Ching-Ling Huang

Adaptive intervehicle communication control for cooperative safety systems

Vehicular ad hoc networks play a critical role in enabling important active safety applications such as cooperative collision warning. These active safety applications rely on continuous broadcast of self-information by all vehicles, which allows each vehicle to track all its neighboring cars in real time. The most pressing challenge in such safety-driven communication is to maintain acceptable tracking accuracy while avoiding congestion in the shared channel. In this article we propose a transmission control protocol that adapts communication rate and power based on the dynamics of a vehicular network and safety-driven tracking process. The proposed solution uses a closed-loop control concept and accounts for wireless channel unreliability. Simulation results confirm that if packet generation rate and associated transmission power for safety messages are adjusted in an on-demand and adaptive fashion, robust tracking is possible under various traffic conditions.

Analysis of Information Dissemination in Vehicular Ad-Hoc Networks With Application to Cooperative Vehicle Safety Systems

Cooperative vehicle safety systems (CVSSs) rely on vehicular ad-hoc networks (VANETs) for the delivery of critical vehicle tracking information. The wireless channel in such systems is shared by vehicles within the transmission range of each other. Due to the near-linear spatial distribution of vehicles in a highway scenario, the vehicular broadcast network is heavily affected by the hidden node interference phenomenon, which considerably limits its capacity. The performance of vehicle tracking application that is the basis for CVSS is therefore significantly affected by the performance of the underlying network. The two main parameters that affect the network condition and performance are the range and rate (frequency) of transmission of safety and tracking messages. In this paper, we analyze the effect of different choices of rate and range and present models that quantify network performance in terms of its ability to disseminate tracking information. Following a thorough analysis of the hidden node affected VANET, we show that channel occupancy or busy ratio can be used as a feedback measure that quantifies the success of information dissemination and, consequently, the CVSS, under different network conditions. These findings are used to design feedback control schemes for transmission range adaptation, which are robust to variations of road and network traffic.

Design of cooperative vehicle safety systems based on tight coupling of communication, computing and physical vehicle dynamics

One of the main characteristics of a Cyber Physical System (CPS) is the tight coupling of the computing and communications aspects of the system with its physical dynamics. In this paper, we examine this characteristic for a cooperative vehicle safety (CVS) system, and identify how the design and operation of such CPSs should consider this tight coupling. In CVS systems, vehicles broadcast their physical state information over a shared wireless network to allow their neighbors to track them and predict possible collisions. The physical dynamics of vehicle movement and the required accuracy from tracking process dictate certain load on the network. The network performance is directly affected by the amount of offered load, and in turn directly affects the tracking process and its required load. The tight mutual dependence of physical dynamics of vehicle (physical component), estimation/tracking process and communication process (cyber components) require a new look at how such systems are designed and operated. We consider these factors and propose methods to simplify the design procedure for such tightly coupled systems. The method includes modeling the subcomponent of the CPS and devising interaction and control algorithms to operate them. The proposed methods are compared with methods based on separate design of components that deal with physical and cyber aspects. Through simulation experiments we show significant gains in performance when CPS design considerations are respected.

Intervehicle Transmission Rate Control for Cooperative Active Safety System

We propose an intervehicle communication framework for the cooperative active safety system (CASS) whose operation is based on the dissemination of each vehicles state information through a wireless network. Such a CASS requires each subject vehicle to be aware of its surroundings, particularly of the motion and position of other vehicles in its proximity. In this paper, we assume that all vehicles are equipped with onboard communication devices. In such situations, the wireless channel is simultaneously shared by a large number of vehicles, and one of the most difficult challenges in designing CASS is to maintain real-time tracking accuracy of neighboring vehicles while avoiding network congestion and failure. To address this issue, we analyze the problem that multiple scalar linear time-invariant dynamical systems track each other over a multiaccess channel, and then, we propose a rate adaptation algorithm to distributively control the self-information broadcast behavior of each vehicle. The proposed algorithm uses a closed-loop control concept and accounts for the lossy channel. Simulation results show that, if the message generation rate is dynamically adjusted in an on-demand fashion, more accurate and robust tracking performance can be achieved under various traffic conditions.

Congestion Control Based on Channel Occupancy in Vehicular Broadcast Networks

Cooperative vehicle safety (CVS) systems rely on vehicular ad-hoc networks operating in broadcast mode to deliver vehicle tracking and safety information to neighboring cars. This information is used to enable collision avoidance and warning systems. One of the main challenges of the eventual large scale deployment of such systems is network congestion, which could critically degrade the quality of a CVS system. In this paper, we present a method for congestion monitoring and control based on limited feedback from the network. We study the relationship between channel occupancy, as a readily available feedback measure, controllable network parameters, and network performance. We describe a performance measure relevant to CVS systems and present a congestion control method, based on channel occupancy measurements, that robustly maintains the system performance near optimal operation points. We examine the convergence properties of the congestion control algorithm and provide guidelines for the design of such systems based on network and traffic density conditions. Through simulation experiments we show significant gains in performance when closed loop congestion control methods are applied.

Information Dissemination Control for Cooperative Active Safety Applications in Vehicular Ad-Hoc Networks

Vehicular Ad-Hoc networks (VANETs) play a critical role in enabling highway active safety applications such as collision warning and vehicle tracking. The most pressing challenge in enabling such applications is to maximize the amount of disseminated vehicle state information while avoiding network congestion. In this paper, we explore the structure of VANET tracking problem and propose an adaptive rate control algorithm based on network condition and tracking error. Proposed algorithm uses a closed-loop control concept and accounts for the lossy channel. This algorithm is shown to achieve better tracking performance than existing solutions. We first analyze the algorithm behavior in small scale Matlab simulations with bounded dynamical systems and then evaluate its performance using OPNET simulations with realistic vehicle trajectories from a microscopic traffic simulator (SHIFT).

Implementation and evaluation of scalable vehicle-to-vehicle safety communication control

Vehicle-to-vehicle (V2V) communications play a critical role in enabling numerous important cooperative safety applications. V2V safety communications rely on broadcast of self-state information (e.g., position, speed, and heading) by each vehicle, which allows a vehicle to track its neighboring vehicles in real time. One of the most pressing challenges in this research is to maintain acceptable tracking accuracy of neighboring vehicles while avoiding congestion in the shared communication channel. In this article we describe the evaluation of a transmission control protocol that adapts the message rate and transmission power for V2V safety communications. This protocol has been implemented on V2V test vehicles with wireless radios and integrated with existing active safety applications. The testing and evaluation results show that proposed communication design works well in practice, its performance matches the observations from previous simulations and shows great promise for a large-scale deployment of V2V cooperative safety systems.

Robustness Evaluation of Decentralized Self-Information Dissemination Control Algorithms for VANET Tracking Applications

The success of VANET safety mechanisms, e.g. pre-crash warning and collaborative collision avoidance, relies on good proximity-awareness of each subject vehicle. Prior analysis (17), (18) shows that a self-information dissemination control on each car, which adapts communication rate based on estimation error and channel congestion, can achieve higher tracking accuracy than others in a multi-access channel. In this paper, we extensively evaluate the robustness of three communication policies under different highway traffics. The worst case scenario, in terms of tracking accuracy, is identified to be a bidirectional highway with one direction congested and another direction being free flowing. Our proposed algorithms are shown to be robust and achieve much better tracking performance than 100-ms beaconing suggested by (24). I. INTRODUCTION

Field Measurements of Vehicle to Roadside Communication Performance

Field measurements were made of vehicle to roadside communication performance at two roadside dedicated short range communication sites in San Francisco. Portable equipment allowing the recording of data gathered from multiple radios on a single vehicle used the wireless access in a vehicular environment radio modules produced by Denso according to the specifications of the crash avoidance metrics programs vehicle safety communications consortium. Software was developed at the application layer to generate artificial loading in order to emulate with one or a few vehicles some of the characteristics of systems with many more vehicles. Experiments were carried out to emulate transaction processing on the service channel and investigate performance in an urban setting.

Analysis of aggregated power level and rate-power control designs for status update messages in VANETs

In vehicular ad-hoc networks, status update messages are used to disseminate vehicle state information so that each vehicle can track movements of its neighbors. In this paper, we focus on the aggregated power level from this kind of active safety messages and its impact on DSRC channel quality. With macroscopic model and LWR (Lighthill-Whitham-Richards) PDE, we show how the probability distribution of aggregated power level propagates along time and space. Based on our analysis, several rate-power control ideas for status update messages are discussed. In addition, microscopic traffic/network simulation results show that our proposed rate-power control algorithm can reduce channel interference and thus enhance tracking accuracy.