Mike Dixon

The constant increase in users’ bandwidth needs, through a large variety of multimedia applications, creates the need for highly effective network traffic control. This need is imperative in wireless networks, where the available bandwidth is limited, but is very important for wired networks as well. In this work we focus on the problem of policing video traffic from sources encoded with H.264 and H.265, given that these are the major state-of-the-art standards currently in the market. Building on work that has shown that classic traffic policing schemes can lead to unnecessarily strict policing for conforming video sources, we propose the use of Artificial Intelligence (AI) – based traffic policing schemes for video traffic. We conduct a performance evaluation of several AI – based schemes with the classic token bucket and we show that our proposed Frame Size Predictor and Policer scheme improve the performance of the classic token bucket by around 90% for conforming users, while providing only slightly worse policing results for non-conforming users.

As video has become the dominant type of traffic over wired and wireless networks, the efficient transmission of video streams is of paramount importance. Hence, especially for wireless networks, the optimum utilization of the available bandwidth while preserving the users’ Quality of Service and Quality of Experience requirements is crucial. Towards this goal, the accurate prediction of upcoming video frame sizes can play a significant role. This work focuses on achieving such an accurate prediction for videos encoded with H.264 and H.265, which are the major state-of-the-art standards based on their current market share. Unlike previous studies, we use single-step and multi-step approaches to capture the long-range dependence and short-range dependence properties of variable bit rate video traces through neural networks-based modeling. We evaluate the accuracy of Long Short Term Memory, Convolutional Neural Networks and Sequence-to-Sequence models and compare them with existing approaches. Our models show significantly higher accuracy for a variety of videos. We also provide a case study on how our model can be used for traffic policing purposes.

Global communications without reliance on an engineered communications network make the ionosphere an attractive medium for wireless sensors in remote deployments. However, ionospheric circuits’ temporary availability is a challenge in scheduling transmissions for a sensor with limited power, communications and computational capacity, particularly where cost and antenna constraints limit operation to a single frequency. We describe a technique for scheduling transmissions based on precomputed propagation models. The models predict the time-varying Signal to Noise Ratio (SNR) at the receiver. We describe methods to determine threshold SNR values, using the Weak Signal Propagation Reporter (WSPR) database to determine if a time slot is suitable for transmission.Two techniques are investigated to quantify the failed receptions: the Inverse Square Law method uses a statistical approach and a sampling measurement technique called Goldilocks. The two approaches yielded threshold SNR values of −21 dB and −19 dB, respectively, for a time slot with a 90% successful reception goal. Applying these thresholds to the modelled SNR, we generate a precomputed hourly transmission schedule. With the schedule determined monthly, a 12-month plan requires 36 bytes of wireless sensor storage. A six-day experiment, using a 1677 km path, found that the schedule resulted in an 83% reception rate when used with a power level of 200 mW.

Analysis of time series data has been a challenging research subject for decades. Email traffic has recently been modelled as a time series function using a Recurrent Neural Network (RNN) and RNNs were shown to provide higher prediction accuracy than previous probabilistic models from the literature. Given the exponential rise of email workloads which need to be handled by email servers, in this paper we first present and discuss the literature on modelling email traffic. We then explain the advantages and limitations of different approaches as well as their points of agreement and disagreement. Finally, we present a comprehensive comparison between the performance of RNN and Long Short Term Memory (LSTM) models. Our experimental results demonstrate that both approaches can achieve high accuracy over four large datasets acquired from different universities’ servers, outperforming existing work, and show that the use of LSTM and RNN is very promising for modelling email traffic.

Passwords are the most widely used method of authentication on the Internet, but users find compliance with password guidelines difficult, and we know little about the long-term effects of attempts to improve compliance. In this paper, we extend the work of fear appeals use in the IS security domain to investigate their longer-term effects. We conducted a longitudinal experimental study to examine fear appeals’ long- and short-term effects. Using a model based on protection motivation theory (Rogers, 1983), we found that fear of threat, perceived password effectiveness, and password self-efficacy predicted compliance. We also found that neither perceived vulnerability to a security attack nor perceived severity of an attack influenced compliance. Providing persuasive communication improved compliance with password guidelines and resulted in significantly stronger passwords, but the effects on compliance intentions were only short term. This study extends our understanding of the factors that influence compliance with password guidelines and how we can modify them to improve compliance. We raise interesting questions about the role of fear in different IS security contexts. We also highlight the need for more research on the long-term impact of persuasive communication.

Information technology has been identified as a vital means for supporting construction project processes, yet the level of adoption in the construction industry has been low relative to other sectors. Mobile Information and communications technology (mICT) allows people to access information from wherever they are, and as work in the construction industry is mainly fieldwork, with workers being highly mobile, mICT holds promise for the sector, particularly in developing countries. The aim of the study reported in this paper was to investigate factors that could impact stakeholders’ adoption of mICT in the Libyan construction industry. A model of mICT adoption was developed, and tested using data collected from a survey of 202 construction industry stakeholders from 15 companies in Libya. The analysis was undertaken using structural equation modelling. It was found that perceived usefulness and ease of use are important in determining intention to adopt mICT, and that they are influenced by self - efficacy and facilitating conditions. The cost of technology was not found to be a barrier to adoption. Recommendations are made to the construction industry in Libya and relevant government authorities, in order to help improve awareness of the potential of mICT and to help improve potential users’ self-efficacy.

A large volume of research has been undertaken to improve short range mesh based sensor networks. This paper considers a contrasting, but less established, research position that long range single hop HF transmissions suit many applications. HF transmissions are, however, not without challenges and this paper details an experiment to evaluate the viability of using spatial diversity to mitigate the intermittent and cyclical availability of HF circuits. Four transmitting nodes were placed across Australia and six days of packet transmissions at 14 MHz were analysed. The findings support the use of spatial diversity to maximise HF circuit availability and the presence of redundant circuits suggests the system has the capacity to accommodate antenna radiation patterns that present nulls in return for targeted gain.

Studies on the composition and nature of Internet protocols are crucial for continued research and innovation. This study used three different methods to investigate the presence and level of support for various Internet protocols. Internet traffic entering and exiting a university network was passively captured, anonymised and analysed to test protocol usage. Active tests probed the Internet's most popular websites and experiments on the default behaviour of popular client, server and mobile operating systems were performed to reconcile the findings of the passive data collection. These results are valuable to research areas, such as those using emulations and simulations, where realism is dependent on the accuracy of the underlying assumptions about Internet traffic. Prior work is leveraged to explore changes and protocol adoption trends. This study shows that the majority of Internet traffic is now encrypted. There has also been an increase in large UDP frames, which we attribute to the Google QUIC protocol. Support for TCP options such as Selective Acknowledgements (SACK) and Maximum Segment Size (MSS) can now be assumed. Explicit Congestion Notification (ECN) usage is still marginal, yet active measurement shows that many servers will support the protocol if requested. Recent IETF standards such as Multipath TCP and TCP Fast Open have small but measurable levels of adoption.

Over half of the transmission time in WiFi networks is dedicated to ensuring that errors are corrected or detected. Despite these mechanisms, many studies have concluded that frame error rates vary. An increased understanding of why frames are lost is a pragmatic approach to improving real world 802.11 throughput. The potential beneficiaries of this research, include rate control algorithms, Modulation and Coding Schemes, simulation models, frame size selection and 802.11 configuration guidelines. This paper presents a measurement study of the factors which correlate with packet loss in 802.11 WiFi. Both passive and active approaches were used to investigate how the frame size, modulation and coding scheme and airtime effect the loss rate. Overall, packet errors were high, but the size of frames were not a major determinant of the loss rate. The loss rate decreased with the airtime but at substantially lower rates than those suggested in simple packet error models. Future work will further try to isolate and investigate specific errors, such as head on collisions in the preamble.

Passwords have long been the preferred method of user authentication, yet poor password practices cause security issues. The study described in this paper investigates how user perceptions of passwords and security threats affect intended compliance with guidelines and explores how these perceptions might be altered in order to improve compliance. It tests a research model based on protection motivation theory [24]. Two groups of internet users were surveyed, one of which received password security information and an exercise to reinforce it. This study suggests effective ways that trainers or employers can improve compliance with password guidelines. In particular, training programs should aim to enhance IS security coping appraisal. The research model proposed in this study has also been shown to be a useful model for explaining IS security behavioral intentions.