Mars: a personalised mobile activity recognition system

Mobile activity recognition focuses on inferring the current activities of a mobile user by leveraging the sensory data that is available on today’s smart phones. The state of the art in mobile activity recognition uses traditional classification learning techniques. Thus, the learning process typically involves: i) collection of labelled sensory data that is transferred and collated in a centralised repository; ii) model building where the classification model is trained and tested using the collected data; iii) a model deployment stage where the learnt model is deployed on-board a mobile device for identifying activities based on new sensory data. In this paper, we demonstrate the Mobile Activity Recognition System (MARS) where for the first time the model is built and continuously updated on-board the mobile device itself using data stream mining. The advantages of the on-board approach are that it allows model personalisation and increased privacy as the data is not sent to any external site. Furthermore, when the user or its activity profile changes MARS enables promptly adaptation. MARS has been implemented on the Android platform to demonstrate that it can achieve accurate mobile activity recognition. Moreover, we can show in practise that MARS quickly adapts to user profile changes while at the same time being scalable and efficient in terms of consumption of the device resources.

Wavelets in High-Capacity DWDM Systems and Modal-Division Multiplexed Transmissions

Optical communications receivers using wavelet signals processing is proposed in this paper for dense wavelength-division multiplexed (DWDM) systems and modal-division multiplexed (MDM) transmissions. The optical signal-to-noise ratio (OSNR) required to demodulate polarization-division multiplexed quadrature phase shift keying (PDM-QPSK) modulation format is alleviated with the wavelet denoising process. This procedure improves the bit error rate (BER) performance and increasing the transmission distance in DWDM systems. Additionally, the wavelet-based design relies on signal decomposition using time-limited basis functions allowing to reduce the computational cost in Digital-Signal-Processing (DSP) module. Attending to MDM systems, a new scheme of encoding data bits based on wavelets is presented to minimize the mode coupling in few-mode (FWF) and multimode fibers (MMF). The Shifted Prolate Wave Spheroidal (SPWS) functions are proposed to reduce the modal interference.

Contribución al análisis eficiente y a la mejora de prestaciones de antenas reflectarray

El diseño de una antena reflectarray bajo la aproximación de periodicidad local requiere la determinación de la matriz de scattering de estructuras multicapa con metalizaciones periódicas para un gran número de geometrías diferentes. Por lo tanto, a la hora de diseñar antenas reflectarray en tiempos de CPU razonables, se necesitan herramientas númericas rápidas y precisas para el análisis de las estructuras periódicas multicapa. En esta tesis se aplica la versión Galerkin del Método de los Momentos (MDM) en el dominio espectral al análisis de las estructuras periódicas multicapa necesarias para el diseño de antenas reflectarray basadas en parches apilados o en dipolos paralelos coplanares. Desgraciadamente, la aplicación de este método numérico involucra el cálculo de series dobles infinitas, y mientras que algunas series convergen muy rápidamente, otras lo hacen muy lentamente. Para aliviar este problema, en esta tesis se propone un novedoso MDM espectral-espacial para el análisis de las estructuras periódicas multicapa, en el cual las series rápidamente convergente se calculan en el dominio espectral, y las series lentamente convergentes se calculan en el dominio espacial mediante una versión mejorada de la formulación de ecuaciones integrales de potenciales mixtos (EIPM) del MDM. Esta versión mejorada se basa en la interpolación eficiente de las funciones de Green multicapa periódicas, y en el cálculo eficiente de las integrales singulares que conducen a los elementos de la matriz del MDM. El novedoso método híbrido espectral-espacial y el tradicional MDM en el dominio espectral se han comparado en el caso de los elementos reflectarray basado en parches apilados. Las simulaciones numéricas han demostrado que el tiempo de CPU requerido por el MDM híbrido es alrededor de unas 60 veces más rápido que el requerido por el tradicional MDM en el dominio espectral para una precisión de dos cifras significativas. El uso combinado de elementos reflectarray con parches apilados y técnicas de optimización de banda ancha ha hecho posible diseñar antenas reflectarray de transmisiónrecepción (Tx-Rx) y polarización dual para aplicaciones de espacio con requisitos muy restrictivos. Desgraciadamente, el nivel de aislamiento entre las polarizaciones ortogonales en antenas DBS (típicamente 30 dB) es demasiado exigente para ser conseguido con las antenas basadas en parches apilados. Además, el uso de elementos reflectarray con parches apilados conlleva procesos de fabricación complejos y costosos. En esta tesis se investigan varias configuraciones de elementos reflectarray basadas en conjuntos de dipolos paralelos con el fin de superar los inconvenientes que presenta el elemento basado en parches apilados. Primeramente, se propone un elemento consistente en dos conjuntos apilados ortogonales de tres dipolos paralelos para aplicaciones de polarización dual. Se ha diseñado, fabricado y medido una antena basada en este elemento, y los resultados obtenidos para la antena indican que tiene unas altas prestaciones en términos de ancho de banda, pérdidas, eficiencia y discriminación contrapolar, además de requerir un proceso de fabricación mucho más sencillo que el de las antenas basadas en tres parches apilados. Desgraciadamente, el elemento basado en dos conjuntos ortogonales de tres dipolos paralelos no proporciona suficientes grados de libertad para diseñar antenas reflectarray de transmisión-recepción (Tx-Rx) de polarización dual para aplicaciones de espacio por medio de técnicas de optimización de banda ancha. Por este motivo, en la tesis se propone un nuevo elemento reflectarray que proporciona los grados de libertad suficientes para cada polarización. El nuevo elemento consiste en dos conjuntos ortogonales de cuatro dipolos paralelos. Cada conjunto contiene tres dipolos coplanares y un dipolo apilado. Para poder acomodar los dos conjuntos de dipolos en una sola celda de la antena reflectarray, el conjunto de dipolos de una polarización está desplazado medio período con respecto al conjunto de dipolos de la otra polarización. Este hecho permite usar solamente dos niveles de metalización para cada elemento de la antena, lo cual simplifica el proceso de fabricación como en el caso del elemento basados en dos conjuntos de tres dipolos paralelos coplanares. Una antena de doble polarización y doble banda (Tx-Rx) basada en el nuevo elemento ha sido diseñada, fabricada y medida. La antena muestra muy buenas presentaciones en las dos bandas de frecuencia con muy bajos niveles de polarización cruzada. Simulaciones numéricas presentadas en la tesis muestran que estos bajos de niveles de polarización cruzada se pueden reducir todavía más si se llevan a cabo pequeñas rotaciones de los dos conjuntos de dipolos asociados a cada polarización. ABSTRACT The design of a reflectarray antenna under the local periodicity assumption requires the determination of the scattering matrix of a multilayered structure with periodic metallizations for quite a large number of different geometries. Therefore, in order to design reflectarray antennas within reasonable CPU times, fast and accurate numerical tools for the analysis of the periodic multilayered structures are required. In this thesis the Galerkin’s version of the Method of Moments (MoM) in the spectral domain is applied to the analysis of the periodic multilayered structures involved in the design of reflectarray antennas made of either stacked patches or coplanar parallel dipoles. Unfortunately, this numerical approach involves the computation of double infinite summations, and whereas some of these summations converge very fast, some others converge very slowly. In order to alleviate this problem, in the thesis a novel hybrid MoM spectral-spatial domain approach is proposed for the analysis of the periodic multilayered structures. In the novel approach, whereas the fast convergent summations are computed in the spectral domain, the slowly convergent summations are computed by means of an enhanced Mixed Potential Integral Equation (MPIE) formulation of the MoM in the spatial domain. This enhanced formulation is based on the efficient interpolation of the multilayered periodic Green’s functions, and on the efficient computation of the singular integrals leading to the MoM matrix entries. The novel hybrid spectral-spatial MoM code and the standard spectral domain MoM code have both been compared in the case of reflectarray elements based on multilayered stacked patches. Numerical simulations have shown that the CPU time required by the hybrid MoM is around 60 times smaller than that required by the standard spectral MoM for an accuracy of two significant figures. The combined use of reflectarray elements based on stacked patches and wideband optimization techniques has made it possible to design dual polarization transmit-receive (Tx-Rx) reflectarrays for space applications with stringent requirements. Unfortunately, the required level of isolation between orthogonal polarizations in DBS antennas (typically 30 dB) is hard to achieve with the configuration of stacked patches. Moreover, the use of reflectarrays based on stacked patches leads to a complex and expensive manufacturing process. In this thesis, we investigate several configurations of reflectarray elements based on sets of parallel dipoles that try to overcome the drawbacks introduced by the element based on stacked patches. First, an element based on two stacked orthogonal sets of three coplanar parallel dipoles is proposed for dual polarization applications. An antenna made of this element has been designed, manufactured and measured, and the results obtained show that the antenna presents a high performance in terms of bandwidth, losses, efficiency and cross-polarization discrimination, while the manufacturing process is cheaper and simpler than that of the antennas made of stacked patches. Unfortunately, the element based on two sets of three coplanar parallel dipoles does not provide enough degrees of freedom to design dual-polarization transmit-receive (Tx-Rx) reflectarray antennas for space applications by means of wideband optimization techniques. For this reason, in the thesis a new reflectarray element is proposed which does provide enough degrees of freedom for each polarization. This new element consists of two orthogonal sets of four parallel dipoles, each set containing three coplanar dipoles and one stacked dipole. In order to accommodate the two sets of dipoles in each reflectarray cell, the set of dipoles for one polarization is shifted half a period from the set of dipoles for the other polarization. This also makes it possible to use only two levels of metallization for the reflectarray element, which simplifies the manufacturing process as in the case of the reflectarray element based on two sets of three parallel dipoles. A dual polarization dual-band (Tx-Rx) reflectarray antenna based on the new element has been designed, manufactured and measured. The antenna shows a very good performance in both Tx and Rx frequency bands with very low levels of cross-polarization. Numerical simulations carried out in the thesis have shown that the low levels of cross-polarization can be even made smaller by means of small rotations of the two sets of dipoles associated to each polarization.

Location of bioelectricity plants in the Madrid Community based on triticale crop: a multicriteria methodology

This paper presents a work whose objective is, first, to quantify the potential of the triticale biomass existing in each of the agricultural regions in the Madrid Community through a crop simulation model based on regression techniques and multiple correlation. Second, a methodology for defining which area has the best conditions for the installation of electricity plants from biomass has been described and applied. The study used a methodology based on compromise programming in a discrete multicriteria decision method (MDM) context. To make a ranking, the following criteria were taken into account: biomass potential, electric power infrastructure, road networks, protected spaces, and urban nuclei surfaces. The results indicate that, in the case of the Madrid Community, the Campiña region is the most suitable for setting up plants powered by biomass. A minimum of 17,339.9 tons of triticale will be needed to satisfy the requirements of a 2.2 MW power plant. The minimum range of action for obtaining the biomass necessary in Campiña region would be 6.6 km around the municipality of Algete, based on Geographic Information Systems. The total biomass which could be made available in considering this range in this region would be 18,430.68 t.

Different mechanisms for suppression of apoptosis by cytokines and calcium mobilizing compounds

Overexpression of wild-type p53 in M1 myeloid leukemia cells induces apoptotic cell death that was suppressed by the calcium ionophore A23187 and the calcium ATPase inhibitor thapsigargin (TG). This suppression of apoptosis by A23187 or TG was associated with suppression of caspase activation but not with suppression of wild-type-p53-induced expression of WAF-1, mdm-2, or FAS. In contrast to suppression of apoptosis by the cytokines interleukin 6 (IL-6) and interferon γ, a protease inhibitor, or an antioxidant, suppression of apoptosis by A23187 or TG required extracellular Ca2+ and was specifically abolished by the calcineurin inhibitor cyclosporin A. IL-6 induced immediate early activation of junB and zif/268 (Egr-1) but A23187 and TG did not. A23187 and TG also suppressed induction of apoptosis by doxorubicin or vincristine in M1 cells that did not express p53 by a cyclosporin A-sensitive mechanism. Suppression of apoptosis by A23187 or TG was not associated with autocrine production of IL-6. Apoptosis induced in IL-6-primed M1 cells after IL-6 withdrawal was not suppressed by A23187 or TG but was suppressed by the cytokines IL-6, IL-3, or interferon γ. The results indicate that these Ca2+-mobilizing compounds can suppress some pathways of apoptosis suppressed by cytokines but do so by a different mechanism.

Peptide-specific cytotoxic T lymphocytes restricted by nonself major histocompatibility complex class I molecules: Reagents for tumor immunotherapy

Studies in melanoma patients have revealed that self proteins can function as targets for tumor-reactive cytotoxic T lymphocytes (CTL). One group of self proteins MAGE, BAGE, and GAGE are normally only expressed in testis and placenta, whilst another group of CTL recognized proteins are melanocyte-specific differentiation antigens. In this study we have investigated whether CTL can be raised against a ubiquitously expressed self protein, mdm-2, which is frequently overexpressed in tumors. The observation that T-cell tolerance is self major histocompatibility complex-restricted was exploited to generate CTL specific for an mdm-2 derived peptide presented by nonself major histocompatibility complex class I molecules. Thus, the allo-restricted T-cell repertoire of H-2d mice was used to isolate CTL specific for the mdm100 peptide presented by allogeneic H-2Kb class I molecules. In vitro, these CTL discriminated between transformed and normal cells, killing specifically Kb-positive melanoma and lymphoma tumors but not Kb-expressing dendritic cells. In vivo, the CTL showed antitumor activity and delayed the growth of melanoma as well as lymphoma tumors in H-2b recipient mice. These experiments show that it is possible to circumvent T-cell tolerance to ubiquitously expressed self antigens, and to target CTL responses against tumors expressing elevated levels of structurally unaltered proteins.

Substance P antagonist (CP-96,345) inhibits HIV-1 replication in human mononuclear phagocytes

Substance P (SP) is a potent modulator of neuroimmunoregulation. We recently reported that human immune cells express SP and its receptor. We have now investigated the possible role that SP and its receptor plays in HIV infection of human mononuclear phagocytes. SP enhanced HIV replication in human blood-isolated mononuclear phagocytes, whereas the nonpeptide SP antagonist (CP-96,345) potently inhibited HIV infectivity of these cells in a concentration-dependent fashion. CP-96,345 prevented the formation of typical giant syncytia induced by HIV Bal strain replication in these cells. This inhibitory effect of CP-96,345 was because of the antagonism of neurokinin-1 receptor, a primary SP receptor. Both CP-96,345 and anti-SP antibody inhibited SP-enhanced HIV replication in monocyte-derived macrophages (MDM). Among HIV strains tested (both prototype and primary isolates), only the R5 strains (Bal, ADA, BL-6, and CSF-6) that use the CCR5 coreceptor for entry into MDM were significantly inhibited by CP-96,345; in contrast, the X4 strain (UG024), which uses CXCR4 as its coreceptor, was not inhibited. In addition, the M-tropic ADA (CCR5-dependent)-pseudotyped HIV infection of MDM was markedly inhibited by CP-96,345, whereas murine leukemia virus-pseudotyped HIV was not affected, indicating that the major effect of CP-96,345 is regulated by Env-determined early events in HIV infection of MDM. CP-96,345 significantly down-regulated CCR5 expression in MDM at both protein and mRNA levels. Thus, SP–neurokinin-1 receptor interaction may play an important role in the regulation of CCR5 expression in MDM, affecting the R5 HIV strain infection of MDM.

Wild-type p53 protein undergoes cytoplasmic sequestration in undifferentiated neuroblastomas but not in differentiated tumors.

Neuroblastoma (NB), a tumor arising from the sympathetic nervous system, is one of the most common malignancies in childhood. Several recent reports on the p53 genotype found virtually exclusive wild-type status in primary tumors, and it was postulated that p53 plays no role in the development of NB. Here, however, we report that the vast majority of undifferentiated NBs exhibit abnormal cytoplasmic sequestration of wild-type p53. This inability of p53 to translocate to the nucleus presumably prevents the protein from functioning as a suppressor. Thirty of 31 cases (96%) of undifferentiated NB showed elevated levels of wild-type p53 in the cytoplasm of all tumor cells concomittant with a lack of nuclear staining. p53 immunoprecipitation from tumor tissues showed a 4.5- to 8-fold increase over normal protein levels. All of 10 tumors analyzed harbored wild-type p53 by direct sequencing of full-length cDNA and Southern blot. In addition, no MDM-2 gene amplification was seen in all 11 tumors analyzed. In contrast, no p53 abnormality was detected in 14 differentiated ganglioneuroblastomas and 1 benign ganglioneuroma. We conclude that loss of p53 function seems to play a major role in the tumorigenesis of undifferentiated NB. This tumor might abrogate the transactivating function of p53 by inhibiting its access to the nucleus, rather than by gene mutation. Importantly, our results suggest that (i) this could be a general mechanism for p53 inactivation not limited to breast cancer (where we first described it) and that (ii) it is found in a tumor previously not thought to be affected by p53 alteration.

Effects of follicular versus luteal phase-based strength training in young women

Hormonal variations during the menstrual cycle (MC) may influence trainability of strength. We investigated the effects of a follicular phase-based strength training (FT) on muscle strength, muscle volume and microscopic parameters, comparing it to a luteal phase-based training (LT). Eumenorrheic women without oral contraception (OC) (N = 20, age: 25.9 ± 4.5 yr, height: 164.2 ± 5.5 cm, weight: 60.6 ± 7.8 kg) completed strength training on a leg press for three MC, and 9 of them participated in muscle biopsies. One leg had eight training sessions in the follicular phases (FP) and only two sessions in the luteal phases (LP) for follicular phase-based training (FT), while the other leg had eight training sessions in LP and only two sessions in FP for luteal phase-based training (LT). Estradiol (E2), progesterone (P4), total testosterone (T), free testosterone (free T) and DHEA-s were analysed once during FP (around day 11) and once during LP (around day 25). Maximum isometric force (Fmax), muscle diameter (Mdm), muscle fibre composition (No), fibre diameter (Fdm) and cell nuclei-to-fibre ratio (N/F) were analysed before and after the training intervention. T and free T were higher in FP compared to LP prior to the training intervention (P < 0.05). The increase in Fmax after FT was higher compared to LT (P <0.05). FT also showed a higher increase in Mdm than LT (P < 0.05). Moreover, we found significant increases in Fdm of fibre type ΙΙ and in N/F only after FT; however, there was no significant difference from LT. With regard to change in fibre composition, no differences were observed between FT and LT. FT showed a higher gain in muscle strength and muscle diameter than LT. As a result, we recommend that eumenorrheic females without OC should base the periodization of their strength training on their individual MC.