Duncan Hodges

An attenuation time series model for propagation forecasting

A key problem in the efficient use of higher (Ka- and V-band) frequencies lies in the mitigation of propagation impairments caused by meteorological phenomena. The traditional approach to this problem is based upon a relatively simplistic statistical model in the form of a fade margin. At higher frequencies this traditional approach becomes inefficient due to the large margin required. This inefficiency has lead to the introduction of dynamic fade mitigation techniques (FMTs). We present a method of generating attenuation time series that can be used for the development and evaluation of FMTs. The method we propose is based on the use of proven numerical weather prediction models in conjunction with a propagation model. This approach has two unique aspects. First, the spatial correlation and dynamic behavior of the attenuation fields are inherited from the meteorological environment. Second, the model can provide forecasts of attenuation. It is foreseen that this a priori knowledge of the occurrence of fades, their likely depth and likely duration can be exploited to manage the resource control of entire networks. This paper presents a description of the method and demonstrates the ability to generate attenuation time series. Conclusions are drawn regarding its use in real-time for network resource management.

A Model for Identity in the Cyber and Natural Universes

In this paper we describe a model for evaluating and investigating holistic identities across both the cyber- and natural-worlds. The model is taxonomy-agnostic and helps guide analysts through the creation of a rich super identity whilst also being flexible enough for users to deploy their intuition and experience. Super identities are created by iterative stages of enrichment and relaxation, which allow the super identity to grow and include more elements of identity. The confidence in individual elements can be propagated so as to provide an overall level of confidence in the super identity. The quality of the final, rich super identity can be metricized, identifying good qualities of the super identity such as reinforcement of elements and consensus within element types. This work is a key enabling component of a large inter-disciplinary project called Super identity, which is seeking to develop the model through understanding the elements of identity which might be available and how their reliability might vary according to context. In addition to this aim the project looks to develop visual analytic support to help validate the modelling approach. In particular demonstrating the models use by analysts seeking to make identity attributions.

Tools for understanding identity

We present two tools for analysing identity in support of homeland security. Both are based upon the Superi-dentity model that brings together cyber and physical spaces into a single understanding of identity. Between them, the tools provide support for defensive, information gathering and capability planning operations. The first tool allows an analyst to explore and understand the model, and to apply it to risk-exposure assessment activities for a particular individual, e.g. an influential person in the intelligence or government community, or a commercial company board member. It can also be used to understand critical capabilities in an organizations identity-attribution process, and so used to plan resource investment. The second tool, referred to as Identity Map, is designed to support investigations requiring enrichment of identities and the making of attributions. Both are currently working prototypes.

What does your profile really say about you?: privacy warning systems and self-disclosure in online social network spaces

This paper reports current progress on the design and initial evaluation of an innovative privacy feedback system aimed to provide social network users with tailor-made feedback about their identity exposure online. Preliminary results suggest our feedback system, based on a research driven model of identity, appears to reduce the amount of information individuals disclose about themselves in social network spaces. We discuss the impact of our feedback system on the way individuals share information online, as well as suggestions for a more fine-grained evaluation and future development of this feedback system.

Individual Differences in Cyber Security Behaviors: An Examination of Who Is Sharing Passwords

In spite of the number of public advice campaigns, researchers have found that individuals still engage in risky password practices. There is a dearth of research available on individual differences in cyber security behaviors. This study focused on the risky practice of sharing passwords. As predicted, we found that individuals who scored high on a lack of perseverance were more likely to share passwords. Contrary to our hypotheses, we found younger [corrected] people and individuals who score high on self-monitoring were more likely to share passwords. We speculate on the reasons behind these findings, and examine how they might be considered in future cyber security educational campaigns.

Relationships between password choices, perceptions of risk and security expertise

‘Despite technological advances, humans remain the weakest link in Internet security’ [1], this weakness is typically characterised in one of two domains. First, systems may not enable humans to interface securely, or the security mechanisms themselves are unusable or difficult to use effectively. Second, there may be something fundamental about the behaviour of some people which leads them to become vulnerable.

Estimation of rainfall rate from terrestrial microwave link measurements

The large number of millimetre and microwave terrestrial links that are in operation has lead to the belief that it may be possible to estimate two-dimensional rainfall rate fields. This paper outlines a reconstruction algorithm that can be used to estimate the rainfall field from a number of measurements of path attenuation on terrestrial links. In this paper we adopt a different approach to the regular-ization of this inverse problem. The main focus of the paper is a description of the retrieval algorithm which is demonstrated through simulation. The approach has been shown to be robust and relatively insensitive to errors and quantisation in the link measurements.

Generation of attenuation time-series for EHF SATCOM simulation

In this paper, a method for deriving time-series of attenuation on fixed satellite or terrestrial links is described. The method uses meteorological model forecast data and radar data and hence permits time coincident derivations for multiple sites that include the spatial correlation properties inherent in weather systems. The success of the technique depends upon insertion of the short interval temporal properties (varying typically over 1 second to 15 minutes) that are statistically independent between stations. The technique has immediate application to simulation of the performance of multi-media satellite to earth links and has longer-term significance in terms of operational real-time performance prediction.

Breaking the Arc: Risk control for Big Data

The use of Big Data technologies and analytics have the potential to revolutionise the world. The mass instrumentation of the planet and society is providing intelligence that is not only enhancing our personal lives, but also opening up new opportunities for addressing some of key environmental, social and economic challenges of the 21st century. Unfortunately, as with all technology, there is the potential for misuse; in the case of personal data the ability to gather, enrich and mine at extreme pace and volume could result in societal-scale privacy intrusions. We apply a model for identity across cyber and physical spaces to the question of risk control for personal-data in the context of big data analytics. Using a graphical model for identity we reflect on the response options we have and how such risk controls may or may not be effective.

Building a better intelligence machine: A new approach to capability review and development

In this paper we propose a new approach for managing capability within organisations that are engaged in identity-attribution or identity-enrichment exercises. Specifically, we believe that a modelling framework which uses a bottom-up data-driven approach encapsulates the most appropriate abstraction of capability. In particular, that this should be agnostic (but aware) of the type of capability provider (whether a technology, service or a human) and that the capability framework be independent of specific staff or skill groups. In this way our approach could avoid the limitations of solely driving strategic capability enhancements through siloed role or skills-based staff pools, which are necessarily biased towards the staff management frameworks and potentially limit the strategic outcome.