| 1. |
- Blanch, Krister, 1991
(författare)
-
Beyond-application datasets and automated fair benchmarking
- 2023
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Beyond-application perception datasets are generalised datasets that emphasise the fundamental components of good machine perception data. When analysing the history of perception datatsets, notable trends suggest that design of the dataset typically aligns with an application goal. Instead of focusing on a specific application, beyond-application datasets instead look at capturing high-quality, high-volume data from a highly kinematic environment, for the purpose of aiding algorithm development and testing in general. Algorithm benchmarking is a cornerstone of autonomous systems development, and allows developers to demonstrate their results in a comparative manner. However, most benchmarking systems allow developers to use their own hardware or select favourable data. There is also little focus on run time performance and consistency, with benchmarking systems instead showcasing algorithm accuracy. By combining both beyond-application dataset generation and methods for fair benchmarking, there is also the dilemma of how to provide the dataset to developers for this benchmarking, as the result of a high-volume, high-quality dataset generation is a significant increase in dataset size when compared to traditional perception datasets. This thesis presents the first results of attempting the creation of such a dataset. The dataset was built using a maritime platform, selected due to the highly dynamic environment presented on water. The design and initial testing of this platform is detailed, as well as as methods of sensor validation. Continuing, the thesis then presents a method of fair benchmarking, by utilising remote containerisation in a way that allows developers to present their software to the dataset, instead of having to first locally store a copy. To test this dataset and automatic online benchmarking, a number of reference algorithms were required for initial results. Three algorithms were built, using the data from three different sensors captured on the maritime platform. Each algorithm calculates vessel odometry, and the automatic benchmarking system was utilised to show the accuracy and run-time performance of these algorithms. It was found that the containerised approach alleviated data management concerns, prevented inflated accuracy results, and demonstrated precisely how computationally intensive each algorithm was.
|
|
| 2. |
- Fan, Yuchuan, et al.
(författare)
-
Feedforward Neural Network-Based EVM Estimation : Impairment Tolerance in Coherent Optical Systems
- 2022
-
Ingår i: IEEE Journal of Selected Topics in Quantum Electronics. - : Institute of Electrical and Electronics Engineers Inc.. - 1077-260X .- 1558-4542. ; 28:4
-
Tidskriftsartikel (refereegranskat)abstract
- Error vector magnitude (EVM) is commonly used for evaluating the quality of m-ary quadrature amplitude modulation (mQAM) signals. Recently proposed deep learning techniques for EVM estimation extend the functionality of conventional optical performance monitoring (OPM). In this article, we evaluate the tolerance of our developed EVM estimation scheme against various impairments in coherent optical systems. In particular, we analyze the signal quality monitoring capabilities in the presence of residual in-phase/quadrature (IQ) imbalance, fiber nonlinearity, and laser phase noise. We use feedforward neural networks (FFNNs) to extract the EVM information from amplitude histograms of 100 symbols per IQ cluster signal sequence captured before carrier phase recovery. We perform simulations of the considered impairments, along with an experimental investigation of the impact of laser phase noise. To investigate the tolerance of the EVM estimation scheme to each impairment type, we compare the accuracy for three training methods: 1) training without impairment, 2) training one model for all impairments, and 3) training an independent model for each impairment. Results indicate a good generalization of the proposed EVM estimation scheme, thus providing a valuable reference for developing next-generation intelligent OPM systems.
|
|
| 3. |
- Shafiq, ur Réhman, 1978-, et al.
(författare)
-
Using Vibrotactile Language for Multimodal Human Animals Communication and Interaction
- 2014
-
Ingår i: Proceedings of the 2014 Workshops on Advances in Computer Entertainment Conference, ACE '14. - New York, NY, USA : Association for Computing Machinery (ACM). - 9781450333146 ; , s. 1:1-1:5
-
Konferensbidrag (refereegranskat)abstract
- In this work we aim to facilitate computer mediated multimodal communication and interaction between human and animal based on vibrotactile stimuli. To study and influence the behavior of animals, usually researchers use 2D/3D visual stimuli. However we use vibrotactile pattern based language which provides the opportunity to communicate and interact with animals. We have performed experiment with a vibrotactile based human-animal multimodal communication system to study the effectiveness of vibratory stimuli applied to the animal skin along with audio and visual stimuli. The preliminary results are encouraging and indicate that low-resolution tactual displays are effective in transmitting information.
|
|
| 4. |
- Zhang, Kewei, et al.
(författare)
-
Protecting GNSS Open Service-Navigation Message Authentication against Distance-Decreasing Attacks
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- As the security of global navigation satellite systems (GNSS) for civilian usage is increasingly important, navigation message authentication (NMA) significantly improves resilience to spoofing attacks. However, not all attacks can be effectively countered: a strong variant of replay/relay attacks, distance-decreasing (DD) attacks, can shorten pseudorange measurements, without manipulating the cryptographically protected navigation message, thus manipulating the position, velocity, and time solution undetected. First, we discuss how DD attacks can tamper with GNSS signals, demonstrating the attack effectiveness on a recorded Galileo signal. DD attacks might introduce bit errors to the forged signals, but the adversary can keep this error rate very low with proper attack parameter settings. Then, based on our mathematical model of the prompt correlator output of the tracking phase at the victim receiver, we find that the correlator output distribution changes in the presence of DD attacks. This leads us to apply hypothesis testing to detect DD attacks, notably a Goodness of Fit (GoF) test and a generalized likelihood ratio test (GLRT), depending on the victim’s knowledge on the DD attacks. Monte Carlo simulations are used to evaluate the detection probability and the receiver operating characteristic (ROC) curves for two tests, for different adversary configuration and noise settings. Then, we evaluate the effectiveness of the GoF and GLRT tests with a synthesized DD signal. Both tests can detect DD attacks with similar performance in high signal-to-noise ratio (SNR) environments. The GLRT detection probability is approximately 20% higher than that of the GoF test in low SNR environments.
|
|
| 5. |
- Nopchinda, Dhecha, 1991
(författare)
-
mm-Wave Data Transmission and Measurement Techniques: A Holistic Approach
- 2019
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- The ever-increasing demand on data services places unprecedented technical requirements on networks capacity. With wireless systems having significant roles in broadband delivery, innovative approaches to their development are imperative. By leveraging new spectral resources available at millimeter-wave (mm-wave) frequencies, future systems can utilize new signal structures and new system architectures in order to achieve long-term sustainable solutions. This thesis proposes the holistic development of efficient and cost-effective techniques and systems which make high-speed data transmission at mm-wave feasible. In this paradigm, system designs, signal processing, and measurement techniques work toward a single goal; to achieve satisfactory system level key performance indicators (KPIs). Two intimately-related objectives are simultaneously addressed: the realization of efficient mm-wave data transmission and the development of measurement techniques to enable and assist the design and evaluation of mm-wave circuits. The standard approach to increase spectral efficiency is to increase the modulation order at the cost of higher transmission power. To improve upon this, a signal structure called spectrally efficient frequency division multiplexing (SEFDM) is utilized. SEFDM adds an additional dimension of continuously tunable spectral efficiency enhancement. Two new variants of SEFDM are implemented and experimentally demonstrated, where both variants are shown to outperform standard signals. A low-cost low-complexity mm-wave transmitter architecture is proposed and experimentally demonstrated. A simple phase retarder predistorter and a frequency multiplier are utilized to successfully generate spectrally efficient mm-wave signals while simultaneously mitigating various issues found in conventional mm-wave systems. A measurement technique to characterize circuits and components under antenna array mutual coupling effects is proposed and demonstrated. With minimal setup requirement, the technique effectively and conveniently maps prescribed transmission scenarios to the measurement environment and offers evaluations of the components in terms of relevant KPIs in addition to conventional metrics. Finally, a technique to estimate transmission and reflection coefficients is proposed and demonstrated. In one variant, the technique enables the coefficients to be estimated using wideband modulated signals, suitable for implementation in measurements performed under real usage scenarios. In another variant, the technique enhances the precision of noisy S-parameter measurements, suitable for characterizations of wideband mm-wave components.
|
|
| 6. |
- Mahdavi, Mojtaba
(författare)
-
Baseband Processing for 5G and Beyond: Algorithms, VLSI Architectures, and Co-design
- 2021
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- In recent years the number of connected devices and the demand for high data-rates have been significantly increased. This enormous growth is more pronounced by the introduction of the Internet of things (IoT) in which several devices are interconnected to exchange data for various applications like smart homes and smart cities. Moreover, new applications such as eHealth, autonomous vehicles, and connected ambulances set new demands on the reliability, latency, and data-rate of wireless communication systems, pushing forward technology developments. Massive multiple-input multiple-output (MIMO) is a technology, which is employed in the 5G standard, offering the benefits to fulfill these requirements. In massive MIMO systems, base station (BS) is equipped with a very large number of antennas, serving several users equipments (UEs) simultaneously in the same time and frequency resource. The high spatial multiplexing in massive MIMO systems, improves the data rate, energy and spectral efficiencies as well as the link reliability of wireless communication systems. The link reliability can be further improved by employing channel coding technique. Spatially coupled serially concatenated codes (SC-SCCs) are promising channel coding schemes, which can meet the high-reliability demands of wireless communication systems beyond 5G (B5G). Given the close-to-capacity error correction performance and the potential to implement a high-throughput decoder, this class of code can be a good candidate for wireless systems B5G. In order to achieve the above-mentioned advantages, sophisticated algorithms are required, which impose challenges on the baseband signal processing. In case of massive MIMO systems, the processing is much more computationally intensive and the size of required memory to store channel data is increased significantly compared to conventional MIMO systems, which are due to the large size of the channel state information (CSI) matrix. In addition to the high computational complexity, meeting latency requirements is also crucial. Similarly, the decoding-performance gain of SC-SCCs also do come at the expense of increased implementation complexity. Moreover, selecting the proper choice of design parameters, decoding algorithm, and architecture will be challenging, since spatial coupling provides new degrees of freedom in code design, and therefore the design space becomes huge. The focus of this thesis is to perform co-optimization in different design levels to address the aforementioned challenges/requirements. To this end, we employ system-level characteristics to develop efficient algorithms and architectures for the following functional blocks of digital baseband processing. First, we present a fast Fourier transform (FFT), an inverse FFT (IFFT), and corresponding reordering scheme, which can significantly reduce the latency of orthogonal frequency-division multiplexing (OFDM) demodulation and modulation as well as the size of reordering memory. The corresponding VLSI architectures along with the application specific integrated circuit (ASIC) implementation results in a 28 nm CMOS technology are introduced. In case of a 2048-point FFT/IFFT, the proposed design leads to 42% reduction in the latency and size of reordering memory. Second, we propose a low-complexity massive MIMO detection scheme. The key idea is to exploit channel sparsity to reduce the size of CSI matrix and eventually perform linear detection followed by a non-linear post-processing in angular domain using the compressed CSI matrix. The VLSI architecture for a massive MIMO with 128 BS antennas and 16 UEs along with the synthesis results in a 28 nm technology are presented. As a result, the proposed scheme reduces the complexity and required memory by 35%–73% compared to traditional detectors while it has better detection performance. Finally, we perform a comprehensive design space exploration for the SC-SCCs to investigate the effect of different design parameters on decoding performance, latency, complexity, and hardware cost. Then, we develop different decoding algorithms for the SC-SCCs and discuss the associated decoding performance and complexity. Also, several high-level VLSI architectures along with the corresponding synthesis results in a 12 nm process are presented, and various design tradeoffs are provided for these decoding schemes.
|
|
| 7. |
- Lian, Mengke, et al.
(författare)
-
What Can Machine Learning Teach Us about Communications
- 2018
-
Ingår i: IEEE International Symposium on Information Theory - Proceedings. - 2157-8095. ; 15 January 2019
-
Konferensbidrag (refereegranskat)abstract
- Rapid improvements in machine learning over the past decade are beginning to have far-reaching effects. For communications, engineers with limited domain expertise can now use off-the-shelf learning packages to design high-performance systems based on simulations. Prior to the current revolution in machine learning, the majority of communication engineers were quite aware that system parameters (such as filter coefficients) could be learned using stochastic gradient descent. It was not at all clear, however, that more complicated parts of the system architecture could be learned as well. In this paper, we discuss the application of machine-learning techniques to two communications problems and focus on what can be learned from the resulting systems. We were pleasantly surprised that the observed gains in one example have a simple explanation that only became clear in hindsight. In essence, deep learning discovered a simple and effective strategy that had not been considered earlier.
|
|
| 8. |
- del Aguila Pla, Pol, 1990-
(författare)
-
Inverse problems in signal processing : Functional optimization, parameter estimation and machine learning
- 2019
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Inverse problems arise in any scientific endeavor. Indeed, it is seldom the case that our senses or basic instruments, i.e., the data, provide the answer we seek. It is only by using our understanding of how the world has generated the data, i.e., a model, that we can hope to infer what the data imply. Solving an inverse problem is, simply put, using a model to retrieve the information we seek from the data.In signal processing, systems are engineered to generate, process, or transmit signals, i.e., indexed data, in order to achieve some goal. The goal of a specific system could be to use an observed signal and its model to solve an inverse problem. However, the goal could also be to generate a signal so that it reveals a parameter to investigation by inverse problems. Inverse problems and signal processing overlap substantially, and rely on the same set of concepts and tools. This thesis lies at the intersection between them, and presents results in modeling, optimization, statistics, machine learning, biomedical imaging and automatic control.The novel scientific content of this thesis is contained in its seven composing publications, which are reproduced in Part II. In five of these, which are mostly motivated by a biomedical imaging application, a set of related optimization and machine learning approaches to source localization under diffusion and convolutional coding models are presented. These are included in Publications A, B, E, F and G, which also include contributions to the modeling and simulation of a specific family of image-based immunoassays. Publication C presents the analysis of a system for clock synchronization between two nodes connected by a channel, which is a problem of utmost relevance in automatic control. The system exploits a specific node design to generate a signal that enables the estimation of the synchronization parameters. In the analysis, substantial contributions to the identifiability of sawtooth signal models under different conditions are made. Finally, Publication D brings to light and proves results that have been largely overlooked by the signal processing community and characterize the information that quantized linear models contain about their location and scale parameters.
|
|
| 9. |
- ur Réhman, Shafiq, 1978-
(författare)
-
Expressing emotions through vibration for perception and control
- 2010
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- This thesis addresses a challenging problem: “how to let the visually impaired ‘see’ others emotions”. We, human beings, are heavily dependent on facial expressions to express ourselves. A smile shows that the person you are talking to is pleased, amused, relieved etc. People use emotional information from facial expressions to switch between conversation topics and to determine attitudes of individuals. Missing emotional information from facial expressions and head gestures makes the visually impaired extremely difficult to interact with others in social events. To enhance the visually impaired’s social interactive ability, in this thesis we have been working on the scientific topic of ‘expressing human emotions through vibrotactile patterns’. It is quite challenging to deliver human emotions through touch since our touch channel is very limited. We first investigated how to render emotions through a vibrator. We developed a real time “lipless” tracking system to extract dynamic emotions from the mouth and employed mobile phones as a platform for the visually impaired to perceive primary emotion types. Later on, we extended the system to render more general dynamic media signals: for example, render live football games through vibration in the mobile for improving mobile user communication and entertainment experience. To display more natural emotions (i.e. emotion type plus emotion intensity), we developed the technology to enable the visually impaired to directly interpret human emotions. This was achieved by use of machine vision techniques and vibrotactile display. The display is comprised of a ‘vibration actuators matrix’ mounted on the back of a chair and the actuators are sequentially activated to provide dynamic emotional information. The research focus has been on finding a global, analytical, and semantic representation for facial expressions to replace state of the art facial action coding systems (FACS) approach. We proposed to use the manifold of facial expressions to characterize dynamic emotions. The basic emotional expressions with increasing intensity become curves on the manifold extended from the center. The blends of emotions lie between those curves, which could be defined analytically by the positions of the main curves. The manifold is the “Braille Code” of emotions. The developed methodology and technology has been extended for building assistive wheelchair systems to aid a specific group of disabled people, cerebral palsy or stroke patients (i.e. lacking fine motor control skills), who don’t have ability to access and control the wheelchair with conventional means, such as joystick or chin stick. The solution is to extract the manifold of the head or the tongue gestures for controlling the wheelchair. The manifold is rendered by a 2D vibration array to provide user of the wheelchair with action information from gestures and system status information, which is very important in enhancing usability of such an assistive system. Current research work not only provides a foundation stone for vibrotactile rendering system based on object localization but also a concrete step to a new dimension of human-machine interaction.
|
|
| 10. |
- Zhang, Kewei
(författare)
-
Secure GNSS-based Positioning and Timing : Distance-Decreasing attacks, fault detection and exclusion, and attack detection with the help of opportunistic signals
- 2021
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- With trillions of devices connected in large scale systems in a wired or wireless manner, positioning and synchronization become vital. Global Navigation Satellite System (GNSS) is the first choice to provide global coverage for positioning and synchronization services. From small mobile devices to aircraft, from intelligent transportation systems to cellular networks, and from cargo tracking to smart grids, GNSS plays an important role, thus, requiring high reliability and security protection. However, as GNSS signals propagate from satellites to receivers at distance of around 20 000 km, the signal power arriving at the receivers is very low, making the signals easily jammed or overpowered. Another vulnerability stems from that civilian GNSS signals and their specifications are publicly open, so that anyone can craft own signals to spoof GNSS receivers: an adversary forges own GNSS signals and broadcasts them to the victim receiver, to mislead the victim into believing that it is at an adversary desired location or follows a false trajectory, or adjusts its clock to a time dictated by the adversary. Another type of attack is replaying GNSS signals: an adversary transmits a pre-recorded GNSS signal stream to the victim receiver, so that the receiver calculates an erroneous position and time. Recent incidents reported in press show that the GNSS functionalities in a certain area, e.g., Black Sea, have been affected by cyberattacks composed of the above-mentioned attack types. This thesis, thus, studies GNSS vulnerabilities and proposes detection and mitigation methods for GNSS attacks, notably spoofing and replay attacks. We analyze the effectiveness of one important and powerful replay attack, the so-called Distance-decreasing (DD) attacks that were previously investigated for wireless communication systems, on GNSS signals. DD attacks are physical layer attacks, targeting time-of-flight ranging protocols, to shorten the perceived as measured distance between the transmitter and receiver. The attacker first transmits an adversary-chosen data bit to the victim receiver before the signal arrives at the attacker; upon receipt of the GNSS signal, the attacker estimates the data bit based on the early fraction of the bit period, and then switches to transmitting the estimate to the victim receiver. Consequently, the DD signal arrives at the victim receiver earlier than the genuine GNSS signals would have, which in effect shortens the pseudorange measurement between the sender (satellite) and the victim receiver, consequently, affecting the calculated position and time of the receiver. We study how the DD attacks affect the bit error rate (BER) of the received signals at the victim, and analyze its effectiveness, that is, the ability to shorten pseudorange measurements, on different GNSS signals. Several approaches are considered for the attacker to mount a DD attack with high probability of success (without being detected) against a victim receiver, for cryptographically unprotected and protected signals. We analyze the tracking output of the DD signals at the victim receiver and propose a Goodness of Fit (GoF) test and a Generalized Likelihood Ratio Test (GLRT) to detect the attacks. The evaluation of the two tests shows that they are effective, with the result being perhaps more interesting when considering DD attacks against Galileo signals that can be cryptographically protected. Moreover, this thesis investigates the feasibility of validating the authenticity of the GNSS signals with the help of opportunistic signals, which is information readily available in modern communication environments, e.g., 3G, 4G and WiFi. We analyze the time synchronization accuracy of different technologies, e.g., Network Time Protocol (NTP), WiFi and local oscillator, as the basis for detecting a discrepancy with the GNSS-obtained time. Two detection approaches are proposed and one testbench is designed for the evaluation. A synthesized spoofing attack is used to verify the effectiveness of the approaches. Beyond attack detection, we develop algorithms to detect and exclude faulty signals, namely the Clustering-based Solution Separation Algorithm (CSSA) and the Fast Multiple Fault Detection and Exclusion (FM-FDE). They both utilize the redundant available satellites, more than the minimum a GNSS receiver needs for position and time offset calculation. CSSA adopts data clustering to group subsets of positions calculated with different subsets of available satellites. Basically, these positions, calculated with subsets not containing any faulty satellites, should be close to each other, i.e., in a dense area; otherwise they should be scattered. FM-FDE is a more efficient algorithm that uses distances between positions, calculated with fixed-size subsets, as test statistics to detect and exclude faulty satellite signals. As the results show, FM-FDE runs faster than CSSA and other solution-separation fault detection and exclusion algorithms while remaining equally effective.
|
|
