30796 research outputs found
Sort by
Machine Learning Based Physical Layer Security for Detecting Active Eavesdropping Attacks
Full text linkThis letter explores machine learning for enhancing physical layer security in a wireless system with an access point, legitimate users, and an active eavesdropper. During uplink training, the eavesdropper mimics pilot signals to compromise communication. We propose a framework to extract statistical features from wireless signals and build physical layer datasets. A one-class Support Vector Machine (OC-SVM) is used to detect such active eavesdropping attacks. Additionally, we introduce a twin-class SVM (TC-SVM) model to evaluate and compare detection performance. Simulation results demonstrate that our proposed approach with OC-SVM achieves a detection accuracy of 99.78%, performing favorably compared to the TC-SVM model and other prior methods.</p
Towards CFD Modelling of Multi-Peak Structure of Liquid Hydrogen Storage Tank “BLEVE”
Full text linkLiquid hydrogen (LH2) storage tanks are equipped with pressure relief devices (PRD) to vent hydrogen and avoid the pressure build-up in a tank due to heat transfer from the ambient, including in case of fire. In the event of a PRD failure, the tank structural integrity may be compromised leading to catastrophic rupture of the storage tank releasing the stored energy and producing destructive blast wave, fireball and projectiles. The present paper presents a CFD model to investigate the underlying physical processes of what is called “BLEVE” and assess the blast wave generated by an LH2 storage tank rupture in a fire. The proposed CFD approach advances the previous model (Cirrone et al., 2023) to include the effect of flash evaporation of LH2 during pressure drop after tank rupture and reproduce the multi-peak overpressure structure observed in experiments performed by BMW. Simulation results show that the observed maximum pressure peak is associated with the gaseous phase “explosion”, whereas the series of secondary pressure peaks, smaller in amplitude and of longer duration, are associated with the flash evaporation of the LH2 fraction stored in the tank. Simulations can reproduce the minimum and maximum overpressures measured at 3 m from the storage tank in BMW experiments. The simulated maximum blast wave pressure is seen to increase with the storage pressure and volumetric fraction of the gaseous hydrogen phase in the tank prior to rupture
Post-Quantum Weighted Anonymous Authentication for Hybrid VANET MAC Protocol
Full text linkEfficient Medium Access Control (MAC) protocols are crucial for time-sensitive transmissions of safety and non-safety messages. The IEEE 802.11p standard requires enhancements for varying channel conditions, including error-prone environments. This paper proposes HVMAC, a hybrid VANET MAC protocol that enhances Quality of Service (QoS) for time-sensitive data traffic applications. It categorizes service channels into contention and scheduled channels. The HVMAC protocol is modeled with Markov chains and evaluated against IEEE 802.11p on offered load, average delay, throughput, reliability, and energy consumption (upto 90%). To improve HVMAC security during safety messages (SM) and service advertisement messages (SAM) transmission, a Post-Quantum Weighted Anonymous Authentication (PQWAA) is proposed. The weighted certificate authority (WCA) assigns priority weights to vehicles, for effective traffic management and resource distribution. PQWAA ensures secure authentication and message integrity using quantum-resistant cryptography and pseudonym-based key derivation. Integrating PQWAA with HVMAC ensures energy efficiency and secure communication for both safety and non-safety applications, offering a comprehensive solution for modern VANET environments. The proposed protocol is analyzed with existing techniques based on the packet delivery ratio, throughput, and average packet delay, values of 99%, (45-50)Mbps, and 50 ms. Also, the HVMAC protocol shows 81.23% lower channel collision than TDMA-MAC
A preliminary psychometric investigation of a brief positive school experiences scale
Full text linkIntroduction Universal screening using measures of mental health symptoms is discounted by UK schools due to stakeholder concerns. This is a preliminary study of the psychometric properties of a new Brief Positive School Experiences (B-PSEs) scale, devised to assess associations between cumulative positive school experiences (PSEs) and adolescent mental health as part of a PhD project that explored PSEs count as a potential proxy measure of adolescent mental health risk. Methods Five teachers developed a brief 19-item PSEs measure. 460 adolescents (aged 14–16) completed a survey based on these items and items measuring mental health ( N = 460). Exploratory factor analysis and measures of internal consistency assessed the scale’s preliminary psychometric properties. Results The B-PSEs scale demonstrated a three-factor structure; (1) Coping with school demands; (2) School belonging, safety and support; and (3) Equity and low social adversity. Composite reliability was measured at 0.91. Conclusion The B-PSEs scale is a reliable measure with a robust factor structure and PSEs scores measured using the scale have demonstrated associations with adolescent mental health outcomes
Use of a BMI-independent biomarker-based prostate cancer risk score to identify and triage individuals at risk of prostate disease
Full text linkProstate cancer (PCa) is the second most common cause of cancer related deaths in men in the UK. A national screening programme for PCa does not exist due to the unsuitability of the total prostate specific antigen (tPSA) test which is not specific for PCa and has a high false positive rate. Serum tPSA was measured in n = 25,356 male Randox Health clients. A biomarker-based (tPSA, EGF, MCP-1, IL-8) prostate cancer risk score (PCRS) was then applied to a retrospective cohort (n = 1,142/25,356) of individuals to assess PCa risk. A comparative analysis between tPSA and PCRS indicated that 90.5% of the cohort were assigned low risk of PCa. Of those with an elevated PCRS, 67.8% (78/115) had a normal tPSA value based on tPSA age-adjusted cut-offs. In addition, we observed a significant negative correlation between increasing body mass index (BMI) in men with high BMI (≥ 30) and tPSA levels. No correlation was observed between BMI and PCRS. The tPSA test is potentially unsuitable for use in males with BMI ≥ 30. Use of PCRS could provide more accurate PCa risk stratification for males with BMI ≥ 30. Future assessment of the clinical utility of PCRS is warranted
Cafeteria diet compromises natural adaptations of islet cell transdifferentiation and turnover in pregnancy
Full text linkAbstractBackgroundPancreatic islet β‐cell mass expands during pregnancy, but underlying mechanisms are not fully understood. This study examines the impact of pregnancy and cafeteria diet on islet morphology, associated cellular proliferation/apoptosis rates as well as β‐cell lineage.MethodsNon‐pregnant and pregnant Ins1Cre/+;Rosa26‐eYFP transgenic mice were maintained on either normal or high‐fat cafeteria diet, with pancreatic tissue obtained at 18 days gestation. Immunohistochemical changes in islet morphology, β‐/α‐cell proliferation and apoptosis, as well as islet cell identity, neogenesis and ductal cell transdifferentiation were assessed.ResultsPregnant normal diet mice displayed an increase in body weight and glycaemia. Cafeteria feeding attenuated this weight gain while causing overt hyperglycaemia. Pregnant mice maintained on a normal diet exhibited typical expansion in islet and β‐cell area, owing to increased β‐cell proliferation and survival as well as ductal to β‐cell transdifferentiation and β‐cell neogenesis, alongside decreased β‐cell dedifferentiation. Such pregnancy‐induced islet adaptations were severely restricted by cafeteria diet. Accordingly, islets from these mice displayed high levels of β‐cell apoptosis and dedifferentiation, together with diminished β‐cell proliferation and lack of pregnancy‐induced β‐cell neogenesis and transdifferentiation, entirely opposing islet cell modifications observed in pregnant mice maintained on a normal diet.ConclusionAugmentation of β‐cell mass during gestation arises through various mechanisms that include proliferation and survival of existing β‐cells, transdifferentiation of ductal cells as well as β‐cell neogenesis. Remarkably, cafeteria feeding almost entirely annuls pregnancy‐induced islet adaptations, which may contribute to the development of gestational diabetes in the setting of dietary provoked metabolic stress
The impact of internal and external factors across language domains and features in sequential bilingual acquisition
Full text linkFactors which impact bilingual language development can often interact with different language features. The current study teases apart the impact of internal and external factors (chronological age, length of exposure, L2 richness, L2 use at home, maternal education and maternal L2 proficiency) across linguistic domains and features (vocabulary, morphology and syntax). Participants were 40 Arabic-speaking sequential bilinguals acquiring English (5;7-12;2, M=8;4). Length of exposure predicted vocabulary and morphology, while chronological age predicted syntax. L2 richness also predicted vocabulary and syntax, although the impact on syntax was selective across structures. This split between syntax on the one hand, and vocabulary and morphology on the other, reflects the more embedded properties of the former; this contrasts with vocabulary and morphology, where transfer from the L1 and L2 may be more strongly dependent on the availability of shared forms across languages. Further implications are considered for sequential bilinguals in education contexts
Cross-population validation of the PreMO risk indicator for predicting myopia onset in children
Full text linkPurpose: The Predicting Myopia Onset and progression (PreMO) risk indicator, developed using data generated from white children in the UK, incorporates age, spherical equivalent refraction (SER), axial length (AL) and parental myopia to stratify the likelihood of developing myopia. This study evaluated the PreMO's predictive accuracy using prospective datasets from independent samples of children in Hong Kong (HK) and an ethnically diverse cohort of children in the United Kingdom. Methods: Non-myopic children (SER > –0.50 D) aged 6–8 and 9–10 years were scored using the PreMO risk indicator framework, integrating baseline cycloplegic SER, AL and parental myopia data. Scores were assigned risk categories as follows: 0 = no risk, 1–3 = low risk, 4–6 = moderate risk and 7–9 = high risk. SER at ≥15 years of age was used to define refractive outcomes as ‘myopic’ or ‘not myopic’. PreMO's predictive accuracy was analysed via Receiver Operator Characteristic curves, with Youden's J-Index identifying the optimal risk score threshold. Sensitivity, specificity and area under the curve were determined and compared with those of singular predictors, that is, SER < +0.75 D and AL ≥ 23.07 mm at 6–8 years. Results: In the cohort of children aged 6–8 years, a PreMO risk score ≥ 4 exhibited high sensitivity in predicting myopia onset in UK (0.97) and HK (0.94) children, with high specificity in UK (0.96) and moderate specificity in HK (0.64) children. In UK children aged 6–8 years, the PreMO outperformed singular predictors such as SER and AL. Among HK children aged 9–10 years, the PreMO score maintained high sensitivity (0.90) and moderate specificity (0.72). Conclusions: A PreMO risk score ≥ 4 is a strong predictive indicator for future myopia onset, particularly in UK children. Despite high sensitivity in both UK and HK cohorts, specificity varied, indicating the need for contextual application of the tool, particularly in pre-myopic Asian children.</p
Social connection and its prospective association with adolescent internalising and externalising symptoms: an exploratory cross-country study using retrospective harmonisation
Full text linkBackground: Social connection factors play a key role for young people's mental health. It is important to understand how their influence may vary across contexts. We investigated structural (e.g. household size), functional (e.g. social support) and quality (e.g. feeling close) social connection factors in relation to adolescent internalising and externalising symptoms, comparing two countries Brazil and the United Kingdom (UK). Methods: We pooled data from the UK Millennium Cohort Study (MCS) and the Brazilian High Risk Cohort Study (BHRCS). We included 12 social connection variables, identified through retrospective harmonisation and lived experience expert involvement. We tested measurement invariance and conducted multiple regressions to analyse associations between the social connection factors (age 14) and later internalising and externalising difficulties (age 17.5) in both cohorts. We investigated country-level interactions and used weights to account for attrition, survey design, population representativeness and sample size. Results: We found pooled main associations with later internalising symptoms for ‘living with half-siblings’ (p <.001), ‘moving address’ (p =.001), ‘mother marital status’ (p <.001–.003), ‘bullying’ (p =.001), ‘being bullied’ (p <.001) and ‘difficulties keeping friends’ (p <.001). For externalising, we found main associations with ‘household size’ (p =.041), ‘moving address’ (p =.041), ‘mother's marital status’ (p =.001–.013), ‘bullying others’ (p <.001) and ‘being bullied’ (p <.001). Country-level interactions suggested higher internalising symptoms were associated with ‘household size’ (p =.001) in Brazil and ‘being bullied’ (p <.001) in MCS. Additionally, ‘half-siblings in household’ (p =.003), ‘poor mother–child relationship’ (p =.018), ‘single mother’ (p =.035), ‘bullying’ (p <.001) and ‘being bullied’ (p <.001) were more strongly linked to externalising difficulties in MCS. Conclusions: Social connection factors, mostly structural, contributed to adolescent internalising and externalising difficulties in both countries. Factors relating to bullying and family composition seem to play a stronger role in each country. Cultural and socioeconomic factors might explain these differences. Future research should investigate cross-regional differences to meaningfully inform global mental health efforts.</p
Protestantism and human capital: Evidence from early 20th century Ireland
Full text linkUsing a large individual-level dataset, we explore the significance of religious affiliation for human capital variation in Ireland at the turn of the twentieth century. We construct a large sample based on the returns of male household heads in the 1901 census and explore variation in literacy across the three principal denominations: Roman Catholicism, Anglicanism and Presbyterianism. Protestantism, particularly Presbyterianism, is associated with higher levels of human capital. This denominational effect is remarkably robust, even when accounting for various control variables and alternative modelling specifications. Supplementary analyses reveal that these literacy disparities existed before the foundation of centralised national schooling in 1831 and were independent of school attendance, as Presbyterians exhibited lower attendance rates than Anglicans. We suggest that denomination mattered because it affected the incentives to accrue literacy ability to fully participate in religious and wider cultural life