Bike sharing demand is increasing in large cities worldwide. The proper functioning of bike-sharing systems is, nevertheless, dependent on a balanced geographical distribution of bicycles throughout a day. In this context, understanding the spatiotemporal distribution of check-ins and check-outs is key for station balancing and bike relocation initiatives. Still, recent contributions from deep learning and distance-based predictors show limited success on forecasting bike sharing demand. This consistent observation is hypothesized to be driven by: i) the strong dependence between demand and the meteorological and situational context of stations; and ii) the absence of spatial awareness as most predictors are unable to model the effects of high-low station load on nearby stations.This work proposes a comprehensive set of new principles to incorporate both historical and prospective sources of spatial, meteorological, situational and calendrical context in predictive models of station demand. To this end, a new recurrent neural network layering composed by serial long-short term memory (LSTM) components is proposed with two major contributions: i) the feeding of multivariate time series masks produced from historical context data at the input layer, and ii) the time-dependent regularization of the forecasted time series using prospective context data. This work further assesses the impact of incorporating different sources of context, showing the relevance of the proposed principles for the community even though not all improvements from the contextaware predictors yield statistical significance.
SUMMARYThis paper focuses on the protection of fixed service (FS) receivers from the aggregate interference produced by the satellites of multiple highly elliptical orbit satellite systems (HEOs). It analyzes the protection given to FS receivers operating in the 18 GHz frequency band by the power flux-density (pfd) mask contained in Article 21 of the 2003 edition of the Radio Regulations [International Telecommunication Union, 2003.]. This mask establishes the maximum allowable value for the pfd produced by any of the satellites of a non-geostationary system at the Earth's surface. The protection offered to FS receivers by this mask is analyzed in four interfering environments, each containing three identical HEO systems. Four types of HEO systems, with different orbital characteristics, are considered: three having satellites that operate only in the northern hemisphere and one having satellites that operate in both hemispheres. All satellite antennas are assumed to radiate 0.31 spot beams. Each HEO satellite is modelled so that the maximum pfd it produces at the Earth's surface just meets the RR Article 21 mask and the analysis takes into account the roll-off characteristics of the satellite antenna beams. To reflect the multiplicity of possibilities concerning the geographical location and technical characteristics of the victim FS receiver (e.g. latitude, longitude, azimuth and elevation of its receiving antenna, antenna gain, receiver noise temperature, etc.) a number of cases were evaluated. The concept of interference in excess [Int. J. Satellite Commun. Networking 2006; 24: 73-95] was used to combine the results corresponding to FS receivers located at the same latitude and having the same receiving antenna elevation angle but for which the location longitude and the azimuth of the pointing direction of its receiving antenna are randomly chosen. Results are expressed in terms of the cumulative distribution function of the interference in excess.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.