Smartphone Models Research Articles

GreenHub: a large-scale collaborative dataset to battery consumption analysis of android devices

The development of solutions to improve battery life in Android smartphones and the energy efficiency of apps running on them is hindered by diversity. There are more than 24k Android smartphone models in the world. Moreover, there are multiple active operating system versions, and a myriad application usage profiles. In such a high-diversity scenario, profiling for energy has only limited applicability. One would need to obtain information about energy use in real usage scenarios to make informed, effective decisions about energy optimization. The goal of our work is to understand how Android usage, apps, operating systems, hardware, and user habits influence battery lifespan. We leverage crowdsourcing to collect information about energy in real-world usage scenarios. This data is collected by a mobile app, which we developed and made available to the public through Google Play store, and periodically uploaded to a centralized server and made publicly available to researchers, app developers, and smartphone manufacturers through multiple channels (SQL, REST API, zipped CSV/Parquet dump). This paper presents the results of a wide analysis of the tendency several smart-phone characteristics have on the battery charge/discharge rate, such as the different models, brands, networks, settings, applications, and even countries. Our analysis was performed over the crowdsourced data, and we have presented findings such as which applications tend to be around when battery consumption is the highest, do users from different countries have the same battery usage, and even showcase methods to help developers find and improve energy inefficient processes. The dataset we considered is sizable; it comprises 23+ million (anonymous) data samples stemming from a large number of installations of the mobile app. Moreover, it includes 700+ million data points pertaining to processes running on these devices. In addition, the dataset is diverse. It covers 1.6k+ device brands, 11.8k+ smartphone models, and more than 50 Android versions. We have been using this dataset to perform multiple analyses. For example, we studied what are the most common apps running on these smartphones and related the presence of those apps in memory with the battery discharge rate of these devices. We have also used this dataset in teaching, having had students practicing data analysis and machine learning techniques for relating energy consumption/charging rates with many other hardware and software qualities, attributes and user behaviors. The dataset we considered can support studies with a wide range of research goals, be those energy efficiency or not. It opens the opportunity to inform and reshape user habits, and even influence the development of both hardware (manufacturers) and software (developers) for mobile devices. Our analysis also shows results which go outside of the common perception of what impacts battery consumption in real-world usage, while exposing new varied, complex, and promising research avenues.

Dynamic Pricing Recognition on E-Commerce Platforms with VAR Processes

In an environment such as e-commerce, characterized by the presence of numerous agents, competition based on product characteristics is a very important aspect. This paper proposes a model based on vector autoregressive processes (VAR) and Lasso penalization to detect and examine the dynamics that govern real-time price competition in electronic marketplaces. Employing this model, an empirical study was performed on the price trends of smartphone models on the major electronic sales platforms of the Italian market. The proposed model detects real-time price variations in single vendors, based on the variations of their direct competitors. The statistical method adopted in this analysis may be useful for e-commerce companies that conduct market analyses of competitors’ pricing strategies.

Remote embedded devices test framework on the cloud

SummaryEmbedded systems have high coupling and dependency among different hardware and software components in heterogeneous layers, which makes location and issue tracking in their testing difficult. Despite these poor verification conditions, even the most important reliability quality verification among embedded system characteristics is verified with insufficient sample size, typical test cases, and general test strategies, following limitations such as development costs and scheduling. As a result, shipments are highly likely to lead to various reliability quality problems because items have not been verified considering reliability quality characteristics. Hence, to address this gap, this study developed remote embedded device test framework on the cloud (RED‐TFC), which has an innovative reliability test manager component that can automatically perform various tests for the evaluation of reliability and performance of distributed shared devices by utilizing the cloud concept. RED‐TFC offers two key enhancements over existing testing services: (i) the adaptive sample scale for reliability test (ASRT), a feature that identifies the most appropriate sample size for performing functionality and reliability tests of remote verification targets connected to the RED‐TFC server; and (ii) the mass sample reliability test (MSRT), which uses a test case that is specific to reliability, with the sample size obtained by ASRT, to perform verification following the Markov prediction process. This paper analyses two Android smartphone models considered the most generic examples, including many embedded components, and presents a method of detecting a high number of reliability problems in smartphones using the proposed RED‐TFC and its implications.

Collaborative 3D Monitoring for Coastal Survey: Conclusive Tests and First Feedbacks Using the SELPhCoAST Workflow

“Structure from Motion” by smart devices photography (SfM-S) is a current promising tool to support 3D participatory monitoring of heritage and geosites. Within the ODySéYeu scientific project, which aims to analyze the sedimentary dynamics around the isle of Yeu (France), a consortium of different local actors was invited to collectively develop an integrated solution to monitor the coastal evolution of the isle in 3D. This new solution is composed of a combination of an application (SENTINELLES de la côte) and a specific workflow (SELPhCoAST), which guides citizens to acquire reliable in-situ photo datasets with their own smart device, facilitates data transmission, and allows both a reliable data treatment and a public presentation of matters and results. This study presents the results of reproducibility and accuracy tests performed on a test-site to validate the SELPhCoAST workflow, over 20 months of monitoring, with multiple operators and devices. Tests show that at least 95% of each SfM-S model is identical to the terrestrial laser scans models within a range of ±5 cm, almost independently of the operator or smartphone model. This study highlights the interest of this cost-effective participatory solution to understand coastal erosion processes, and facilitate the establishment of soft and nature-based solutions of protection, management or restoration of sites.

Evaluation of the best smartphone model in the market by integrating fuzzy-AHP and PROMETHEE decision-making approach

Smartphones nowadays have become the most globally demanded electronic devices. Each year, almost thousands of new smartphone models are launched by different brands having diverse specifications and variable price range. Due to the availability of large smartphone quantity in the market, sometimes the buyers get confused about the most appropriate model to buy at low price with respect to descent hardware specifications. This research article states such a type of decision problem and solves it using multi-criteria decision-making techniques that enable the mobile users to select the best smartphone model. The main objective of the study is to choose the best smartphone model among 10 selected models available in the market with the help of fuzzy analytical hierarchy process (FAHP) and preference ranking organization method for enrichment of evaluations (PROMETHEE) where FAHP based on trapezoidal fuzzy numbers is used to evaluate the criteria weights and PROMETHEE is used to propose the preference ranking order of the alternatives. Seven important specifications (i.e., criteria), namely price, storage capacity, RAM, battery, screen size, camera and clock speed, are considered based on which the selection procedure is carried out. PROMETHEE I provides the partial alternatives ranking and shows the mode of preference of each model over other models, whereas PROMETHEE II delivers the complete ranking of the alternatives which recommends Realme 5i and Samsung J6 Plus as the best and worst model, respectively, among the group.

Predicting Symptoms of Depression and Anxiety Using Smartphone and Wearable Data.

Background: Depression and anxiety are leading causes of disability worldwide but often remain undetected and untreated. Smartphone and wearable devices may offer a unique source of data to detect moment by moment changes in risk factors associated with mental disorders that overcome many of the limitations of traditional screening methods.Objective: The current study aimed to explore the extent to which data from smartphone and wearable devices could predict symptoms of depression and anxiety.Methods: A total of N = 60 adults (ages 24–68) who owned an Apple iPhone and Oura Ring were recruited online over a 2-week period. At the beginning of the study, participants installed the Delphi data acquisition app on their smartphone. The app continuously monitored participants' location (using GPS) and smartphone usage behavior (total usage time and frequency of use). The Oura Ring provided measures related to activity (step count and metabolic equivalent for task), sleep (total sleep time, sleep onset latency, wake after sleep onset and time in bed) and heart rate variability (HRV). In addition, participants were prompted to report their daily mood (valence and arousal). Participants completed self-reported assessments of depression, anxiety and stress (DASS-21) at baseline, midpoint and the end of the study.Results: Multilevel models demonstrated a significant negative association between the variability of locations visited and symptoms of depression (beta = −0.21, p = 0.037) and significant positive associations between total sleep time and depression (beta = 0.24, p = 0.023), time in bed and depression (beta = 0.26, p = 0.020), wake after sleep onset and anxiety (beta = 0.23, p = 0.035) and HRV and anxiety (beta = 0.26, p = 0.035). A combined model of smartphone and wearable features and self-reported mood provided the strongest prediction of depression.Conclusion: The current findings demonstrate that wearable devices may provide valuable sources of data in predicting symptoms of depression and anxiety, most notably data related to common measures of sleep.

Detecting Cataract Using Smartphones.

Objective: Cataract, which is the clouding of the crystalline lens, is the most prevalent eye disease accounting for 51% of all eye diseases in the U.S. Cataract is a progressive disease, and its early detection is critical for preventing blindness. In this paper, an efficient approach to identify cataract disease by adopting luminance features using a smartphone is proposed. Methods: Initially, eye images captured by a smartphone were cropped to extract the lens, and the images were preprocessed to remove irrelevant background and noise by utilizing median filter and watershed transformation. Then, a novel luminance transformation from pixel brightness algorithm was introduced to extract lens image features. The luminance and texture features of different types of cataract disease images could be obtained accurately in this stage. Finally, by adopting support vector machines (SVM) as the classification method, cataract eyes were identified. Results: From all the images that we fed into our system, our method could diagnose diseased eyes with 96.6% accuracy, 93.4% specificity, and 93.75% sensitivity. Conclusion: The proposed method provides an affordable, rapid, easy-to-use, and versatile method for detecting cataracts by using smartphones without the use of bulky and expensive imaging devices. This methodcan be used for bedside telemedicine applications or in remote areas that have medical shortages. Previous smartphone-based cataract detection methods include texture feature analysis with 95 % accuracy, Gray Level Co-occurrence Matrix (GLCM) method with 89% accuracy, red reflex measurement method, and RGB color feature extraction method using cascade classifier with 90% accuracy. The accuracy of cataract detection in these studies is subject to changes in smartphone models and/or environmental conditions. However, our novel luminance-based method copes with different smartphone camera sensors and chroma variations, while operating independently from sensors’ color characteristics and changes in distances and camera angle. Clinical and Translational Impact—This study is an early/pre-clinical research proposing a novel luminance-based method of detecting cataract using smartphones for remote/at-home monitoring and telemedicine application.

Research of the Possibility of Creating a Measuring System Based on a Modern Smartphone for Estimating the Parameters of Object Movement

When studying different fast-moving processes, high-speed video recording is now widely used. The main deterrent for its further distribution remains the high cost of the necessary equipment. Herewith modern mobile devices (including smartphones) are constantly improved, some models are supplied with video cameras with high recording speed - 1000 frames/s and higher. In the paper the possibility of creating a measuring system based on one of the modern smartphone models to analyze the parameters of object motion is studied. The block diagram of the system is given; a layout of such a system has been created. To verify the workability of the developed model of the system, a software and hardware facility was developed to imitate the motion of an object on the Arduino Uno platform with an engine and a carriage moving along the guides. Graphs of changing the coordinates and speed of the object obtained using the program Tracker 5.1.5 are shown. A theoretical estimate of the error in determining the speed of an object using the implemented model of the system is performed. In addition, the estimation of the range of object speeds that can be measured using the developed system, as well as the range of permissible object movements was performed. The results of the performed experiments prove the possibility of creating a measuring system on the basis of the considered model of the smartphone to assess the speed of moving small objects with certain limitations. At the same time, to create a full-fledged measuring system that does not require additional processing of the received resulting video image on a PC and is capable of calculating the necessary results immediately after the end of the measurement process, it is necessary to develop our own software, which is advisable to do using the existing free libraries.

Validity of Consumer Activity Monitors and an Algorithm Using Smartphone Data for Measuring Steps during Different Activity Types.

Background: Consumer activity monitors and smartphones have gained relevance for the assessment and promotion of physical activity. The aim of this study was to determine the concurrent validity of various consumer activity monitor models and smartphone models for measuring steps. Methods: Participants completed three activity protocols: (1) overground walking with three different speeds (comfortable, slow, fast), (2) activities of daily living (ADLs) focusing on arm movements, and (3) intermittent walking. Participants wore 11 activity monitors (wrist: 8; hip: 2; ankle: 1) and four smartphones (hip: 3; calf: 1). Observed steps served as the criterion measure. The mean average percentage error (MAPE) was calculated for each device and protocol. Results: Eighteen healthy adults participated in the study (age: 28.8 ± 4.9 years). MAPEs ranged from 0.3–38.2% during overground walking, 48.2–861.2% during ADLs, and 11.2–47.3% during intermittent walking. Wrist-worn activity monitors tended to misclassify arm movements as steps. Smartphone data collected at the hip, analyzed with a separate algorithm, performed either equally or even superiorly to the research-grade ActiGraph. Conclusion: This study highlights the potential of smartphones for physical activity measurement. Measurement inaccuracies during intermittent walking and arm movements should be considered when interpreting study results and choosing activity monitors for evaluation purposes.

Photogrammetric 3D Model via Smartphone GNSS Sensor: Workflow, Error Estimate, and Best Practices

Geotagged smartphone photos can be employed to build digital terrain models using structure from motion-multiview stereo (SfM-MVS) photogrammetry. Accelerometer, magnetometer, and gyroscope sensors integrated within consumer-grade smartphones can be used to record the orientation of images, which can be combined with location information provided by inbuilt global navigation satellite system (GNSS) sensors to geo-register the SfM-MVS model. The accuracy of these sensors is, however, highly variable. In this work, we use a 200 m-wide natural rocky cliff as a test case to evaluate the impact of consumer-grade smartphone GNSS sensor accuracy on the registration of SfM-MVS models. We built a high-resolution 3D model of the cliff, using an unmanned aerial vehicle (UAV) for image acquisition and ground control points (GCPs) located using a differential GNSS survey for georeferencing. This 3D model provides the benchmark against which terrestrial SfM-MVS photogrammetry models, built using smartphone images and registered using built-in accelerometer/gyroscope and GNSS sensors, are compared. Results show that satisfactory post-processing registrations of the smartphone models can be attained, requiring: (1) wide acquisition areas (scaling with GNSS error) and (2) the progressive removal of misaligned images, via an iterative process of model building and error estimation.

P(VDF-TrFE) Thin-Film-Based Transducer for Under-Display Ultrasonic Fingerprint Sensing Applications

The use of fingerprint for biometric identification is one of the most prevalent authentication methods applied today in smartphones. In the course of pursuing narrow-bezel and full-screen display, the under-display fingerprint sensor is considered to be a user-friendly and practical solution for newer models of smartphone. While under-display optical fingerprint sensor has been commercially available in various smartphones, it demonstrates limitations such as sensitivity to humidity and contaminations including oil and water as well as easy to spoof. Ultrasonic fingerprint sensing has been proved to be able to overcome these limitations. In this study, P(VDF-TrFE) piezoelectric polymer-based transducer was reported, for the first time, for under-display ultrasonic fingerprint sensing applications. In specific, a 40 MHz ultrasonic transducer using a layer of $10~\mu \text{m}$ thick P(VDF-TrFE) thin-film was designed, fabricated, and characterized. The under-display ultrasonic fingerprint sensing capability of the prototyped transducer was experimentally validated using phantoms of real fingerprint. Electronic images of fingerprint with resolution of $500\times500$ DPI were obtained through under-display ultrasonic fingerprint sensing tests. The lateral resolution of the transducer was calculated to be $\sim ~70~\mu \text{m}$ . The results of this study illustrate promising advances in under-display ultrasonic fingerprint sensing applications.

Factors Influencing the Use of Smartphones for Programing: A Structural Equation Modeling Approach

The ubiquitous application of smartphones and their advanced development have created an opportunity for using them as coding platforms. In this study, we compared between the use of personal computers (PCs) and smartphones to investigate the factors affecting the use of smartphones in programing. The behavioral intentions of smartphone end-users are inspired by the ease of use perception, enjoyment perception, programing anxiety, perception of external control, and smartphone design aesthetics. Although the [Formula: see text] value of the smartphone model was lower than that of the PC model, the end-users’ adoption decisions could have shifted toward accepting the use of smartphones for programing had their decisions been free of enforcement. In this study, design aesthetics and programing anxiety were introduced to Technology Acceptance Model 3 in an Arabic environment. Additionally, the model was creatively applied for guiding practitioners in two situations. First, when organizations are in the quest for emerging technology to replace the legacy technology, they might apply the presented side-by-side comparison of the use of PCs with that of smartphones in programing. Second, at a time when decision makers analyze what might hinder the adoption of a recently introduced technology, the model can be a successful hand tool guiding the top management in identifying technology acceptance interventions, an approach this research revealed. Furthermore, in this paper, the research implications and recommendations are presented for technology practitioners and designers.

Two‐tiered face verification with low‐memory footprint for mobile devices

Mobile devices have their popularity and affordability greatly increased in recent years. As a consequence of their ubiquity, these devices now carry all sorts of personal data that should be accessed only by their owner. Even though knowledge-based procedures are still the main methods to secure the owner's identity, recently biometric traits have been employed for more secure and effortless authentication. In this work, the authors propose a facial verification method optimised to the mobile environment. It consists of a two-tiered procedure that combines hand-crafted features and a convolutional neural network (CNN) to verify if the person depicted in a photograph corresponds to the device owner. To train a CNN for the verification task, the authors propose a hybrid-image input, which allows the network to process encoded information of a pair of face images. The proposed experiments show that the solution outperforms state of the art face verification methods, providing a 4× speedup when processing an image in recent smartphone models. Additionally, the authors show that the two-tiered procedure can be coupled with existing face verification CNNs improving their accuracy and efficiency. They also present a new data set of selfie pictures – RECOD Selfie data set – that hopefully will support future research in this scenario.

Comparison of Smartphone and Drone Lidar Methods for Characterizing Spatial Variation in PAI in a Tropical Forest

Estimating leaf area index (LAI) and assessing spatial variation in LAI across a landscape is crucial to many ecological studies. Several direct and indirect methods of LAI estimation have been developed and compared; however, many of these methods are prohibitively expensive and/or time consuming. Here, we examine the feasibility of using the free image processing software CAN-EYE to estimate effective plant area index (PAIeff) from hemispherical canopy images taken with an extremely inexpensive smartphone clip-on fisheye lens. We evaluate the effectiveness of this inexpensive method by comparing CAN-EYE smartphone PAIeff estimates to those from drone lidar over a lowland tropical forest at La Selva Biological Station, Costa Rica. We estimated PAIeff from drone lidar using a method based in radiative transfer theory that has been previously validated using simulated data; we consider this a conservative test of smartphone PAIeff reliability because above-canopy lidar estimates share few assumptions with understory image methods. Smartphone PAIeff varied from 0.1 to 4.4 throughout our study area and we found a significant correlation (r = 0.62, n = 42, p < 0.001) between smartphone and lidar PAIeff, which was robust to image processing analytical options and smartphone model. When old growth and secondary forests are assumed to have different leaf angle distributions for the lidar PAIeff algorithm (spherical and planophile, respectively) this relationship is further improved (r = 0.77, n = 42, p < 0.001). However, we found deviations in the magnitude of the PAIeff estimations depending on image analytical options. Our results suggest that smartphone images can be used to characterize spatial variation in PAIeff in a complex, heterogenous tropical forest canopy, with only small reductions in explanatory power compared to true digital hemispherical photography.

Smartphone sensor accuracy varies from device to device in mobile research: The case of spatial orientation

Smartphone usage is increasing around the globe—in daily life and as a research device in behavioral science. Smartphones offer the possibility to gather longitudinal data at little cost to researchers and participants. They provide the option to verify self-report data with data from sensors built into most smartphones. How accurate this sensor data is when gathered via different smartphone devices, e.g., in a typical experience sampling framework, has not been investigated systematically. With the present study, we investigated the accuracy of orientation data about the spatial position of smartphones via a newly invented measurement device, the RollPitcher. Objective status of pitch (vertical orientation) and roll (horizontal orientation) of the smartphone was compared to data gathered from the sensors via web browsers and native apps. Bayesian ANOVAs confirmed that the deviations in pitch and roll differed between smartphone models, with mean inaccuracies per device of up to 2.1° and 6.6°, respectively. The inaccuracies for measurements of roll were higher than for pitch, d = .28, p < .001. Our results confirm the presence of heterogeneities when gathering orientation data from different smartphone devices. In most cases, measurement via a web browser was identical to measurement via a native app, but this was not true for all smartphone devices. As a solution to lack of sensor accuracy, we recommend the development and implementation of a coherent research framework and also discuss the implications of the heterogeneities in orientation data for different research designs.

Testing Walking-Induced Vibration of Floors Using Smartphones Recordings

Smartphone technology is rapidly evolving, adding sensors of growing accuracy and precision. Structural engineers are among customers who indirectly benefit from such technological advances. This paper tests whether accelerometers installed in new generations of smartphones can reach the accuracy of professional accelerometers created for vibration monitoring of civil engineering structures, and how they can be useful. The paper describes an experimental study designed to measure walking-induced vibrations of a slender prefabricated prestressed concrete slab. Both traditional, high-accuracy, accelerometers and those integrated into commercial smartphones were used for experimental data collection. Direct comparison of the recordings yielded two key findings: the accuracy of smartphone accelerometers largely depends on the specific smartphone model, and nevertheless is satisfactory for preliminary modal testing at the very least. Furthermore, the smartphone measured accelerations of the lower back were used successfully to indirectly measure pedestrian walking loads.

Predictors of problematic smartphone use among university students

Predictors of problematic smartphone use have been found mainly in studies on elementary and high school students. Few studies have focused on predictors related to social network and messaging apps or smartphone model. Thus, the objective of our study was to identify predictors of problematic smartphone use related to demographic characteristics, loneliness, social app use, and smartphone model among university students. This cross-sectional study involved 257 Brazilian university students who answered a smartphone addiction scale, a questionnaire about smartphone usage patterns, and the Brazilian version of the UCLA-R loneliness scale. Women, iPhone owners, and users of Instagram and Snapchat had significantly higher smartphone addiction scores. We found correlations between scores for the Brazilian version of smartphone addiction scale and the importance attributed to WhatsApp, Facebook, Instagram, and Snapchat, and the Brazilian version of the UCLA-R loneliness scale. Our hierarchical regression model predicted 32.2% of the scores of the Brazilian version of the smartphone addiction scale, with the greatest increase in predictive capability by the step that added smartphone social app importance, followed by the step that added loneliness. Adding the smartphone model produced the smallest increase in predictive capability. The theoretical and practical implications of these results are discussed.

Using smartphone leaf area index data acquired in a collaborative context within vineyards in southern France

Aim: The objective of this short communication is to study the opportunity of using a smartphone application for leaf area index (LAI) observations within vineyards in southern France in a farmsourcing context, i.e. when several operators make parallel acquisitions over the same area. In this context, several sources of variability are likely to affect measurement quality, such as the smartphone model or the operator. Understanding these sources of variability will enhance the ability to properly interpret LAI observations to produce relevant information for decision-making.Methods and material: A study was specifically designed to evaluate the ability of a smartphone application to differentiate sites with different LAI and to determine the origin and the relative importance of different sources of variation in a context of farmsourcing data collection. This focused on the VitiCanopy application, which has been developed specifically for viticulture LAI measurements. Measurements were performed by 18 operators with 11 different smartphone models, on three different vines presenting controlled canopy size to evaluate the ability of the smartphone application to differentiate sites under varying acquisition conditions. Controlled repetitions over seven vines by seven operators with seven smartphone models were performed to further determine the sources of variation and their relative importance.Results: LAI estimations made with VitiCanopy were consistent with the different levels of controlled vine size in the experiment. The operator and the smartphone model had a significant effect on the variance of the estimated LAI. The variance caused by the observation protocol was relatively low compared to the variability between plants within the observation site (seven vines).Conclusions: This study showed that the VitiCanopy application was relevant for ordering or classifying vines according to LAI. In an operational context, the results of this study support the use of this smartphone application for relative measurements. However, the best results were achieved when smartphone model differences were minimised or avoided and with homogeneous acquisition conditions between operators. This last condition will require the organisation of group training sessions to minimise an observed operator effect on measurement variability.Significance and impact of the study: This short communication demonstrated the potential of LAI observations collected with smartphones by several operators for decision-making in a context of farmsourcing. The results showed that this new source of observations, which is inexpensive to collect, made it possible to characterise vine size (LAI) differences in vineyards of southern France. This shows the potential of this app for large production areas such as cooperatives. Further investigations are needed to understand how different training systems may affect the measurement. This source of observations could be complementary to other information sources that are more precise or more accurate, but also more expensive (i.e. destructive methods).

CORONAVIRUS COVID-19 ROILS DISPLAY SUPPLY CHAIN.

CORONAVIRUS COVID-19 ROILS DISPLAY SUPPLY CHAIN.

BioTouchPass2: Touchscreen Password Biometrics Using Time-Aligned Recurrent Neural Networks

Passwords are still used on a daily basis for all kind of applications. However, they are not secure enough by themselves in many cases. This work enhances password scenarios through two-factor authentication asking the users to draw each character of the password instead of typing them as usual. The main contributions of this study are as follows: i) We present the novel MobileTouchDB public database, acquired in an unsupervised mobile scenario with no restrictions in terms of position, posture, and devices. This database contains more than 64K on-line character samples performed by 217 users, with 94 different smartphone models, and up to 6 acquisition sessions. ii) We perform a complete analysis of the proposed approach considering both traditional authentication systems such as Dynamic Time Warping (DTW) and novel approaches based on Recurrent Neural Networks (RNNs). In addition, we present a novel approach named Time-Aligned Recurrent Neural Networks (TA-RNNs). This approach combines the potential of DTW and RNNs to train more robust systems against attacks. A complete analysis of the proposed approach is carried out using both MobileTouchDB and e-BioDigitDB databases. Our proposed TA-RNN system outperforms the state of the art, achieving a final 2.38% Equal Error Rate, using just a 4-digit password and one training sample per character. These results encourage the deployment of our proposed approach in comparison with traditional typed-based password systems where the attack would have 100% success rate under the same impostor scenario.