The particle filter algorithm: parallel implementations and performance analysis over Android mobile devices

The limited computational resources of mobile devices significantly constrain their ability to effectively execute resource-intensive computer vision tasks, such as , video processing, augmented reality, and face recognition. Handwritten character recognition (HCR) is a critical application of machine learning and computer vision, with wide-ranging implications for digital transformation and accessibility. This proposed study explores the implementation of a deep learning-based HCR system tailored for mobile phones using cloudlet. A cloudlet is a novel technology aimed at boosting the computational power of mobile devices to manage resource-intensive tasks. It enables mobile clients to interact with nearby servers, allowing only minimally processed data to be transmitted from the mobile application to the server. The server handles the complex computations and sends the results back to the mobile device in real-time. By utilizing Convolutional Neural Networks (CNNs), the proposed method achieves high accuracy and efficiency on resource-constrained mobile devices. The research details the system architecture, dataset preprocessing, model training, and integration into an Android application, offering a robust solution for real-time handwritten character recognition.

The widely development of mobile devices is contributing to a high demand in 3D graphics, as they have also become a very important requirement of modern applications. Virtual Globes integrating environmental data at any time or place, remain a challenge within the technical constraints imposed by mobile devices. We present Glob3 Mobile, an open source framework for the development of virtual globes on familiar iOS and Android mobile devices. The paper discusses the design and development choices for each platform. The aim of this work is twofold. First, to provide an efficient Virtual Globe application, testable and freely accessible from the web and providing a truly 3D navigation experience with smooth flying. Second, to provide the main software components to easily design and implement 3D Virtual Globes based applications, on both iOS and Android platforms.KeywordsVirtual globeSmartphone applicationiOSAndroid

The study aims for the development of a control system for D.C motors through an application made for Android mobile devices. The D.C motors are represented by a robot model car and the control application communicates with the system via Bluetooth technology. The Android mobile device user must install an application on his mobile device, then, the user must turn on the Bluetooth communication on the mobile device. The user can use various commands to control the robot model car such as front, back, stop, left, right. These commands are sent from the Android mobile device to the Bluetooth receiver, which is interfaced with the control system. The control system has a Bluetooth HC-06 receiver that receives commands from the control application and sends them to the Arduino Nano microcontroller to control the motors through two L298N drivers.

Advances in ubiquitous displays and wireless communications have fueled the emergence of exciting mobile graphics applications including 3D virtual product catalogs, 3D maps, security monitoring systems and mobile games. Current trends that use cameras to capture geometry, material re?ectance and other graphics elements mean that very high resolution inputs are accessible to render extremely photorealistic scenes. However, captured graphics content can be many gigabytes in size, and must be simpli?ed before they can be used on small mobile devices, which have limited resources, such as memory, screen size and battery energy. Scaling and converting graphics content to a suitable rendering format involves running several software tools, and selecting the best resolution for target mobile device is often done by trial and error, which all takes time. Wireless errors can also affect transmitted content and aggressive compression is needed for low-bandwidth wireless networks. Most rendering algorithms are currently optimized for visual realism and speed, but are not resource or energy ef?cient on a mobile device. This chapter focuses on the improvement of rendering performance by reducing the impacts of these problems with UbiWave, an end-to-end framework to enable real time mobile access to high resolution graphics using wavelets. The framework tackles the issues including simpli?cation, transmission, and resource ef?cient rendering of graphics content on mobile device based on wavelets by utilizing 1) a Perceptual Error Metric (PoI) for automatically computing the best resolution of graphics content for a given mobile display to eliminate guesswork and save resources, 2) Unequal Error Protection (UEP) to improve the resilience to wireless errors, 3) an Energy-ef?cient Adaptive Real-time Rendering (EARR) heuristic to balance energy consumption, rendering speed and image quality and 4) an Energy-ef?cient Streaming Technique. The results facilitate a new class of mobile graphics application which can gracefully adapt the lowest acceptable rendering resolution to the wireless network conditions and the availability of resources and battery energy on mobile device adaptively.

Video verification on mobile devices, such as cell phones and tablet PCs, allows users to remotely check what is happening in a particular place without the need to constantly monitor the site and record video from the camera. This requires appropriate stream encoding to be applied. Unfortunately, existing encoding standards, widely used in video monitoring, have several drawbacks. In the paper, we show some remarks on choosing video stream encoding for video verification on Android-based mobile devices. Results of performed tests on bandwidth usage have shown that encoding based on passing static frames in JPEG format translates very well into an environment where the camera's resolution is often higher than the mobile device screen.

Video-based moving vehicle detection and tracking is an important prerequisite for vehicle counting under complex transportation environments. However, in the complex natural scene, the conventional optical flow method cannot accurately detect the boundary of the moving vehicle due to the generation of the shadow. In addition, traditional vehicle tracking algorithms are often occluded by trees, buildings, etc., and particle filters are also susceptible to particle degradation. To solve this problem, this paper proposes a kind of moving vehicle detection and tracking based on the optical flow method and immune particle filter algorithm. The proposed method firstly uses the optical flow method to roughly detect the moving vehicle and then uses the shadow detection algorithm based on the HSV color space to mark the shadow position after threshold segmentation and further combines the region-labeling algorithm to realize the shadow removal and accurately detect the moving vehicle. Improved affinity calculation and mutation function of antibody are proposed to make the particle filter algorithm have certain adaptivity and robustness to scene interference. Experiments are carried out in complex traffic scenes with shadow and occlusion interference. The experimental results show that the proposed algorithm can well solve the interference of shadow and occlusion and realize accurate detection and robust tracking of moving vehicles under complex transportation environments, which has the potentiality to be processed on a cloud computing platform.

Internet protocol datacast (IPDC) (ETSI, 2006a) is a framework for the distribution of digital data services to wireless, mobile handsets via a broadcast infrastructure based on digital video broadcasting for handhelds (DVB-H). The technical extensions of DVB-H (ETSI, 2004; Faria et al., 2006) to the widely used terrestrial DVB transmission system called DVB-T (ETSI, 2009a; Reimers, 2005) allow for specific requirements of rather small mobile devices, like a smaller screen size compared to television sets, or their dependency on accumulators for electricity and therefore their special need for low power consumption. As a broadcasting infrastructure does not automatically offer the possibility of data transmission in the “upstream” path from a mobile device to the sender, in the context of IPDC this sole one-way system is enriched with upstream paths using cellular networks or wireless local area networks (WLAN) – two types of networks which are typically used by mobile devices. Using these upstream paths for interactivity or signalling purposes, handsets are given the possibility to inform the sender about lost data segments allowing it to retransmit portions of data already sent. This chapter will give an introduction into the digital video broadcasting (DVB) technology, especially into the DVB transmission system useful for mobile devices called DVB-H. It will describe the architecture of IP datacast, which is based on DVB-H, and the protocols used by this technology. One means of using IPDC is the broadcasting of binary data to a number of receivers using IP multicast protocols. In order to achieve correct reception of all of the sent binary data, a signalling mechanism has been specified, which enables receivers to inform the sender about lost or irreparable data fragments. This so-called file repair mechanism allows the setting of different parameters which affect the amount of data which can be corrected at the receiver and therefore have impact on the retransmission costs. The main topics of the research presented in this chapter are studies on the parameterization of the IPDC file repair mechanism and the effects of retransmissions on financial repair costs. In order to accomplish these studies, a simulation model has been designed and implemented which simulates an IP datacast network including a state-of-the-art error model for the wireless transmission channel and a versatilely parametrizable implementation of the file repair mechanism.

We present the parallel particle filtering (PPF) software library, which enables hybrid shared-memory/distributedmemory parallelization of particle filtering (PF) algorithms combining the Message Passing Interface (MPI) with multithreading for multi-level parallelism. The library is implemented in Java and relies on OpenMPI's Java bindings for inter-process communication. It includes dynamic load balancing, multi-thread balancing, and several algorithmic improvements for PF, such as input-space domain decomposition. The PPF library hides the difficulties of efficient parallel programming of PF algorithms and provides application developers with a tool for parallel implementation of PF methods. We demonstrate the capabilities of the PPF library using two distributed PF algorithms in two scenarios with different numbers of particles. The PPF library runs a 38 million particle problem, corresponding to more than 1.86 TB of particle data, on 192 cores with 67% parallel efficiency.

This paper is motivated by work on a real-time Android navigation system using accelerometer, gyroscope, and camera data. We observed that our attempts at combining data from these different sensors yielded time series with jitter, i.e., inaccurate timestamps, that negatively affected program correctness and reliability. Thus, the aim of this paper is to help Android developers anticipate, assess, and understand problems with data quality in sensor-dependent applications. Our analysis focuses on jitter in timestamps obtained from Android's SensorManager and CameraManager services. We demonstrate the presence and severity of jitter obtained through these services and explain the characteristics of Android that originate jitter, resulting in a taxonomy of the various sources of jitter. Finally, we discuss the results of a quantitative analysis of jitter performed on three Android mobile devices (HTC One M8, Google Nexus 5, and Asus Zenfone 2) and provide recommendations on how Android developers can design sensing applications that will behave correctly across the majority of mobile devices.

Purpose – The purpose of this paper is to propose an approach that helps in acquisition of live data as well as data stored in the internal/external memory of android mobile device considering that the data on the device are not much altered during the extraction process. Also, the emphasis is laid on testing the validity of existing forensic tools against the data obtained manually and by using this approach. Smartphones have spurred the mobile computing technology, and Android is widely used as an Operating System in these devices. These days, users store most of their personal information like emails, images, contacts etc., on Phones/Tablets as their data would be readily accessible and thus convenient for them. Design/methodology/approach – Android Operating System is built on the Linux Kernel and scripts to extract data from Android Mobile Device with the use of Android Debugging Bridge have been written. The approach is more focused on the logical acquisition of data from devices rather than acquisition using physical methods. Findings – Live data of the Facebook application running on the device can be extracted. Also, the password of the LuksManager application (used to create an encrypted volume on the device), which is stored in the internal memory, is also extracted and identified. Research limitations/implications – The study has been conducted in an academic environment, thereby limiting external validity. Another limitation is the limited edition of some of the software forensics tools that are used. The full access to these software tools are restricted by Law enforcement and Investigation policies. The research provides a different approach which could aid in criminal investigation activities on mobile devices. Practical implications – The devices which have the latest versions of Android not only store messages and mails, but a lot of information about GPS, as well as information about popular applications like Facebook, WhatsApp, etc. This could practically help a lot in criminal investigation. Originality/value – This study is important because very few works have been done on recent versions (Jellybean and Kitkat) of Android. The proposed approach could extract large amounts of information as compared to earlier approaches with the newer versions of Android having larger memory and new features.

This article presents Andromaly--a framework for detecting malware on Android mobile devices. The proposed framework realizes a Host-based Malware Detection System that continuously monitors various features and events obtained from the mobile device and then applies Machine Learning anomaly detectors to classify the collected data as normal (benign) or abnormal (malicious). Since no malicious applications are yet available for Android, we developed four malicious applications, and evaluated Andromaly's ability to detect new malware based on samples of known malware. We evaluated several combinations of anomaly detection algorithms, feature selection method and the number of top features in order to find the combination that yields the best performance in detecting new malware on Android. Empirical results suggest that the proposed framework is effective in detecting malware on mobile devices in general and on Android in particular.

The work reported here focuses on creating, composing, and recognizing multisensor gestures and their use in terms of remotely controlling another device. The approach allows to compose gestures synthetically, i.e., by recording complex gestures directly as they are performed, and analytically, i.e., by selecting and composing existing gestures to define more complex ones. Analytical gesture composition doesn't have to be sequential: the gestures can be superposed. We explored this in a setting of using a mobile device to control a remote device. In our case, this was a smartphone controlling a computer, which we assume to be most probable setting in practice. For this, a client-server system was developed consisting of the application running on the Android mobile device using the accelerometer, proximity sensor, touchscreen, and ambient light sensor to capture gestures. The server-side application that runs on a computer was developed to store and recognize gestures. The preliminary evaluation results are promising exhibiting the 85% rate in the ability of users to replicate gestures not created by them. According to informal observations, the user seem to be quite comfortable working with their own multisensor gestures.

Deploying mobile devices to frontline troops presents many potential benefits, e.g. situational awareness, enhanced communication capabilities, etc. However, security remains an impediment to realizing such capability. In this research, we develop and evaluate an approach to securing the non-volatile storage of a collection of mobile devices. Our technique relies on well-established cryptographic primitives, combining them in a unique way to meet military mission specific security and resiliency requirements. Specifically, we create MDFS, a distributed mobile file system using erasure coding, Shamir's threshold secret sharing, and the symmetric AES block cipher. The resulting system provides two important properties: (1) data at rest is protected even after total compromise of up to k devices, and (2) data is replicated within an infrastructureless ad hoc network and, as such, resilient to device outages. We implement MDFS on Android mobile devices and achieve ≃10Mbps throughput in real-world performance experiments, suggesting that MDFS is suitable for a variety of practical workloads.

The space around the body provides a large interaction volume that can allow for big interactions on small mobile devices. However, interaction techniques making use of this opportunity are underexplored, primarily focusing on distributing information in the space around the body. We demonstrate three types of around-body interaction including canvas, modal and context-aware interactions in six demonstration applications. We also present a sensing solution using standard smartphone hardware: a phone's front camera, accelerometer and inertia measurement units. Our solution allows a person to interact with a mobile device by holding and positioning it between a normal field of view and its vicinity around the body. By leveraging a user's proprioceptive sense, around-body Interaction opens a new input channel that enhances conventional interaction on a mobile device without requiring additional hardware.

Using soft keyboard is the main text entry method on mobile devices. However, texting on small mobile devices, such as smartphones remains challenging and inconvenient. It’s because of the small size of keys, the difference between view angle and touch point, the tilt of the device etc. When entering text, the human eye conducted visual search for the target key, then casted their attention on the symbol on the button. In this research we would like to understand how the touch points would be affected by the symbol positions on the button. A click-based App has been designed to observe, record, and analyze the data and behaviors of the subjects when they clicked on the keypad buttons. In the experiments the subjects have been asked to perform a series of clicking on the target points under five conditions that the symbols appeared at the center of the buttons, the upper right portions, the lower right portions, the upper left portions and the lower left portions. According to analyses we fund that under the circumstances of different symbol positions, the subjects tended to click on the positions where the symbols appeared instead of the centers of the buttons.