Binary Files Research Articles

Hydra image processor: 5-D GPU image analysis library with MATLAB and python wrappers.

Light microscopes can now capture data in five dimensions at very high frame rates producing terabytes of data per experiment. Five-dimensional data has three spatial dimensions (x, y, z), multiple channels (λ) and time (t). Current tools are prohibitively time consuming and do not efficiently utilize available hardware. The hydra image processor (HIP) is a new library providing hardware-accelerated image processing accessible from interpreted languages including MATLAB and Python. HIP automatically distributes data/computation across system and video RAM allowing hardware-accelerated processing of arbitrarily large images. HIP also partitions compute tasks optimally across multiple GPUs. HIP includes a new kernel renormalization reducing boundary effects associated with widely used padding approaches. HIP is free and open source software released under the BSD 3-Clause License. Source code and compiled binary files will be maintained on http://www.hydraimageprocessor.com. A comprehensive description of all MATLAB and Python interfaces and user documents are provided. HIP includes GPU-accelerated support for most common image processing operations in 2-D and 3-D and is easily extensible. HIP uses the NVIDIA CUDA interface to access the GPU. CUDA is well supported on Windows and Linux with macOS support in the future.

Mimicking Anti-Viruses with Machine Learning and Entropy Profiles.

The quality of anti-virus software relies on simple patterns extracted from binary files. Although these patterns have proven to work on detecting the specifics of software, they are extremely sensitive to concealment strategies, such as polymorphism or metamorphism. These limitations also make anti-virus software predictable, creating a security breach. Any black hat with enough information about the anti-virus behaviour can make its own copy of the software, without any access to the original implementation or database. In this work, we show how this is indeed possible by combining entropy patterns with classification algorithms. Our results, applied to 57 different anti-virus engines, show that we can mimic their behaviour with an accuracy close to 98% in the best case and 75% in the worst, applied on Windows’ disk resident malware.

LDAF: Low-Bandwidth Distributed Applications Framework in a Use Case of Blockchain-Enabled IoT Devices.

In this paper, we present Low-Bandwidth Distributed Applications Framework (LDAF)—an application-aware gateway for communication-constrained Internet of things (IoT) devices. A modular approach facilitates connecting to existing cloud backend servers and managing message formats and APIs’ native application logic to meet the communication constraints of resource-limited end devices. We investigated options for positioning the LDAF server in fog computing architectures. We demonstrated the approach in three use cases: (i) a simple domain name system (DNS) query from the device to a DNS server, (ii) a complex interaction of a blockchain—based IoT device with a blockchain network, and (iii) difference based patching of binary (system) files at the IoT end devices. In a blockchain smart meter use case we effectively enabled decentralized applications (DApp) for devices that without our solution could not participate in a blockchain network. Employing the more efficient binary content encoding, we reduced the periodic traffic from 16 kB/s to ~1.1 kB/s, i.e., 7% of the initial traffic. With additional optimization of the application protocol in the gateway and message filtering, the periodic traffic was reduced to ~1% of the initial traffic, without any tradeoffs in the application’s functionality or security. Using a function of binary difference we managed to reduce the size of the communication traffic to the end device, at least when the binary patch was smaller than the patching file.

Design and Implementation of a Method and Framework for Dynamic Loading of the Function Methods in Cyber-physical Systems Nodes

In order to satisfy the need of the function methods of slaver nodes for dynamic load after the nodes in the Cyber-physical Systems (CPS) network been dynamically reconstructed, and provide a unified specification to develop and call the different function method, this paper puts forward a method and framework of dynamic loading of the function methods belong to the CPS nodes. In this method and framework, provide uniform interface specifications, compile the code of the function methods into binary executable files and store it in the storage space of the slave node. During initialize the node, read the binary data to master node, and then load the function methods by parsing it directly and execute by name. Establishing an experiment system to verify the method and framework, the function methods could be loaded successful and it could realize accurate control of slave nodes.

RECODE: revision control for digital images

Revision control is a vital component in the collaborative development of artifacts such as software code and multimedia. While revision control has been widely deployed for text files, very few attempts to control the versioning of binary files can be found in the literature. This can be inconvenient for multimedia applications that use a significant amount of binary data, such as images, videos, meshes, and animations. Existing strategies such as storing whole files for individual revisions or simple binary deltas, respectively consume significant storage and complex semantic information. To overcome these limitations, in this paper we present RECODE, a revision control system for digital images. It stores revisions in the form of a DAG (directed acyclic graph) where nodes represent editing operations, and edges describe the spatial and temporal relationships between operations. Being integrated with GitHub, the largest project hosting platform, RECODE also facilitates the artistic creation process of distributed teams with different workflows that include image editing and digital painting. A preliminary user study was performed to assess the perceived usability of the proposed system.

Low-Cost FPGA-Based Electronic Control Unit for Vehicle Control Systems.

The development of new control algorithms in vehicles requires high economic resources, mainly due to the use of generic real-time instrumentation and control systems. In this work, we proposed a low-cost electronic control unit (ECU) that could be used for both development and implementation. The proposed electronic system used a hybrid system on chip (SoC) between a field-programmable gate array (FPGA) and an Advanced RISC (reduced instruction set computer) Machine (ARM) processor that allowed the execution of parallel tasks, fulfilling the real-time requirements that vehicle controls demand. Another feature of the proposed electronic system was the recording of measured data, allowing the performance of the implemented algorithm to be evaluated. All this was achieved by using modular programming that, without the need for a real-time operating system, executed the different tasks to be performed, exploiting the parallelism offered by the FPGA as well as the dual core of the ARM processor. This methodology facilitates the transition between the designing, testing, and implementation stages in the vehicle. In addition, our system is programmed with a single binary file that integrates the code of all processors as well as the hardware description of the FPGA, which speeds up the updating process. In order to validate and demonstrate the performance of the proposed electronic system as a tool for the development and implementation of control algorithms in vehicles, a series of tests was carried out on a test bench. Different traction control system (TCS) algorithms were implemented and the results were compared.

Using Arm’s scalable vector extension on stencil codes

Data-level parallelism is frequently ignored or underutilized. Achieved through vector/SIMD capabilities, it can provide substantial performance improvements on top of widely used techniques such as thread-level parallelism. However, manual vectorization is a tedious and costly process that needs to be repeated for each specific instruction set or register size. In addition, automatic compiler vectorization is susceptible to code complexity, and usually limited due to data and control dependencies. To address some of these issues, Arm recently released a new vector ISA, the scalable vector extension (SVE), which is vector-length agnostic (VLA). VLA enables the generation of binary files that run regardless of the physical vector register length. In this paper, we leverage the main characteristics of SVE to implement and optimize stencil computations, ubiquitous in scientific computing. We show that SVE enables easy deployment of textbook optimizations like loop unrolling, loop fusion, load trading or data reuse. Our detailed simulations using vector lengths ranging from 128 to 2048 bits show that these optimizations can lead to performance improvements over straightforward vectorized code of up to 1.57$$\times$$. In addition, we show that certain optimizations can hurt performance due to reduced arithmetic intensity and instruction overheads, and provide insight useful for compiler optimizers.

Bidirectional LSTM Malicious webpages detection algorithm based on convolutional neural network and independent recurrent neural network

This paper proposes a bidirectional LSTM algorithm (CBIR) based on convolutional neural network and independent recurrent neural network. The algorithm extracts the “texture fingerprint” feature used to express the similarity of the content of the URL binary file of the malicious webpages, and uses the word vector tool word2vec to train the URL word vector feature and extract the URL static vocabulary feature. The “texture fingerprint” feature, the URL word vector feature and the URL static vocabulary feature are merged, and the malicious webpages is analyzed and detected based on the CBIR algorithm model. Experimental results show that compared with other methods, the proposed CBIR algorithm has improved the accuracy of malicious webpages detection.

Methods for automated formation of a disassembled listing structure

The subject of the research is the method of splitting a disassembled code into logical blocks in automatic mode, searching for software vulnerabilities without using source code (using a binary file or its equivalent, obtained by reverse engineering).The object of the research is the existing code analyzers and features of their functionality.The aim of the study is to consider the possibility of splitting a disassembled code into logical blocks in automatic mode and some of the possible difficulties associated with this.Formulation of the problem. The complexity of analyzing large software products at the level of machine code necessitates the automation of this process. The research methodology is based on a combination of theoretical and empirical approaches using the methods of static and dynamic analysis, comparison, generalization, algorithmization, modeling, synthesis. Key findings. Splitting the code into blocks by sequential in line-by-line analysis of machine code in some cases can lead to misinterpretation. In addition, the analysis of the code according to the conclusions of the functions also does not guarantee the correctness of the determination of the boundaries of the functions. However, in general, the matrix method can be applied to analyze the dependencies of functions on the blocks of code thus selected.The scientific novelty is connected with the determination of promising vectors for the study of software code for vulnerability, the rationale for the approach (building the transition matrix from integer values), which may be the initial stage of preparation for the automated analysis of the disassembled code.

Validation study of voxel phantom monte carlo simulations with EGSnrc C++ class library

The aim of this work is to validate an in-house tool which writes voxel phantoms input files according to EGSnrc C++ class library (egspp) for Monte Carlo simulations. This tool was developed to read a phantom binary file and write a voxel phantom input deck file according to egspp structures. For the validation of the new tool, three voxel phantoms from literature considering different levels of complexity were used. They are the DM_BRA mouse phantom, Golem anthropomorphic phantom, and Case 5 XCAT model phantom from AAPM TG 195 report. For the different cases of study, internal and external photon sources were set and the energy deposition for different source and target tissue/organs were calculated. The results showed good agreement when comparing to dose calculates obtained with other Monte Carlo codes and published in the literature. The new tool was then validated for the egspp Monte Carlo studies with voxel phantoms.

Implementation of interactive three-dimensional visualization of air pollutants using WebGL

The visualization of a three-dimensional (3D) distribution of high-density data for scientific computations with limited computing resources under a network environment is difficult. This paper describes the implementation of an interactive 3D display of air pollutants in a network environment using the open-source WebGL class library, Three. js. A kd-tree is used to find the nearest point data to interpolate the 3D grid data of pollutants because the source data are scattered, which helps locate the data conveniently at the client's end. The binary file format is used to save and transfer data to reduce the number of network transmissions and to improve the speed of the image display. The results show that the visual effects and rendering speed are significantly improved on the web with the use of the proposed technology.

HDF5 and row-wise ntuple in analysis tools in Geant4 10.4

The analysis category was introduced in Geant4 release 9.5 to help users capture statistical data in the form of histograms and ntuples and store these in files in various formats. Up to release 10.3 the following formats had been introduced: csv, AIDA/XML and the binary ROOT file format. We present here the work done to handle, in Geant4 10.4, the binary HDF5 file format, a format/library widely used in other domains of science but quite ignored in HEP for the moment. Work has been done also to support the management of a single file in a multi-thread or MPI parallel environment for the ROOT format; we present the introduction of a row-wise way to manage paginated ntuples in order to restore an “event view” lost by our today column-wise implementation for this format.

BiN: A Two-Level Learning-Based Bug Search for Cross-Architecture Binary

With the popularity of IoT (Internet of Things) devices, the security risks of these devices are increasing. However, due to the multisource heterogeneity of IoT devices, there are significant differences between the vulnerability detection of the Internet of Things and the PC-based vulnerability search method. Therefore, determining how to accurate search for vulnerabilities in large-scale cross-platform binary executable files is an urgent problem to be solved. At present, the solution to this problem mostly calculates code similarities by generating a CFG (control flow graph) from binary code, but due to the choice of architecture, OS (operating system) or compilation options, the same source code will be compiled into different assembly codes. The performance of existing vulnerability search methods for cross-architecture binaries has been challenged. To alleviate the vast differences in the assembly codes caused by different compilation scenarios, this paper proposes a cross-platform large-scale binary vulnerability search method based on two-level feature semantic learning. The contribution is that we have defined a new functional structured signature method to mitigate the massive grammatical and structural differences of binary files caused by different compilation environments. Moreover, we reasonably integrate the hierarchical model of Structure2Vec and GAT (graph attention network) and implement training from the internal control flow characteristics of the function and the call relationship between functions to obtain a more accurate functional semantic expression.

AppAuth: Authorship Attribution for Android App Clones

Android app clone detection has been extensively studied in our community, and a number of effective approaches and frameworks were proposed and released. However, there still remains one open challenge that has not been well addressed in previous work, i.e., the authorship attribution for the detected app clones. Although state-of-the-art approaches could accurately identify repackaged apps in one way or another, no convincing method has been proposed to identify the original app and the authentic author from the repackaged app pairs, which greatly limits the usage scenario of app clone detection techniques. For example, app market maintainers have to manually confirm the identified repackaged app pairs, while in most cases, it is challenging for them to make an accurate decision. In this paper, we propose AppAuth, a novel learning-based approach to predict the authorship of app clones. To be specific, for a given Android app clone pair (or a group of repackaged apps identified), AppAuth could accurately infer the original author of the plagiarized apps. Our approach is motivated by the traditional authorship attribution studies on binary files. AppAuth first extracts a number of coding-style-related features from the executable .apk files, and then relies on machine learning techniques to train a classification model. We have conducted extensive experiments to evaluate the effectiveness of AppAuth. The experiment results suggest that we are able to infer the authorship for Android app clones with high precision. Our work is the first one that tackles the problem systematically and we believe our efforts could positively contribute to the research community and boost the research of app repacking detection and authorship attribution studies.

Run Length Encoding Based Lossless MRI Image Compression Using LZW and Adaptive Variable Length Coding

Medical image compression is considered one of the most important research fields nowadays in biomedical applications. The majority of medical images must be compressed without loss because each pixel information is of great value. With the widespread use of applications concerning medical imaging in the health-care context and the increased significance in telemedicine technologies, it has become crucial to minimize both the storage and bandwidth requirements needed for archiving and transmission of medical imaging data, rather by employing means of lossless image compression algorithms. Furthermore, providing high resolution and image quality preservation of the processed image data has become of great benefit. The proposed system introduces a lossless image compression technique based on Run Length Encoding (RLE) that encodes the original magnetic resonance imaging (MRI) image into actual values and their numbers of occurrence. The actual image data values are separated from their runs and they are stored in a vector array. Lempel–Ziv–Welch (LZW) is used to provide further compression that is applied to values array only. Finally the Variable Length Coding (VLC) will be applied to code the values and runs arrays for the precise amount of bits adaptively into a binary file. These bit streams are reconstructed using inverse LZW of the values array and inverse RLE to reconstruct the input image. The obtained compression gain is enhanced by 25% after applying LZW to the values array.

Towards a serverless CernVM-FS

The CernVM File System (CernVM-FS) provides a scalable and reliable software distribution and—to some extent—a data distribution service. It gives POSIX access to more than a billion binary files of experiment application software stacks and operating system containers to end user devices, grids, clouds, and supercomputers. Increasingly, CernVM-FSalso provides access to certain classes of data, such as detector conditions data, genomics reference sets, or gravitational wave detector experiment data. For most of the high- energy physics experiments, an underlying HTTP content distribution infrastructure is jointly provided by universities and research institutes around the world. In this contribution, we will present recent developments and future plans. For future developments, we put a focus on evolving the content distribution infrastructure and at lowering the barrier for publishing into CernVM-FS. Through so-called serverless computing, we envision cloud hosted CernVM-FS repositories without the need to operate dedicated servers or virtual machines. An S3 compatible service in conjunction with a content delivery network takes on data provisioning, replication, and caching. A chainof time-limited and resource-limited functions (so called “lambda function” or “function-as- a-service”) operate on the repository and stage the updates. As a result, any CernVM-FS client should be able to turn intoawriter, possession of suitable keys provided. For repository owners, we aim at providing cost transparency and seamless scalability from very small to very large CernVM-FS installations.

Simulation and applications of cumulative anisotropic sky radiance patterns

The prediction of the incident solar energy on a surface under a partially obstructed sky is an important task for most solar energy-related architectural and engineering applications. In previous contributions, Ivanova described an approach that solves this problem using cumulative sky radiance patterns. The method considers the numerical integration of fisheye orthographic projections of these patterns on a considered surface, taking into account the shading caused by all existing obstructing objects in the environment. Further developments of the approach are proposed here to improve the method’s accuracy. The resolution system called SOLARES includes two separate units: a pre-processor and a main program. For each new site, the sky radiance patterns have to be computed once with the pre-processor. The latter prepares daily cumulative sky radiance patterns based on the PVGIS TMY solar database. The direct and diffuse components of the sky radiance are modeled separately. The values of diffuse radiance are generated using an anisotropic model for the luminance and radiance sky distributions. The results are saved in binary files for ulterior use in the main program. The fisheye orthographic projections of the cumulative radiance patterns of the sky hemisphere onto a receiving surface are discretized into matrices of pixels and corresponding direct and diffuse radiance values. The shading objects are projected onto that surface, and the results are saved in a separate shading matrix generated for a specific viewpoint or averaged over the whole surface with the help of the finite-element method. The present numerical method can be used to generate accurate long-term time series of irradiance on any tilted exterior or interior surface in complex urban environments with many neighboring shading surfaces of any tilt and orientation. Additionally, the method allows the visualization of the different system components to enhance architectural or solar applications.

ANALYSIS OF ADDITION OF DATA IN INFORMATION PROCESSING AND DECODING GROUND SYSTEM

This article discusses the question of adding data in the information processing and decoding ground system. The scheme of registration of messages by the onboard registration device is resulted. Describes the subsequent recording of registered messages in binary files. For processing the received files the information processing and decoding ground system parses messages by their type. The problem of adding a large amount of data to the database management system «Linter-VS» 7.0. The article identifies and describes various methods of adding data, their advantages and disadvantages. To estimate the execution time of the data loading process was chosen the model of the messages registration system. The table of results of adding data by the various methods is presented and the histogram demonstrating comparison of adding data by different methods are given.

Stress flow analysis of bio-structures using the finite element method and the flow network approach

Bio-structures have been optimized by nature to possess remarkable resiliency and strength yet retain a light-weight composition. The hierarchical geometrical arrangements that are prevalent in bio-structures, coupled with heterogeneous constituents, present uncertainty as to what dictates the structural response. Although computational mechanics experiments give an overall distribution of stresses in the structural system, the underlying details of the stress flow patterns in the structure are hard to identify. This paper discusses the development of software that integrates the results from computational mechanics experiments with advanced mathematical algorithms to capture the stress distribution in a bio-structure subjected to an external load. A fundamental issue addressed in this work is the procedure used to prepare data in appropriate format for the seamless transition from finite element binary database files to an abstract mathematical domain. The robust, platform-independent procedure developed for this transdisciplinary strategy efficiently handles the large datasets produced by the high performance computational mechanics experiments.

A Type Information Reconstruction Scheme Based on Long Short-Term Memory for Weakness Analysis in Binary File

Due to increasing use of third-party libraries because of the increasing complexity of software development, the lack of management of legacy code and the nature of embedded software, the use of third-party libraries which have no source code is increasing. Without the source code, it is difficult to analyze these libraries for vulnerabilities. Therefore, to analyze weaknesses inherent in binary code, various studies have been conducted to perform static analysis using intermediate code. The conversion from binary code to intermediate language differs depending on the execution environment. In this paper, we propose a deep learning-based analysis method to reconstruct missing data types during the compilation process from binary code to intermediate language, and propose a method to generate supervised learning data for deep learning.