation-based channels were more frequency-selective than the soil-based channels examined.

Wireless wearable applications which allow for remote biomedical data gathering, analysis and visualisation present a significant opportunity to improve patient care [10] . However, the design and manufacture of wearable systems does present a challenge as relates to their flexibility [11] and ensuring their safe operation [12] – [13] . In this work our goal was to examine the behaviour of a traditional Planar Inverted-F Antenna (PIFA) structure across the frequency band of 700MHz - 4GHz. The specific scenario was a wireless wearables application where nylon was used as a flexible substrate and the antenna was placed directly against the underlying tissue. The RF characteristics were considered for the structure under the action of applied forces which produced multiplanar deformations that correspond to the natural modalities of the antenna and underlying tissue. A shift in the characteristics of the antenna structure was noted due to the change in substrate and effects of the underlying tissue, which was not unexpected [13] . However our specific interest was in understanding how the physical changes to the antenna geometry and surrounding tissue resulted in changes to the RF behaviour. It was observed that the electric field energy varied based on the zone of application of the applied force to the antenna surface. Furthermore, there was greater concentration of the electric field energy into the underlying tissue at higher frequencies.

Power Transformers are the most critical assets in a power system after generators. They are costly, have long lead times for acquisition and their failure compromises the security of the system. As a result of this, a number of testing methods have been developed over time to appraise their condition with a view to extending their lives as well as predicting their end of life as early as possible. Typically these tests are administered at different times depending on the maintenance scheme adopted by the utility and analysed individually. This paper describes the development of a multi-dimensional fingerprint for transformer health estimation. Test results from a number of testing methods are analysed and a binary code system developed to more precisely pinpoint the transformer state.

Dynamic spectrum access deployments potentially enable non-deterministic operation across the electrospace, minimally defined by frequency, space and time axes. This poses several challenges for traditional spectrum management, particularly approaches to spectrum measurement and planning, as well as compliance enforcement. Interference limit policies such as the recently proposed harm claim threshold approach, shift from the traditional transmitter-centric approaches to regulation as they enable quantitative specification of acceptable interference environments for receiver operation. However, a key challenge to successfully leverage such policies is the ability to design and conduct practical and appropriate measurement plans for compliance enforcement. In this paper extreme value theory and the use of regional frequency analysis techniques are proposed for characterizing the receiver environment. A framework is presented for determination of the receiver environment, particularly for estimating order statistics and tail distributions, in support of the enforcement of harm claim thresholds. Measurement data and simulated scenarios are used to demonstrate the approach. Results highlight the importance of, as well as expand the work on, the use of extreme value theory and regional frequency analysis as a powerful mathematical tool for proving claims of interference and specifying appropriate confidence intervals for such claims. This paper provides valuable input into the current discussions on regulatory policies for deployment of harm claim thresholds, as well as for compliance enforcement.

Condition monitoring of power transformers are becoming critical as there is a thrust to operate the grid closer to its limits. Frequency response analysis has emerged as one of the principal methods in appraising the state of the physical structure of the transformer core and winding assembly. Changes in winding signatures, however, can still only be interpretted by experts. This paper explores the use of network synthesis as a means to interpret changes in winding signatures. A large set of transformer windings is simulated using complex RLC meshes. Neural networks are then trained to match the responses to network parameters. A number of different neural networks with varying parameters were tested and results show that Radial Basis Networks perform the best in correctly matching circuit parameters.

Transmission line models for simulations are necessaryfor designing and controlling modern power systems. However, the orders of these models are very large and theirsimulations are recognized to be computationally intensive. Thispaper compares ten model-order reduction methods with eightmeasures for a three phase transmission line system. Two testcases were developed, the first forces all model-order reductionmethods to reduce the original system to a reduced system withthe same order and the second keeps a constant chosen errorbound and compares the performance of the reduced obtainedsystems. The main differences between the different proceduresare identified and illustrated on quantitative levels.

Heart disease contributes significantly to annual deaths. Personal health monitoring systems can be useful in reducing mortality rates as well as for treatment and prevention of associated complications. A current challenge is the limited accessibility to devices which offer advanced feature extraction for cardiovascular diagnosis. In this work, investigations are presented using non-invasive and continuous monitoring using photoplethysmography (PPG), a simple, low-cost, popular choice for wearable devices, from which features other than pulse rate and blood pressure can be extracted from acquired PPG signals. The investigations outlined in this paper advance work in automated feature extraction for cardiovascular diagnosis, demonstrated through identification of the 'a', 'b' and 'e' waves derived from the second derivative of the PPG waveform, followed by calculation of indices associated with patient arterial stiffness using these waves. Results suggest that the implemented approach is a useful tool in assessing physiological features which could be utilized to determine cardiovascular conditions. The approach is important in providing insights into the emerging design space for cardiovascular monitoring and feature extraction as part of personalized healthcare systems.

Growth in the electric vehicle and renewable energyindustries has created a corresponding interest in energy storagesystems. Since storage systems have finite reserves, new toolsare required in order to properly examine and optimize theireffects in power system studies. Time stepped power flow analysesin particular are crucial to understanding the implications ofincorporating storage into a system. This paper deals withaugmenting the native MATLAB power flow capabilities toprovide time stepped power flow capabilities. The tool gives theuser the ability to model storage in steady state and control theirbehaviour in time based on system parameters. To illustrate itsuse, a simple load levelling study is performed using a modifiedIEEE 9 bus system.

Advances in power electronics, communications and control technologies employed by Electric Utilities, have been gradually shifting the way the grid is being operated for many years. These, together with small scale renewable energy and micro-generation capabilities are now forcing the philosophy of grid operation to be rethought from the ground up. The future is likely to be a network of power prosumers (producer-consumers) who must be managed by a central coordinating body. Often overlooked, however, is the fact that increased data can give us not only additional information about the system but also about customers. This paper posits the idea of building more social sensitivity into the the operation of the grid to enhance demand side management, emergency management and building Rawlsian principles into prosumer interactions with the utility. It also gives a conceptual overview of the infrastructure topologies required to enable this shift, but does not concern itself with auxiliary matters such as cyber security.

The Internet of Things offers tremendous potential to increasingly harness wireless and portable systems in everyday life. Low-dropout (LDO) linear regulators are essential components in numerous wireless and portable electronic devices and their performance in systems can be very important, given the need to power such devices particularly in scenarios with limited energy available. In this paper, the basic low–dropout regulator utilizing fractional order capacitors is investigated. A fractional order model is presented for the voltage regulator, and the corresponding transfer function is derived. From the transfer function, it is shown that the system stability performance depends upon the fractional order, offering increased flexibility in the design process, compared to the integer-order LDO regulator. Furthermore using fractional order components offers increased degrees of freedom in shaping the transfer function, beyond that of traditional pole and zero placement in integer order systems. Analytical findings are explored and validated via numerical simulations. This work provides valuable insights into the design space for using fractional order circuits to meet targeted LDO regulator system performance.

Globally, chronic kidney disease (CKD) burden healthcare systems. Current treatment systems in many countries have presented a serious socioeconomic situation for low- to middle-income countries. As such, monitoring mechanisms present a clinical need for effective diagnosis, prevention and treatment of numerous non communicable diseases, such as CKD. Non-invasive, continuous monitoring of renal function using a complementary split ring resonator (CSRR) microwave sensor confers potential benefits to meeting this clinical need. The sensor was sensitive to changes in the relative permittivity of a material under test with the electrical properties of blood, for the frequency range investigated in COMSOL Multiphysics®. The correlation of the electrical parameters with relative permittivity changes highlights the applicability of CSRRs as a sensor module for CKD monitoring.

Multi-channel access and spectrum agility represent significant challenges to temporal occupancy modelling for compliance enforcement of individual users/networks in opportunistic spectrum access (OSA) scenarios. In previous work, temporal occupancy estimation within spectrum bands was examined, when there was no requirement to track individual user/network behaviors. This approach was in line with the predominantly licensed-based access models where individual users/networks did not need to be tracked. However, depending upon the specific measurement objectives, discrimination between different users/networks may be required. In envisioned OSA scenarios, such tracking may be more difficult, given the dynamics of proposed OSA techniques, as well as the different network management infrastructures. In this work, the random temporal sampling approach for estimation of temporal spectrum occupancy parameters is expanded. A frame-based sampling inversion technique for spectrum occupancy estimation is proposed, which can be used for estimating temporal parameters for a given user/network of interest. In this approach, parameters from randomly sampled frames are used to characterize temporal occupancy. The performance of the proposed approach is examined for temporal occupancy characterization, including circumstances where the monitoring node misses frames due to the sampling design, sensor limitations or through error conditions on a given channel. The work further motivates the use of probabilistic characterization of spectrum occupancy for compliance enforcement, given the non-deterministic behavior of dynamic spectrum access mechanisms in emerging wireless network deployment scenarios.

In this paper opportunities for wireless devices to harvest out-of-band RF energy are explored. Emerging multi-standard wireless communications devices such as mobile handsets currently include multiple antennas and multiple transceiver systems. Typical device usage leverages a subset of available antenna and transceiver systems at any given time. Additionally, receiver chains in use typically filter out-of-band energies to improve communications system performance. These out-of-band energies can be collected in multi-antenna and single-antenna systems or simpler devices such as wireless sensor nodes. In this paper, opportunities for harvesting out-of-band energy are investigated. Investigations use both data collected from an extensive measurement campaign spanning 88–2686MHz across 4 mid-sized US cities as well as simulations to provide insights into the opportunities for RF energy harvesting. An architecture is proposed using a 3-port out-of-band RF harvester to enable opportunistic capture of ambient RF energy. The proposed approach is demonstrated via simulation using a planar inverted-F antenna with feed line coupled to a split ring resonator, for harvesting out-of-band energy. Results indicate that potentially significant proportions of ambient energy remain untapped and can be harvested using the proposed approach. This work provides valuable insight into the design space for harvesting out-of-band RF energy.

Spectrum measurements are a crucial element of compliance enforcement in dynamic spectrum access (DSA) networks and for emerging interference limit policies such as harm claim thresholds. Such measurements should provide better spatio-temporal resolution than traditional compliance enforcement approaches for characterisation of spectrum usage. Using current techniques, even in distributed monitoring networks, monitoring nodes typically sense contiguous spectral bands, including sub-bands which may not be of interest. This can be energy inefficient, particularly for use in denser monitoring networks, such as those envisioned for DSA within the VHF and UHF bands. In this paper a knowledge-based, non-contiguous, multi-resolution sensing approach for characterization of spectrum occupancy is proposed for compliance enforcement networks. The knowledge framework is introduced for scenarios involving heterogeneous radio access technologies (RAT), and non-uniform spectral sampling using the Non-Uniform Fast Fourier Transform (NUFFT) is demonstrated. The performance of the proposed approach is compared to the typical contiguous sensing approach used in practice, using simulations. Results suggest that the proposed approach can decrease the number of computations required, which would improve energy efficiency for individual nodes in monitoring networks.

The estimation of temporal occupancy statistics is a common monitoring output for spectrum management. In emerging dynamic spectrum access (DSA) networks as well as for recent interference limit policies such as harm claim threshold mechanisms, this is even more crucial for compliance enforcement since operational parameters additionally can include temporal constraints. In this paper, random temporal sampling is explored, for probabilistic characterization of temporal channel occupancy. The precision and bias performance due to various random temporal measurement plan designs are examined, both analytically as well as through experiments implemented using Software Defined Radio technology. A framework for performance analysis of random temporal sampling for compliance enforcement is presented, and a lower bound on sensing performance in terms of estimator precision is derived for spectrum occupancy modeled as an alternating renewal process. Using random sampling, estimator variance is seen to approach the theoretical lower bound using larger sample sizes, or through sparser sampling. Results further suggest that the detection error impacts the performance of random temporal sampling for average temporal occupancy estimation. The work further motivates the use of probabilistic characterization of spectrum occupancy for compliance enforcement, given the non-deterministic behavior of dynamic spectrum access mechanisms in emerging wireless network deployment scenarios.

One challenge faced in the engineering education, is the need to imbue both technical competence and critical thinking skills within the bounds of academic program delivery. While technical competence can be built and assessed using structured and/or quantitative exercises, critical thinking is a skill that is both difficult to cultivate and to assess. Critical thinking involves the possession of both an expert mental model as well as the ability to leverage this model in various tasks. In engineering education, these tasks include making and justifying design choices, system optimization, and predicting system performance. Prior work, by one of the authors, explored the role of graphic organizers in the development of student's mental models. This paper describes action research underway, to explore the use of the argument map, as a structured means of leveraging mental models to promote critical thinking. In this paper, interactive small-group tasks based on argument maps are presented, and outputs generated by the initial cohort of undergraduate senior learners on these tasks are examined for evidence of critical thinking. These items form the basis of a longer-term longitudinal study in which the most effective means of deploying argument maps for promoting critical thinking will be examined.

Extended Common Mode LVDS components from SPACE IC have been tested for their radiation hardness. The collected test results are introduced and discussed.

In this paper, we propose a channel selection algorithm for vehicular dynamic spectrum access (VDSA) networks employing instance-based learning methods. Due to the high mobility and spatially variant spectrum availability across large geographical regions associated with this transmission environment, we propose using VDSA methods for non-safety applications such as traffic efficiency and local information dissemination. Additionally, we propose a distance-based multidimensional indexing approach to enable learning of a vehicle communications environment. Our results suggest that the multidimensional approach can improve the channel selection and channel switching performance, especially in either unknown environments or when limited learning information is available due to circumstances such as reduced storage requirements.