Searcharxiv⌕ Search

arXiv subjects

Mashrur Chowdhury

Publications and source records attributed to Mashrur Chowdhury.

83 records · Page 5Linked to original sources
arXiv2019

Real-time Pedestrian Detection Approach with an Efficient Data Communication Bandwidth Strategy

Vehicle-to-Pedestrian (V2P) communication can significantly improve pedestrian safety at a signalized intersection. It is unlikely that pedestrians will carry a low latency communication enabled device and activate a pedestrian safety application in their hand-held device all the time. Because of this limitation, multiple traffic cameras at the signalized intersection can be used to accurately detect and locate pedestrians using deep learning and broadcast safety alerts related to pedestrians to warn connected and automated vehicles around a signalized intersection. However, unavailability of high-performance computing infrastructure at the roadside and limited network bandwidth between traffic cameras and the computing infrastructure limits the ability of real-time data streaming and processing for pedestrian detection. In this paper, we develop an edge computing based real-time pedestrian detection strategy combining pedestrian detection algorithm using deep learning and an efficient data communication approach to reduce bandwidth requirements while maintaining a high object detection accuracy. We utilize a lossy compression technique on traffic camera data to determine the tradeoff between the reduction of the communication bandwidth requirements and a defined object detection accuracy. The performance of the pedestrian-detection strategy is measured in terms of pedestrian classification accuracy with varying peak signal-to-noise ratios. The analyses reveal that we detect pedestrians by maintaining a defined detection accuracy with a peak signal-to-noise ratio (PSNR) 43 dB while reducing the communication bandwidth from 9.82 Mbits/sec to 0.31 Mbits/sec, a 31x reduction.

cs.CV↗
arXiv2018

Connected Vehicle Application Development Platform (CVDeP) for Edge-centric Cyber-Physical Systems

Connected vehicle (CV) application developers need a development platform to build, test and debug CV applications, such as safety, mobility, and environmental applications, in an edge-centric Cyber-Physical Systems. Our study objective is to develop and evaluate a scalable and secure CV application development platform (CVDeP) that enables the CV application developers to build, test and debug CV applications in real-time. CVDeP ensures that the functional requirements of the CV applications meet the latency requirements imposed by corresponding CV applications. We conducted a case study to evaluate the efficacy of CVDeP using two CV applications (one safety and one mobility application) and validated them through a field evaluation at the Clemson University Connected and Autonomous Vehicle Testbed (CU-CAVT). The analysis outcome proves the efficacy of CVDeP, which satisfies the functional requirements (e.g., latency, throughput) of a CV application while maintaining scalability, and security of the platform and applications.

cs.NI↗
arXiv2018

Change Point Models for Real-time V2I Cyber Attack Detection in a Connected Vehicle Environment

Connected vehicle (CV) systems are cognizant of potential cyber attacks because of increasing connectivity between its different components such as vehicles, roadside infrastructure and traffic management centers. However, it is a challenge to detect security threats in real-time and develop appropriate/effective countermeasures for a CV system because of the dynamic behavior of such attacks, high computational power requirement and a historical data requirement for training detection models. To address these challenges, statistical models, especially change point models, have potentials for real-time anomaly detections. Thus, the objective of this study is to investigate the efficacy of two change point models, Expectation Maximization (EM) and Cumulative Sum (CUSUM), for real-time V2I cyber attack detection in a CV Environment. To prove the efficacy of these models, we evaluated these two models for three different type of cyber attack, denial of service (DOS), impersonation, and false information, using basic safety messages (BSMs) generated from CVs through simulation. Results from numerical analysis revealed that EM and CUSUM could detect these cyber attacks, DOS, impersonation, and false information, with an accuracy of 99\%, 100\%, and 98\%, and 100\%, 100\% and 98\%, respectively.

cs.CR↗
arXiv2018

Development and Evaluation of Recurrent Neural Network based Models for Hourly Traffic Volume and AADT Prediction

The prediction of high-resolution hourly traffic volumes of a given roadway is essential for transportation planning. Traditionally, Automatic Traffic Recorders (ATR) are used to collect this hourly volume data. These large datasets are time series data characterized by long-term temporal dependencies and missing values. Regarding the temporal dependencies, all roadways are characterized by seasonal variations that can be weekly, monthly or yearly, depending on the cause of the variation. Regarding the missing data in a time-series sequence, traditional time series forecasting models perform poorly under the influence of seasonal variations. To address this limitation, robust, Recurrent Neural Network (RNN) based, multi-step ahead forecasting models are developed for time-series in this study. The simple RNN, the Gated Recurrent Unit (GRU) and the Long Short-Term Memory (LSTM) units are used to develop the model and evaluate its performance. Two approaches are used to address the missing value issue: masking and imputation, in conjunction with the RNN models. Six different imputation algorithms are then used to identify the best model. The analysis indicates that the LSTM model performs better than simple RNN and GRU models, and imputation performs better than masking to predict future traffic volume. Based on analysis using 92 ATRs, the LSTM-Median model is deemed the best model in all scenarios for hourly traffic volume and AADT prediction, with an average RMSE of 274 and MAPE of 18.91% for hourly traffic volume prediction and average RMSE of 824 and MAPE of 2.10% for AADT prediction.

stat.AP↗
arXiv2018

Towards Secure Infrastructure-based Cooperative Adaptive Cruise Control

Cooperative Adaptive Cruise Control (CACC) is a pivotal vehicular application that would allow transportation field to achieve its goals of increased traffic throughput and roadway capacity. This application is of paramount interest to the vehicular technology community with a large body of literature dedicated to research within different aspects of CACC, including but not limited to security with CACC. Of all available literature, the overwhelming focus in on CACC utilizing vehicle-to-vehicle (V2V) communication. In this work, we assert that a qualitative increase in vehicle-to-infrastructure (V2I) and infrastructure-to-vehicle (I2V) involvement has the potential to add greater value to CACC. In this study, we developed a strategy for detection of a denial-of-service (DoS) attack on a CACC platoon where the system edge in the vehicular network plays a central role in attack detection. The proposed security strategy is substantiated with a simulation-based evaluation using the ns-3 discrete event network simulator. Empirical evidence obtained through simulation-based results illustrate successful detection of the DoS attack at four different levels of attack severity using this security strategy.

cs.CR↗
arXiv2018

Feasibility of 5G mm-wave communication for connected autonomous vehicles

The internet-of-things (IoT) environment holds different intelligent components networked together and will enable seamless data communication between the connected components. Connected autonomous vehicles or CAVs are major components of the IoT, and the smooth, reliable, and safe operation of CAVs demands a reliable wireless communication system, which can ensure high connectivity, high throughput and low communication latency. The 5G millimeter-wave or mm-wave communication network offers such benefits, which can be the enabler of CAVs, especially for dense congested areas. In this research, we have evaluated the 5G mm-wave and Dedicated Short Range Communication (DSRC) for different CAV applications in a Network Simulator-3 (ns-3). For the CAV applications, we have evaluated the end-to-end latency, packet loss, and data rate (for both the CAV receiver and transmitter) of the 5G mm-wave. We have found that 5G mm-wave can support CAV safety applications by ensuring lower latency compared to the required minimum latency of 200 milliseconds for the forward collision warning application. For mobility applications, we have found that 5G mm-wave can support multiple CAVs with a high data receive rate, which is enough for real-time high definition video streaming for in-vehicle infotainment, with mean packet delay of 13 milliseconds. The findings from this study establish that 5G mm-wave can be the enabler of future CAVs in congested areas. Using the evaluation framework developed in this paper, public agencies can evaluate 5G mm-wave to support CAVs in congested areas, such as heavy pedestrian areas like downtown, commercial zones, under their jurisdiction.

cs.NI↗
arXiv2017

Lessons Learned from the Real-world Deployment of a Connected Vehicle Testbed

The connected vehicle (CV) system promises unprecedented safety, mobility, environmental, economic and social benefits, which can be unlocked using the enormous amount of data shared between vehicles and infrastructure (e.g., traffic signals, centers). Real world CV deployments including pilot deployments help solve technical issues and observe potential benefits, both of which support the broader adoption of the CV system. This study focused on the Clemson University Connected Vehicle Testbed (CUCVT) with the goal of sharing the lessons learned from the CUCVT deployment. The motivation of this study was to enhance early CV deployments with the objective of depicting the lessons learned from the CUCVT testbed, which includes unique features to support multiple CV applications running simultaneously. The lessons learned in the CUCVT testbed are described at three different levels: i) the development of system architecture and prototyping in a controlled environment, ii) the deployment of the CUCVT testbed, and iii) the validation of the CV application experiments in the CUCVT. Our field experiments with a CV application validated the functionalities needed for running multiple diverse CV applications simultaneously under heterogeneous wireless networking, and realtime and non real time data analytics requirements. The unique deployment experiences, related to heterogeneous wireless networks, real time data aggregation, a distribution using broker system and data archiving with big data management tools, gained from the CUCVT testbed, could be used to advance CV research and guide public and private agencies for the deployment of CVs in the real world.

cs.CY↗
arXiv2017

Development of Statewide AADT Estimation Model from Short-Term Counts: A Comparative Study for South Carolina

Annual Average Daily Traffic (AADT) is an important parameter used in traffic engineering analysis. Departments of Transportation (DOTs) continually collect traffic count using both permanent count stations (i.e., Automatic Traffic Recorders or ATRs) and temporary short-term count stations. In South Carolina, 87% of the ATRs are located on interstates and arterial highways. For most secondary highways (i.e., collectors and local roads), AADT is estimated based on short-term counts. This paper develops AADT estimation models for different roadway functional classes with two machine learning techniques: Artificial Neural Network (ANN) and Support Vector Regression (SVR). The models aim to predict AADT from short-term counts. The results are first compared against each other to identify the best model. Then, the results of the best model are compared against a regression method and factor-based method. The comparison reveals the superiority of SVR for AADT estimation for different roadway functional classes over all other methods. Among all developed models for different functional roadway classes, the SVR-based model shows a minimum root mean square error (RMSE) of 0.22 and a mean absolute percentage error (MAPE) of 11.3% for the interstate/expressway functional class. This model also shows a higher R-squared value compared to the traditional factor-based model and regression model. SVR models are validated for each roadway functional class using the 2016 ATR data and selected short-term count data collected by the South Carolina Department of Transportation (SCDOT). The validation results show that the SVR-based AADT estimation models can be used by the SCDOT as a reliable option to predict AADT from the short-term counts.

cs.LG↗
arXiv2017

Cybersecurity Attacks in Vehicle-to-Infrastructure (V2I) Applications and their Prevention

A connected vehicle (CV) environment is composed of a diverse data collection, data communication and dissemination, and computing infrastructure systems that are vulnerable to the same cyberattacks as all traditional computing environments. Cyberattacks can jeopardize the expected safety, mobility, energy, and environmental benefits from connected vehicle applications. As cyberattacks can lead to severe traffic incidents, it has become one of the primary concerns in connected vehicle applications. In this paper, we investigate the impact of cyberattacks on the vehicle-to-infrastructure (V2I) network from a V2I application point of view. Then, we develop a novel V2I cybersecurity architecture, named CVGuard, which can detect and prevent cyberattacks on the V2I environment. In designing CVGuard, key challenges, such as scalability, resiliency and future usability were considered. A case study using a distributed denial of service (DDoS) on a V2I application, i.e., the Stop Sign Gap Assist (SSGA) application, shows that CVGuard was effective in mitigating the adverse effects created by a DDoS attack. In our case study, because of the DDoS attack, conflicts between the minor and major road vehicles occurred in an unsignalized intersection, which could have caused potential crashes. A reduction of conflicts between vehicles occurred because CVGuard was in operation. The reduction of conflicts was compared based on the number of conflicts before and after the implementation and operation of the CVGuard security platform. Analysis revealed that the strategies adopted by the CVGuard were successful in reducing the inter-vehicle conflicts by 60% where a DDoS attack compromised the SSGA application at an unsignalized intersection.

cs.NI↗
arXiv2017

Adaptive Queue Prediction Algorithm for an Edge Centric Cyber Physical System Platform in a Connected Vehicle Environment

In the early days of connected vehicles (CVs), data will be collected only from a limited number of CVs (i.e., low CV penetration rate) and not from other vehicles (i.e., non-connected vehicles). Moreover, the data loss rate in the wireless CV environment contributes to the unavailability of data from the limited number of CVs. Thus, it is very challenging to predict traffic behavior, which changes dynamically over time, with the limited CV data. The primary objective of this study was to develop an adaptive queue prediction algorithm to predict real-time queue status in the CV environment in an edge-centric cyber-physical system (CPS), which is a relatively new CPS concept. The adaptive queue prediction algorithm was developed using a machine learning algorithm with a real-time feedback system. The algorithm was evaluated using SUMO (i.e., Simulation of Urban Mobility) and ns3 (Network Simulator 3) simulation platforms to illustrate the efficacy of the algorithm on a roadway network in Clemson, South Carolina, USA. The performance of the adaptive queue prediction application was measured in terms of queue detection accuracy with varying CV penetration levels and data loss rates. The analyses revealed that the adaptive queue prediction algorithm with feedback system outperforms without feedback system algorithm.

cs.CY↗