64623 research outputs found
Sort by
Using Real-Time Kinematics Algorithm in Mission Critical Communication for Accurate Positioning and Time Correction over 5G and Beyond Networks
Get PDF—At 5G and beyond networks, accurate localization services and nanosecond time synchronization are crucial to enabling mission-critical wireless communications technologies and techniques such as autonomous vehicles and distributed multiple-input and multiple-output (MIMO) antenna systems. This paper investigates how to improve wireless time synchronization by studying time correction based on the Real-Time Kinematics (RTK) positioning algorithm. Using the multiple Global Navigation Satellite System (GNSS) receiver references and the proposed binary GNSS satellite formation to reduce the effect of the ionosphere and troposphere delays and recede the measurement phase-range and pseudorange errors. As a result, it improves user equipment's (UE) localization and measures the time difference between the Base Station (BS) and the UE local clocks. The results show that the positioning accuracy has been increased, and a millimetre accuracy has been achieved while attaining the sub-nanosecond time error (TE) between the UE's and BS local clocks
Implementation of green infrastructure for improving the building environment of elderly care centres
No full textThe elderly population is relatively vulnerable to air pollution and thermal stress due to their low mobility and high prevalence of chronic disorders. Appropriate green infrastructure (GI) deployment can improve both the indoor and outdoor air quality and thermal environments of elderly care centres (ECCs), yet a systematic review on this topic area is lacking. This review aims to fill this gap by investigating the impacts of GI on ECC building environment and presents the approaches for integrating GI into the building environment design. We discussed the significance of linking air quality with the thermal environment to ECCs and the effects of GI on the elderly's physical health. We investigated the key design considerations for GI in ECC buildings (e.g., spatial layout, species, aesthetics and fire prevention). Also, the diversity of monitoring and modelling approaches for evaluating the benefits of GI in indoor and outdoor environments was assessed. Finally, we evaluated the associated challenges and provided design recommendations for improving the environments in and around the ECC buildings (e.g., bedrooms, indoor gardens, green roofs and courtyards). The quantitative evidence for linking GI with indoor and outdoor air pollution and extreme heat around the ECC buildings are limited. However, this evidence-base is important for providing generic advice to the building designers and the elderly. Further studies such as the evaluation criteria and monitoring standard are required to develop holistic design recommendations for ECC buildings. The empirical research about the social and economic impacts is also necessary to facilitate the sustainable development of the ageing societies
UK Vice Chancellor compensation: Do they get what they deserve?
No full textThe compensation received by UK Vice Chancellors (VCs) has been on an upward trend in recent years and attracted a lot of negative media attention. In this paper, we examine whether VCs receive the compensation they deserve. Using a panel dataset covering the academic years 2007/2008 to 2018/2019, we develop a model to predict expected VC compensation to determine whether VCs are over-or undercompensated. Our model finds that VCs are not overcompensated regarding their base salary, but some are overcompensated in terms of their benefits and pension contributions. However, there is very little difference in terms of characteristics of over-and un-dercompensated VCs, indicating that on average, UK VCs receive the compensation they deserve. For robustness purposes, we employ a variety of alternative model specifications and subsamples which all support our previous findings. as well as seminar participants at Queen's University Belfast and SOAS University of London for their valuable comments. All remaining errors are our own
Single Ion Free Energy Calculation in ASIC1: The Importance of the HG loop
Get PDFAcid Sensing Ion Channels (ASICs) are one of the most studied channels of the Epithelial Sodium Channel/Degenerin (ENaC/DEG) superfamily. They are responsible for excitatory responses following acidification of the extracellular medium and are involved in several important physiological roles. The ASIC1 subunit can form a functional homotrimeric channel and its structure is currently the most characterised of the whole ENaC/DEG family. Here we computed the free energy profiles for single ion permeation in two different structures of ASIC1 using both Na+ and Cl- as permeating ions. The first structure is the open structure of the channel from the PDB entry 4NTW, and the second structure is the closed structure with the re-entrant loop which contains the highly conserved `HG' motif form PDB entry 6VTK. Both structures show cation selective free energy profiles, however the profiles of the permeating Na+ differ significantly between the two structures. Indeed, whereas there is only a small energetically favorable (-0.5 kcal mol-1) location for Na+ in the open channel (4NTW) near the end of the pore, we observed a clear ion binding site (-7.8 kcal mol-1) located in between the `GAS' belt and the `HG' loop for the channel containing the re-entrant loop (6VTK). Knowing that the `GAS' motif was determined as the selectivity filter, our results support previous observations while addressing the importance of the `HG' motif for the interactions between the pore and the permeating cations
Deep Learning with Noisy Samples
Get PDFThe remarkable success of deep learning is largely attributed to the collection of large datasets with human-annotated labels. However, it is extremely expensive and time-consuming to label extensive data with high-quality annotations. In other words, noisy samples are inevitable in large datasets. In this thesis, we aim to tackle the issues caused by noisy training samples under two noise-sensitive practical settings: instance recognition task and few-shot classification task.Three contributions are made in this thesis. First, in chapter 3 we investigate the negative effect caused by noisy training samples in instance recognition task, which has been largely neglected by existing works. Specifically, we focus on person re-identification (re-ID)---a cross-domain instance matching problem---and propose to model uncertainty for features of input samples. This extra dimension allows the model to focus more on the clean inliers rather than overfitting to noisy training samples, resulting in better class separability and better generalisation to test data. Second, inspired by the ability of modelling uncertainty, in chapter 4 we focus on making full use of modelling uncertainty by encouraging neuron variance to build a unified framework for multiple applications, including network pruning, adversarial defence, and model calibration. Finally, we move on to few-shot classification problem in chapter 5. Since simultaneously optimising for high per-activation variability/uncertainty and predictive accuracy used in chapter 4 improves the few-shot learning model marginally, we propose hybrid graph neural networks to respectively overcome noisy training samples and class overlapping issues
Learning hybrid ranking representation for person re-identification
No full textContemporary person re-identification (re-id) methods mostly compute independentlya feature representation of each person image in the query set and the gallery set. This strategy fails to consider any ranking context information of each probe image in the query set represented implicitly by the whole gallery set. Some recent re-ranking re-id methods therefore propose to take a post-processing strategy to exploit such contextual information for improving re-id matching performance. However, post-processing is independent of model training without jointly optimising the re-id feature and the ranking context information for better compatibility. In this work, for the first time, we show that the appearance feature and the ranking context information can be jointly optimised for learning more discriminative representations and achieving superior matching accuracy. Specifically, we propose to learn a hybrid ranking representation for person re-id with a two-stream architecture: (1) In the external stream, we use the ranking list of each probe image to learn plausible visual variations among the top ranks from the gallery as the external ranking information; (2) In the internal stream, we employ the part-based fine-grained feature as the internal ranking information, which mitigates the harm of incorrect matches in the ranking list. Assembling these two streams generates a hybrid ranking representation for person matching. Extensive experiments demonstrate the superiority of our method over the state-of-the-art methods on four large-scale re-id benchmarks (Market-1501, DukeMTMC-ReID, CUHK03 and MSMT17), under both supervised and unsupervised settings
RIS Assisted Wireless Powered IoT Networks with Phase Shift Error and Transceiver Hardware Impairment
Get PDFConsidering a reconfigurable intelligent surface (RIS) aided wireless powered Internet of Things (WP IoT) network. To address the energy-limitation issue, IoT devices in such a network can be wirelessly powered by a power station (PS) first and then connect with an access point (AP) using their own harvested energy. The RIS helps enhance energy and information receptions in the downlink wireless energy transfer (WET) and uplink wireless information transfer (WIT), respectively. This work unveils the impact of phase shift error (PSE) and transceiver hardware impairment (THI) on the considered network. Our investigation starts with a scenario where only the impact of the PSE on system under study is considered, then moves toward a scenario with the compound effect of both PSE and THI. A maximization problem of the system sum throughput is formulated to evaluate the overall performance for these two scenarios, subject to the constraints of the adjustable RIS phase shifts, the statistical PSE and the transmission time scheduling. To handle the non-convexity of the formulated problem due to those coupled variables, we first adopt the Lagrange dual method and Karush-Kuhn-Tucker (KKT) conditions to derive the optimal time scheduling in closed-form. Next, we recast the stochastic PSE into the deterministic counterpart for its tractability. Then, we adopt a successive convex approximation (SCA) to iteratively derive the optimal WIT’s phase shifts, and element-wise block coordinate decent (EBCD) and complex circle manifold (CCM) methods to iteratively derive the optimal WET’s phase shifts. Finally, we complete our solution approach for the scenario with both PSE and THI. Simulation results highlight the performance of the proposed scheme and the benefits induced by the RIS in comparison to benchmark schemes
Assessing variability in vascular response to cocoa with personal devices: a series of double-blind randomized cross-over n-of-1 trials
Get PDFControlled clinical intervention studies have demonstrated that cocoa flavanols (CF) can decrease blood pressure and arterial stiffness in healthy humans, although a large variability in the effect size across trials has been reported. Here, we evaluated intra- and inter-individual variability of responses to CF in everyday life using a series of n-of-1 trials in healthy free-living individuals with normal blood pressure carrying personal devices. Eleven healthy young humans participated in a repeated cross-over randomized controlled double-blind n-of-1 trial. On eight consecutive days, each volunteer consumed on alternating days 6 CF capsules (862 mg CF) on four days and 6 matched placebo capsules (P, 0 mg CF/day) on another 4 days in one of two randomized sequences (CF-P-CF-P-CF-P-CF-P or P-CF-P-CF-P-CF-P-CF). On each day the capsules were taken at the same time in the morning with breakfast after baseline measurements. Each subject was provided with an upper arm blood pressure monitor and a finger clip that measures pulse wave velocity (PWV). Measurements of blood pressure, heart rate and PWV were taken at least hourly over 12 hours during the day by the participants. On the first 2 days measurements were performed under supervision to provide training. The overall mixed model analysis showed that CF significantly decreased 12 h systolic blood pressure and PWV by -1.4±0.3 mmHg and - -0.11±0.03 m/s, respectively. Peak effects were observed within the first 3 hours (1.5 h SBP: -4.9±2.2 mmHg, PWV: -0.32±0.17 m/s)and again after 8 h post ingestion. Large inter-individual variation in responses was found (intra-cluster correlation coefficients [ICC]: 0.41, 0.41). When analysing single individuals’ datasets, there was also considerable between-day variation in individual responses that varied greatly between subjects (ICC: 0-0.30, 0-0.22, 0-0.45). Effect sizes inversely correlated with baseline blood pressure values both between-- and within-subjects. The data confirm that cocoa can decrease blood pressure and arterial stiffness in everyday life when elevated within the normal range. The large inter- and intra-individual variation in responses call for more personalized nutritional intervention strategies.</p
Data sources:
Access Repository Dashboard
Do you manage University of Surrey? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!