(Semi-)Automatically Parsing Private Protocols for In-Vehicle ECU Communications

In-vehicle electronic control unit (ECU) communications generally count on private protocols (defined by the manufacturers) under controller area network (CAN) specifications. Parsing the private protocols for a particular vehicle model would be of great significance in testing the vehicle’s resistance to various attacks, as well as in designing efficient intrusion detection and prevention systems (IDPS) for the vehicle. This paper proposes a suite of methods for parsing ECU private protocols on in-vehicle CAN network. These methods include an algorithm for parsing discrete variables (encoded in a discrete manner, e.g., gear state), an algorithm for parsing continuous variables (encoded in a continuous manner, e.g., vehicle speed), and a parsing method based on upper-layer protocols (e.g., OBD and UDS). Extensive verifications have been performed on five different brands of automobiles (including an electric vehicle) to demonstrate the universality and the correctness of these parsing algorithms. Some parsing tips and experiences are also presented. Our continuous-variables parsing algorithm could run in a semi-automatic manner and the parsing algorithm from upper-layer protocols could execute in a completely automatic manner. One might view the results obtained by our parsing algorithms as an important indicator of penetration testing on in-vehicle CAN network.

A derivative-based parser generator for visibly Pushdown grammars

In this paper, we present a derivative-based, functional recognizer and parser generator for visibly pushdown grammars. The generated parser accepts ambiguous grammars and produces a parse forest containing all valid parse trees for an input string in linear time. Each parse tree in the forest can then be extracted also in linear time. Besides the parser generator, to allow more flexible forms of the visibly pushdown grammars, we also present a translator that converts a tagged CFG to a visibly pushdown grammar in a sound way, and the parse trees of the tagged CFG are further produced by running the semantic actions embedded in the parse trees of the translated visibly pushdown grammar. The performance of the parser is compared with a popular parsing tool ANTLR and other popular hand-crafted parsers. The correctness of the core parsing algorithm is formally verified in the proof assistant Coq.

An Intelligent Code Search Approach Using Hybrid Encoders

The intelligent code search with natural language queries has become an important researching area in software engineering. In this paper, we propose a novel deep learning framework At-CodeSM for source code search. The powerful code encoder in At-CodeSM, which is implemented with an abstract syntax tree parsing algorithm (Tree-LSTM) and token-level encoders, maintains both the lexical and structural features of source code in the process of code vectorizing. Both the representative and discriminative models are implemented with deep neural networks. Our experiments on the CodeSearchNet dataset show that At-CodeSM yields better performance in the task of intelligent code searching than previous approaches.

Video content analysis based on statistical modeling

This thesis is aimed at finding solutions and statistical modeling techniques to analyze the video content in a way such that intelligent and efficient interaction with video is possible. In our work, we investigate several fundamental tasks for content analysis of video. Specifically, we propose an outline video parsing algorithm using basic statistical measures and an off-line solution using Independent Component Analysis (ICA). A spatiotemporal video similarity model based on dynamic programming is developed. For video object segmentation and tracking, we develop a new method based on probabilistic fuzzy c-means and Gibbs random fields. Theoretically, we develop a generic framework for sequential data analysis. The new framework integrates both Hidden Markov Model and ICA mixture model. The re-estimation formulas for model parameter learning are also derived. As a case study, the new model is applied to golf video for semantic event detection and recognition.

Video content analysis based on statistical modeling

This thesis is aimed at finding solutions and statistical modeling techniques to analyze the video content in a way such that intelligent and efficient interaction with video is possible. In our work, we investigate several fundamental tasks for content analysis of video. Specifically, we propose an outline video parsing algorithm using basic statistical measures and an off-line solution using Independent Component Analysis (ICA). A spatiotemporal video similarity model based on dynamic programming is developed. For video object segmentation and tracking, we develop a new method based on probabilistic fuzzy c-means and Gibbs random fields. Theoretically, we develop a generic framework for sequential data analysis. The new framework integrates both Hidden Markov Model and ICA mixture model. The re-estimation formulas for model parameter learning are also derived. As a case study, the new model is applied to golf video for semantic event detection and recognition.

A note on the incorrect operation of the Covington parsing algorithm

Рассмотрен алгоритм Ковингтона для синтаксического анализа проективных зависимостей LSUP. Показано, что алгоритм анализирует не все проективные структуры (в частности, которые указаны в самой статье), поэтому использование данного алгоритма приводит к некорректным результатам. The Covington algorithm LSUP for the syntactic analysis of projective dependencies is considered. It is shown that the algorithm does not analyze all the projective structures (in particular, those specified in the article itself), so the use of this algorithm leads to incorrect results.

Parsing Character Arrays by Recursive Scoping and Structuring Using the Example of VRML Data

Parsing character arrays by recursive scoping and structuring using the example of VRML data. The article presents a new method for structuring, segmentation and algorithmic design of the parser of character arrays using the example of VRML data. The key feature of the method is the ability to form a hierarchically complex object by means of recursive data structuring, which makes it possible to cover in aggregate the entire contents of the object, including its arbitrary nesting of child objects. This leads to high controllability of the development of the parsing algorithm, allowing you to focus each time on a specific piece of data, while not losing sight of the entire aggregate connectivity of information. The results obtained can easily be used in plans for creating convenient data storage structures related to information security, solving the problem of containing the amount of data in files, managing big data in heterogeneous systems, and processing hierarchical data in the Internet of Things. Keywords: data analysis method, parsing, structuring, se

Syntactic Analysis of Graphical Information Using Recursive Scoping

The report presents a new method for structuring, segmentation, and algorithmic design of the character array parser using VRML data as an example. The key feature of the method is the possibility of forming a hierarchically complex object by means of recursive data structuring, which allows you to cover the entire contents of the object, including its arbitrary nesting of child objects. This leads to a highly manageable development of the parsing algorithm, allowing you to focus each time on a specific piece of data, while not losing sight of the entire aggregate coherence of the information. The results obtained can easily be used in plans for creating convenient data storage structures related to information security, solving the problem of containing the amount of data in files, data management problems in heterogeneous systems, and hierarchical data solutions in the Internet of Things.