Klaus D. McDonald-Maier

University of Essex, Colchester, ENG, United Kingdom

Are you Klaus D. McDonald-Maier?

Claim your profile

Publications (97)15.91 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: ICmetrics is a new approach that exploits the characteristic and behaviour of an embedded system to obtain a collection of properties and features, which aims to uniquely identify and secure an embedded system based on its own behavioural identity. In this paper, an algorithm based on a self-organising map (SOM) neural network is proposed to extract and analyse the features derived from a processor's performance profile (i.e. average cycles per instruction (CPI)), where the extracted features are used to help finding the main behaviours of the system. The proposed algorithm has been tested with different programs selected from the MiBench benchmark suite, and the results achieved show that it can successfully segment each program into different main phases based on the unique extracted features, which confirms its utility for the ICmetrics technology.
    Emerging Security Technologies (EST), 2013 Fourth International Conference on; 01/2013
  • R. Tahir, Huosheng Hu, Dongbing Gu, K. McDonald-Maier, G. Howells
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a scheme for the generation of strong session based ICMetrics key pairs for security critical embedded system applications. ICMetrics generates the security attributes of the sensor node based on measurable hardware and software characteristics of the integrated circuit. In the proposed scheme a random session ID is assigned by a trusted party to each participating network entity, which remains valid for a communication session. Our work is based on the design of a key derivation function that uses an ICMetrics secret key and a session token assigned by the trusted party to derive strong cryptographic key pairs for each entity. These session tokens also serve the purpose of identification/authentication between the trusted parties and the respective nodes in each network. The main strength of our proposed scheme rests on the randomness feature incorporated via the random session ID's, which makes the generated strong private/public key pair highly resistant against exhaustive search and rainbow table attacks. Our proposed approach makes use of key stretching using random session tokens via SHA-2 and performs multiple iterations of the proposed key derivation function to generate strong high entropy session key pairs of sufficient length. The randomness of the assigned ID's and the session based communication hinders the attacker's ability to launch various sorts of cryptanalytic attacks, thereby making the generated keys very high in entropy. The randomness feature has also been very carefully tuned according to the construction principles of ICMetrics, so that it doesn't affect the original ICMetrics data. The second part of the proposed scheme generates a corresponding public session key by computing the Hermite Normal Form of the high entropy private session key.
    Advanced Information Networking and Applications Workshops (WAINA), 2013 27th International Conference on; 01/2013
  • Source
    Shoaib Ehsan, Adrian F Clark, Klaus D McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: A vision system that can assess its own performance and take appropriate actions online to maximize its effectiveness would be a step towards achieving the long-cherished goal of imitating humans. This paper proposes a method for performing an online performance analysis of local feature detectors, the primary stage of many practical vision systems. It advocates the spatial distribution of local image features as a good performance indicator and presents a metric that can be calculated rapidly, concurs with human visual assessments and is complementary to existing offline measures such as repeatability. The metric is shown to provide a measure of complementarity for combinations of detectors, correctly reflecting the underlying principles of individual detectors. Qualitative results on well-established datasets for several state-of-the-art detectors are presented based on the proposed measure. Using a hypothesis testing approach and a newly-acquired, larger image database, statistically-significant performance differences are identified. Different detector pairs and triplets are examined quantitatively and the results provide a useful guideline for combining detectors in applications that require a reasonable spatial distribution of image features. A principled framework for combining feature detectors in these applications is also presented. Timing results reveal the potential of the metric for online applications.
    Sensors 01/2013; 13(8):10876-10907. · 2.05 Impact Factor
  • R. Tahir, Huosheng Hu, Dongbing Gu, K. McDonald-Maier, G. Howells
    [Show abstract] [Hide abstract]
    ABSTRACT: Cryptography has become an essential for providing security in embedded system applications. The employed cryptographic primitives should provide strong protection such that the security of the system is not compromised at any point in the lifecycle of a secure operation. This particularly includes the secure generation and maintenance of cryptographic keys. In general this assumption is difficult to accomplish, since there are attacks that come under this umbrella ranging from brute force attacks on the key to capturing the node to extract the key. In this paper we investigate and analyze ICMetrics and its counterpart scheme referred to as the scheme for the generation of strong high entropy ICMetrics session key pairs. ICMetrics is a key technology that computes the secret key based on hardware/ software properties of a device, thereby providing resilience against node capture attacks, while high entropy key pair generation scheme is employed to strengthen the generated ICMetrics basis number, so as to safeguard the generated strong key pairs from brute force and rainbow table attacks.
    Communications, Signal Processing, and their Applications (ICCSPA), 2013 1st International Conference on; 01/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Integrated Circuit Metrics (ICmetrics) technology is concerned with the extraction of measurable features of an embedded system, capable of uniquely identifying the system's behaviour. Any changes in these identifiers (profiles) during consequent devices' operation would signal about a possible safety or security breach within the electronic system. This paper explores the combination of program counter (PC) and Cycles per Instructions (CPI) of a processor core as a potential ICmetrics source for embedded system security. The use of this combination exhibits that while isolated PC values may not always generate a stable identifier (profile) for a device that would distinguish the device from the rest in a considered set, the PC and CPI sequences and frequencies in the execution flow may serve as suitable ICmetrics features.
    Emerging Security Technologies (EST), 2013 Fourth International Conference on; 01/2013
  • Sen Wang, Ling Chen, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Traditionally, it is difficult for elderly and disabled people to control a wheelchair passing a narrow doorway manually. This paper presents a dynamic trajectory planning algorithm for wheelchairs to pass a door smoothly and automatically. It is a sensor-based approach in which two laser rangefinders are deployed in the wheelchair for real-time door detection. To generate smooth trajectories that enable a wheelchair to pass a door perpendicularly, Bézier curve based trajectories are calculated repeatedly during the whole course of door passing. The proposed approach is tested on a real wheelchair and the experimental results are presented to show the good performance and effectiveness of our proposed automatic door passing strategy.
    Computer Science and Electronic Engineering Conference (CEEC), 2013 5th; 01/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper explores means to increase efficiency in performing tasks with multi-robot teams, in the context of natural Human-Multi-Robot Interfaces (HMRI) for command and control. The motivating scenario is an emergency evacuation by a transport convoy of unmanned ground vehicles (UGVs) that have to traverse, in shortest time, an unknown terrain. In the experiments the operator commands, in minimal time, a group of rovers through a maze. The efficiency of performing such tasks depends on both, the levels of robots' autonomy, and the ability of the operator to command and control the team. The paper extends the classic framework of levels of autonomy (LOA), to levels/hierarchy of autonomy characteristic of Groups (G-LOA), and uses it to determine new strategies for control. An UGVoriented command language (UGVL) is defined, and a mapping is performed from the human-friendly gesture-based HMRI into the UGVL. The UGVL is used to control a team of 3 robots, exploring the efficiency of different G-LOA; specifically, by (a) controlling each robot individually through the maze, (b) controlling a leader and cloning its controls to followers, and (c) controlling the entire group. Not surprisingly, commands at increased G-LOA lead to a faster traverse, yet a number of aspects are worth discussing in this context.
    Collaboration Technologies and Systems (CTS), 2013 International Conference on; 01/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Integrated Circuit Metrics is mainly concerned with the extraction of measurable properties or features of a given hardware device, capable of uniquely identifying the system's behaviour. This paper presents features that can be extracted from software executing on a device and identify the very software in execution. The main contribution of this paper is in two folds. The ability to extract features whiles the software is in execution as well as correctly identifying the software to divulge any tampering or malicious exploitation. Our aim is to use program counter values generated during program execution to train a k-means algorithm optimized for vector quantization, and later use the system to associate program counter values with an application.
    Emerging Security Technologies (EST), 2013 Fourth International Conference on; 01/2013
  • A. Stoica, D.F. Barrero, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper reports a first step toward a brain-computer interface (BCI) for collaborative targeting. Specifically, we explore, from a broad perspective, how the collaboration of a group of people can increase the performance on a simple target identification task. To this end, we requested a group of people to identify the location and color of a sequence of targets appearing on the screen, and measured the time and the accuracy of the response. The individual results are compared to a collective identification result determined by simple majority voting, with random choice in case of drawn. The results are promising, as the identification becomes significantly more reliable even with this simple voting, and with a small number of people (either odd or even) involved in the decision. In addition, the paper briefly analyzes the role of brain-computer interfaces in collaborative targeting, extending the targeting task by using a BCI instead of a mechanical response.
    Collaboration Technologies and Systems (CTS), 2013 International Conference on; 01/2013
  • Source
    Ling Chen, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Localization plays a significant role in the autonomous navigation of a mobile robot. This paper investigates mobile robot localization based on Extended Kalman Filter(EKF) algorithm and a feature based map. Corner angles in the environment are detected as the features, and the detailed processes of feature extraction are described. Then the motion model and odometry information are elaborated, and the EKF localization algorithm is presented. Finally, the experimental result is given to verify the feasibility and performance of the proposed localization algorithm.
    Emerging Security Technologies (EST), 2012 Third International Conference on; 01/2012
  • Y. Kovalchuk, W.G.J. Howells, H. Hu, D. Gu, K.D. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a novel technique, termed ICmetrics (Integrated Circuit metrics), that can be used for the purposes of generating encryption keys, electronic signatures, detecting attempts of frauds, or preventing malfunction of hardware components and systems. The ICmetrics technology is based on employment of measurable features derived from characteristics of a given electronic device in order to generate an identifier that uniquely determines or describes the device. Any changes in the identifier during consequent device's operation would signal about a possible safety or security breach within the electronic system. After a detailed overview of the ICmetrics technology and comparing it to the alternative techniques commonly used for securing electronic systems, the paper discusses challenges of developing the technology and brings an example to demonstrate how these issues are being addressed.
    System Safety, incorporating the Cyber Security Conference 2012, 7th IET International Conference on; 01/2012
  • Source
    E.J. Rechy-Ramirez, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a user-friendly human machine interface (HMI) for hands-free control of an electric powered wheelchair (EPW). Its two operation modes are based on head movements: Mode 1 uses only one head movement to give the commands, and Mode 2 employs four head movements. An EEG device, namely Emotiv EPOC, has been deployed in this HMI to obtain the head movement information of users. The proposed HMI is compared with the joystick control of an EPW in an indoor environment. The experimental results show that Control Mode 2 can be implemented at a fast speed reliably, achieving a mean time of 67.90 seconds for the two subjects. However, Control Mode 1 has inferior performance, achieving a mean time of 153.20 seconds for the two subjects although it needs only one head movement. It is clear that the proposed HMI can be effectively used to replace the traditional joystick control for disabled and elderly people.
    Robotics and Biomimetics (ROBIO), 2012 IEEE International Conference on; 01/2012
  • R. Tahir, Huosheng Hu, Dongbing Gu, K. McDonald-Maier, G. Howells
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a scheme for the generation of strong high entropy keys based on ICMetrics. ICMetrics generates the security attributes of the sensor node based on measurable hardware and software characteristics of the integrated circuit. This work is based on key derivation functions to derive cryptographic key pairs from ICMetrics values. The proposed ICMetrics based key derivation function makes use of ICMetrics basis numbers and authentication tokens from the trusted third party to generate high entropy public/private key pairs. The proposed approach makes use of key stretching using SHA-2 and performs multiple iterations of the proposed key derivation function to generate strong high entropy keys of sufficient length, so as to prevent exhaustive search attacks. The novelty of this work lies in the fact that the entire key generation scheme has been designed keeping in mind the construction principles of ICMetrics, which does not store keys but computes these for every session based on ICMetrics value, therefore use of a random value anywhere in the protocol will compromise the purpose of ICMetrics. The proposed scheme generates high entropy key pairs while concealing the original ICMetrics data, such that it is impossible to recover the ICMetrics basis data in the system.
    Internet Technology And Secured Transactions, 2012 International Conference for; 01/2012
  • Sen Wang, Ling Chen, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Door passing is the basic capability of an intelligent wheelchair. This paper presents a novel strategy to address the door passing issue by dynamically generating the Bezier curve based trajectory. It consists of door finding, optimization based trajectory generation and tracking control, which are executed repeatedly to increase the ability of passing the door and improve the performance. Whenever the door is detected, the optimization method produces a new smooth reference trajectory in real time for the wheelchair to follow. The proposed approach is tested in reality to verify its feasibility and efficiency, and the experimental results show its good performance in terms of the accuracy of finding the door and passing the doorway.
    Robotics and Biomimetics (ROBIO), 2012 IEEE International Conference on; 01/2012
  • Source
    Ling Chen, Sen Wang, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Door passing is the basic capability of an intelligent wheelchair. This paper presents a novel approach to address the door passing issue using Bézier curve based trajectory planning. The planed path consists of two segments: one from corridor to door and the other from door to the goal position. For each segment, an optimal Bézier curve is generated as a reference trajectory for an intelligent wheelchair to travel smoothly and accurately subject to corridor constraints, curvature limitation and obstacles. The simulation is conducted to verify the feasibility of the proposed approach, and the results show a good performance in terms of tracking accuracy and good maneuverability.
    Control (CONTROL), 2012 UKACC International Conference on; 01/2012
  • Source
    Bowen Lu, Dongbing Gu, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Gaussian process (GP) is well researched and used in machine learning field. Comparing with artificial neural network (ANN) and support vector regression (SVR), it provides additional covariance information for regression results. By exploiting this feature, an uncertainty based locational optimisation strategy combining with an entropy based data selection method for mobile sensor networks is presented in this paper. Centroidal Voronoi tessellation (CVT) is used as a locational optimisation framework and Informative Vector Machine (IVM) is applied for data selection. Simulations with different locational optimisation criteria are conducted and the results are given, which proved the effectiveness of presented strategy.
    Emerging Security Technologies (EST), 2012 Third International Conference on; 01/2012
  • Source
    Sen Wang, Huosheng Hu, K. McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Localization is critical for various applications of Wireless Sensor Networks. This paper presents a 3D localization algorithm for high accuracy localization of a wireless sensor network, which consists of three parts: pre-localization, refinement and sequence search. It is clear that the localization accuracy partly depends on the localization sequence of the unknown nodes, which has not been investigated so far. The proposed novel algorithm aims to address this problem by searching a localization sequence corresponding to a high localization accuracy and a robust algorithm. The simulation results show that the proposed algorithm can get rid of the flip ambiguity and is more robust than several existing algorithms in terms of the localization accuracy.
    Emerging Security Technologies (EST), 2012 Third International Conference on; 01/2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The ICmetrics technology is based on extracting features from digital devices' operation that may be integrated together to generate unique identifiers for each of the devices or create unique profiles that describe the devices' actual behaviour. Any changes in these identifiers (profiles) during consequent devices' operation would signal about a possible safety or security breach within the electronic system. This paper explores the program counter (PC) of a processor core as a potential source for ICmetrics features and discusses several methods of feature values acquisition with the aim to achieve a maximum level of information gain with a minimal impact on a system's performance. The main finding of this study is that while isolated PC values may not always allow to generate a stable identifier (profile) for a device that would distinguish the device from the rest in the considered set, the PC sequences and frequencies in the execution flow may serve as suitable ICmetrics features, which has yet to be tested in complex scenarios.
    Emerging Security Technologies (EST), 2012 Third International Conference on; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: In embedded vision systems, parallel computation of the integral image presents several design challenges in terms of hardware resources, speed and power consumption. Although recursive equations significantly reduce the number of operations for computing the integral image, the required internal memory becomes prohibitively large for an embedded integral image computation engine for increasing image sizes. With the objective of achieving high-throughput with minimum hardware resources, this paper proposes a memory-efficient design strategy for a parallel embedded integral image computation engine. Results show that the design achieves nearly 35% reduction in memory for common HD video.
    Proceedings of the 2011 Irish Machine Vision and Image Processing Conference; 09/2011
  • Shoaib Ehsan, Nadia Kanwal, Adrian F Clark, Klaus D McDonald-Maier
    [Show abstract] [Hide abstract]
    ABSTRACT: Speeded-Up Robust Features is a feature extraction algorithm designed for real-time execution, although this is rarely achievable on low-power hardware such as that in mobile robots. One way to reduce the computation is to discard some of the scale-space octaves, and previous research has simply discarded the higher octaves. This paper shows that this approach is not always the most sensible and presents an algorithm for choosing which octaves to discard based on the properties of the imagery. Results obtained with this best octaves algorithm show that it is able to achieve a significant reduction in computation without compromising matching performance.
    IEEE Transactions on Image Processing 06/2011; 21(1):297-304. · 3.20 Impact Factor

Publication Stats

367 Citations
15.91 Total Impact Points

Institutions

  • 2006–2013
    • University of Essex
      • School of Computer Science and Electronic Engineering
      Colchester, ENG, United Kingdom
  • 2004–2009
    • University of Kent
      Cantorbery, England, United Kingdom
  • 2008
    • CTTC Catalan Telecommunications Technology Centre
      Barcino, Catalonia, Spain